Post #593: Expedient masks, part 1, background.

Posted on April 1, 2020



The bottom line of this posting is that, yes, home made masks work to block much of your potential for transmitting disease to others.  Wear a real medical mask if you own one.  Buy (and potentially improve) a non-medical one if you don’t.  Or make one from scratch.  Even paper towels have some efficacy in filtering the air.

This post is really just background.  If you already know that you need to wear a mask, there is no need to read this.  Next post will be about what you can still buy, for now.  Final post will be on making them.


The drumbeat in favor of universal mask use outside the home is getting louder. Some people figured this out sooner than others.  I finally wised up less than a week ago, and did my 3/27/2020 grocery shopping masked.  And every time I enter a public place from now on, I’m wearing a mask.

That said, you can see a summary of the achingly slow progress on this front in this Washington Post article.

So, unless spectacular stupidity triumphs, you will soon have official guidance from the CDC that you ought to wear a mask in public.  I’ve already written down what I would use as the official rules, FWIW (Post #589).  I’ve now posted a comment on the Whole Foods Facebook page, asking them to consider some form of “no mask, no service” policy.

Everybody who is smart, ethical, or both is withholding N95 respirators from the general public.  AKA, the best masks.  Those need to be kept for health care providers.   Here’s Home Depot’s policy.

Executed a "Stop-Sale" on all N95 masks in stores and and redirected all shipments to be donated to hospitals, healthcare providers and first responders around the country

I highlight Home Depot because that’s the first place I detected panic buying/hoarding behavior, back in Post #535. Even if they were too late to stop that, they have had the good sense not to restock.

My point is that mask use is coming.  You can’t buy the ones that protect you with some security (if they are properly fitted).  What can you do?  What are your options for an expedient mask?  What is your best expedient mask choice?  Adapt what you have?  Buy?  Buy and modify?  Make from scratch?

From the standpoint of not infecting others, it’s all good, pretty much.  Any mask beats no mask, hands down.  When you talk, and even to a small extent when you breath, you emit tiny droplets (discussed at length in Post #573, Post #585). Droplet transmission of disease is believed to be the primary way in which the infection is spread.  Any sort of substantial cloth barrier over your face will reduce the spread of those droplets significantly, both by catching some, and by reducing the velocity/range of others, as the mask slows the velocity of your breath.  (See wet finger whistle test in Post #589 to prove this to yourself.)

See this reference for a scholarly look at the ability of home-made masks to stop spread of disease.  But please note that they used the wrong cloth (cotton t-shirts) — nobody recommends that.  They concluded that homemade masks aren’t as good as surgical masks, but they are much better than nothing.  I’ll get into that paper in detail, at the end of this posting.

So the only thing to discuss, really, is how you can get some sort of mask, and then, how well can you expect that mask to protect you.  This assumes you are smart enough to do everything the CDC already says, and in addition, minimize your trips to enclosed public spaces, and the time you spend there, on those trips.

While you can buy some type of (e.g.) cloth face mask now, well, by this time we all know the drill.  The supply chain is set up to deliver product only at the rate at which it is used.  As soon as the CDC says “wear a mask of any sort”, it’s a fair bet that what’s currently for sale will get hoovered up.

So, for most of you who don’t have a mask, I’m afraid this is going to turn into a DIY project.  Or, having your neighbors make one for you.

So I thought that what I might do is offer a kind of mask tutorial.  This is mostly by way of helping me get my thoughts together on whether or not I could build a high-quality expedient mask from available materials.  Anybody can make a mask.  The question is, can I make a good mask, in quantity.

This is part 1:  Background.  Part 2 will be, what can you buy (for now) and modify.  Part 3 will be, what can you make from scratch.

Step1:  What are we talking about?  Definition of N95, PM2.5, MERV, HEPA, and so on.

I’m kind of tired of writing, at this point, so this will largely be done without citation as to sources.  I’m just summarizing things here that I have duly cited in earlier posts, mostly. New information will be cited as needed.

Aerosol versus droplet transmission

Micron (micrometer):  One millionth of a meter.  1000 nanometers.  Droplets from a sneeze range from under about one micron to hundreds of microns.

Nanometer: One billionth of a meter. The diameter of an individual coronavirus particle is around 120 nanometers (Wikipedia).  An obsolete term for it is “millimicron” or “millimicrometer”.

Aerosol versus droplet transmission.  Here’s a key distinction.  Conventionally, anything under 5 microns is an “aerosol”, meaning, it will stay suspended in the air and float around on air currents.  See Post #585.  Particles that size are far too small to be seen. When professionals talk about “airborne transmission”, then mean transmission of the disease in this fashion. If you inhale enough of them, from an infected person, and they land in the right place in your respiratory tract, you get infected.

Studies of influenza showed that aerosol particles were quite potent at spreading lower respiratory tract infections.  It takes less total volume of virus to cause an inflection of you aerosolize it, versus leaving it in larger droplets.  The reason is that if you inhale these tiny droplets, they have the ability to penetrate far into the lungs.  (This same ability is what makes very fine particulates from (say) diesel exhaust particularly harmful air pollution).  And, for SARS (currently SARS-CoV-2 aka COVID-19), that’s where this virus wants to be.  So aerosols put the virus directly into its most favorable habitat within the human body.

Sneezing and coughing produce some aerosol-sized droplets.  But, surprising, just breathing produces a few.  Talking produces them at a rate somewhat less than coughing (but you talk a lot more than you cough, so total production can easily be the same, when integrated over time).  The louder you talk, the more aerosols you produce.  And singing is easily the equivalent of coughing, in terms of the rate of aerosol production.

One fact you need to know is that, for aerosols, some people are “superemitters”.  They produce orders-of-magnitude more aerosol than the average person.  And it’s not really all that rare, but I’m not going to stop and look that up right now.  My recollection is that in any group of 100, you are more than likely to find a few.

And so, likely that event that I wrote about — 45 persons of a 60 person choir got COVID-19, after a single carefully-done practice — it’s a very good guess that this was the result of aerosol transmission of disease by an infected superemitter.

That said, I need to be clear here, because this is now a hot topic:  As far as anyone knows, aerosol transmission of this disease is rare outside the hospital setting.  How do they know?  Well, they think they know that because of what they think they know of the epidemiology of it. If aerosol transmission were common, you probably wouldn’t be able to trace cases back and figure out how they got infected.  Because they think they can do that tracing, and they find that most transmission (that they can trace) appears to occur when people are symptomatic (i.e., coughing), they infer that most transmission is, in fact, droplet transmission, not aerosol.

Let us each now solemnly pray to our respective God or Gods that they are correct, and not just kidding themselves.

The situation in a hospital is different, though.  There, you can get instances where the concentration of aerosol particles is high enough, and exposure times are long enough, that people can get sick literally by simply breathing the same air as the patient.  You’ve got a combination of a very sick patient (so, high virus shedding rate), in a small room (so that person can build up high density of aerosol per cubic foot), and you have health care personnel who spend considerable time in that tainted air.

The non-negligible potential for aerosol transmission in that setting is one of the reasons that health care practitioners need N95 masks.  I’ll get to the definition of N95 below.

By contrast, health professionals use the term “droplets” to mean drops larger than 5 microns.  To a greater or lesser degree, these settle out of the atmosphere fairly quickly.  (Or, at least, that’s the conventional wisdom).  But, if you inhale them, or they land on you in the right place (in your mouth, eyes, nose), or they land on a surface that you touch and then touch your face (mouth, eyes, nose), they can infect you.

You produce droplets by sneezing or coughing, but you also produce them by talking.  (And, because we talk a lot more than we cough, arguably, at small droplet sizes, we’re producing more in total by talking.  Which is among the reasons I called for us all to shut up in public spaces.)

The fact that droplets settle pretty quickly is the basis for the 6-foot social distancing rule.  In theory, if somebody coughs from six feet away, the droplets produced won’t hit you.

But social distancing alone is not good enough for the Chinese.  They use masks, both to absorb some of the droplets, and to slow down the velocity with which they are projected away from the source.  And they are absolutely right in doing that.  Which is why US policy is about to change.

Filtration standards and such.

An incredibly helpful and succinct discussion of single-use masks, surgical masks, and N95 respirators, along with considerable other helpful information (e.g., can you wash disposable masks) can be found at Smart Air.  If you want to get up to speed on what’s what, for actual medical supplies, that’s the place to start.

A different guide to the different types of “surgical” masks can be found at Crosstex (.pdf).  The takeaway  from both of these is that just because it looks like a surgical mask does not mean that it filters viruses and droplets like surgical mask.  It may or may not.  And if it were certified to filter like a surgical mask, it would probably not be on sale to the general public.

N95:  A filtration standard meaning that, when new and carefully tested, a mask will stop 95% of very tiny (0.3 micron or 300 nanometer) particles.  Source:  FDA.  Note that, in real life, you don’t actually achieve that because air leaks around the edge of the mask.  That’s why, for protecting the mask user, a properly-fitted mask is important.  For the mask maker, that means you need to construct it so it seals well.

But that’s the primary reason that health care professionals need N95s.  They really need to be able to stop everything, all the droplet sizes, and stop particles down the size of a few viruses stuck together.  That’s why they need those masks more than we do.

N99, P95, P99:  Additional mask standards.  First one filters 99% of particles that size.  The Ps have the indicated filtration, but are good for oily particles as well, where the Ns aren’t (I think of it as P as in paint).

How in the heck do you filter out something that small, and still be able to breathe through the mask?  Mainly, through the miracle of Melt-blown cloth.  This is a nonwoven cloth made from small-diameter plastic fibers fused together.  It is the heart of most common medical masks, including both N95 and true surgical masks.  This is the element in the mask that filters out fine particles.  And, of course, many different types are made, so you can’t just buy melt-blown cloth and assume it’s good enough for a mask.

PM2.5 (particulate matter 2.5 microns) refers to air pollution particles of around 2.5 microns in size.  In essence, PM2.5 is shorthand for all the common air pollutants that are aerosols, i.e., can stay suspended in the air.  Note that, by definition, an N95 mask filters out at least 95% of PM2.5.

PM2.5 mask.  Near as I can tell, this is not a standard.  This just means that the manufacturer claims that the mask was designed to filter out PM2.5.  It’s a mask that claims to filter out (some, all, most?) particles of that size.   Near as I have been able to tell, a) nobody tests those claims, b) a lot of PM2.5 masks and respirators don’t do what they claim (but some do), and c) near as I can tell, these are not approved for use by health care personnel.

Mask versus respirator.  Near as I can tell, anything that is truly designed to seal up against your face, so that air doesn’t leak around the device, is technically a respirator. By contrast, if it’s designed that air likely leaks around the edges and/or the principal purpose of it is to stop fluids, that’s a mask.

That’s why you’ll hear the terms N95 respirator and N95 mask used interchangeably.  Even though the thing looks like a mask, it’s supposed to function like a respirator.  You are supposed to fit the thing to your face so that air leaks are minimized.  Otherwise, really, what’s the point?

Edit:  Single-use mask versus surgical mask.  I didn’t even realize there were standards here, but see this page at Smart Air for a very helpful discussion.  Single-use face masks are typical single-layer thin masks.  If manufactured to the Chinese standard, these will stop large droplets (3 micron) fairly well, but not viruses.  The US does not have a standard for these.  The US only had standards for surgical masks:

Surgical mask standards:  BFE and PFE.  An excellent summary of the difference between single-use masks, surgical masks, and respirators can be found at Smart Air Filters.  They explain it much better than I do.

Edit:  CORRECTION.  I have now located a proper description of surgical mask standards, from Primed, which I am going to crib here.

1. BFE (Bacterial Filtration Efficiency): BFE measures how well a surgical mask mask filters out an aerosol consisting of 3 micron droplets containing staph.  In order to be certified as a surgical mask, the cloth has to filter out 95% of those droplets.  Better grades of mask (mderate and high protection masks) must filter out at least 98% of those droplets.

2. PFE (Particulate Filtration Efficiency): PFE measures how well a mask filters out virus-sized particles.  They are supposed to be tested with particles of 0.1 micron size (about the size of coronavirus).  The higher the percentage, the better the mask filtration.  Apparently, some masks are tested with somewhat larger particles, and can show a misleadingly high PPE.

Note that the actual performance of a surgical mask, in use, will not be as good as these filtration rates suggest, because the mask does not seal up against the face.  (See Mask versus Respirator).  Air leakage around the edges of the mask compromises the overall filtration.  The standards above show the filtering ability of the cloth, not the overall mask assembly as a whole.

Edit: WRONG. Note that surgical masks are NOT tested for their ability to stop penetration by bacteria in the air.  They ARE tested against penetration by fluids, and penetration by bacteria and such in fluids.  They are designed for health care workers who need to avoid being infected by fluid-borne bacteria and viruses.  As far as I can tell, that’s why a proper surgical mask, tested and certified for health care use, has some ability to filter particles.   Most (some?) are made with three-ply construction, the middle ply of which is some form of melt-blown cloth.

And, helpfully, melt-blown cloth is also used in some (but by no means all) home furnace filters ( (see this manufacturer’s page)Which gives me the segue to comparing the standards above to two that homeowners are familiar with:  MERV and HEPA.  And here, I assume you all know I am not talking about standard fiberglass, very-open-weave filters.  I’m talking about the ones that look like a sheet of fuzzy cloth.

MERV:  Minimum Efficiency Reporting Value.  From Wikipedia.

The scale is designed to represent the worst-case performance of a filter when dealing with particles in the range of 0.3 to 10 micrometers. The MERV value is from 1 to 16. Higher MERV values correspond to a greater percentage of particles captured on each pass, with a MERV 16 filter capturing more than 95% of particles over the full range.  (That little factoid will be important for tomorrow’s post.)

Do two MERV 8s make a 16?  No, absolutely not.  The MERV rating is like a pore size.  As you go up the scale, in groups of four, they start filtering smaller particles.  The 8s simply have larger pores, in effect, and let the smallest particles pass through.  That’s clear from this full explanation of MERV ASHRAE chart.  (But, if I’m reading that right, two MERV 14s, in sequence, come very close to achieving the same filtration as a MERV 16.  And two MERV 15s in sequence exceed that).

HEPA:  High-efficiency particulate arresting.  Again, from Wikipedia:

“Filters meeting the HEPA standard must satisfy certain levels of efficiency. Common standards require that a HEPA air filter must remove—from the air that passes through—at least ..  99.97% (ASME, U.S. DOE)[5][6] of particles whose diameter is equal to 0.3 μm; with the filtration efficiency increasing for particle diameters both less than and greater than 0.3 μm.[7] 

The little μm thing is microns (micrometers).

MPR:  Microparticle Performance Rating.  This is a propriety rating system developed by 3M.  It reflects the ability of a filter to capture the smallest airborne particles—from 0.3 to 1 µm in size (Wikipedia).  So you have to get the detailed 3M literature, if you want to look at Filtrete electrostatic filters:

Filtrete (r) filters differ from MERV-rated filters in that all varieties of Filtrete capture some small particles.  If you look at E1s (the smallest particles, down to 0.3 micron), it appears that you need two layers of MPR 2800 Filtrete to achieve 95% or more of filtration of E1 particles.  Although, for particles one micron and up, one layer of MPR 2800 or one layer of MPR 2500 would achieve 95% capture of those particles in a single pass.

I am uncertain as to how that Filtrete electrostatic material behaves under adverse conditions, such as when damp.

Filtration properties of common household materials. 

This is where the rubber hits the road.  Start with this article, where they actually tested cloth and made masks.

Here’s one key table, below

This is a pretty good setup, because they literally aerosolized the bacterium and virus, then tested what happened when they pushed that aerosol through a cloth panel at about the rate you would if you were breathing.  These particles are certainly on a par with the size of the coronavirus itself, and presumably the aerosol droplets are about the best proxy you are going to find for … well, aerosol droplets.

First, note the similarity of the first two columns, despite the differing size of the bacterium and the virus.  That’s because, by and large,  the masks are catching the droplets, not the individual bacteria and viruses themselves. The standard here is the surgical mask, circled in red.  I note that a vacuum cleaner bag (not stated as to type, likely not HEPA, because I’m pretty sure they are talking about cloth bags) was just about as good as a surgical mask — I put a red line there.  And a tea towel, doubled over, was just about as good.  But in the right-hand column, that’s the back-pressure you would face, in breathing through those materials.  The vacuum-cleaner bag and the doubled-over tea towel were 2 to 2.5 times harder to breathe through than the surgical mask material.

Source: Testing the efficacy of homemade masks: would they protect in an influenza pandemic? Davies A1, Thompson KA, Giri K, Kafatos G, Walker J, Bennett A.  Disaster Med Public Health Prep. 2013 Aug;7(4):413-8. doi: 10.1017/dmp.2013.43.

This is important, because the harder the material is to breathe through, the more air will leak around the mask, rather than through it, and reduce the overall filtration efficiency.

So, to be clear, yeah, you can find stuff around the home that will give as much filtration efficiency as a good surgical mask (but not an N95 mask).  But it’s going to be somewhat hard to breathe through.

And, if you care, read the section of that paper on fitting the mask.  All of the home-made masks were much leakier than a good-quality surgical mask.  That’s important, and that’s what you need to focus on if you make a mask.

Finally, how much better is it to wear a home-made mask, made out of a cotton t-shirt, than to have no mask at all, in terms of containing the spread of disease (from people coughing, in this case).  Well, they tested that empirically:  Literally had people cough, through a mask, into a sterile box, and counted the crap that came out.


Source:  Same as prior table.

Do home-made masks help prevent the spread of contagion?  Heck yes.  Look at the bottom line:  No mask, 200, homemade mask, 43, proper surgical mask, 30.  That’s the count of bacterial “colony-forming units” that they observed.  Do the math, and a home-made mask gets you (200-43/200-30) = 92% of the reduction that you would get from a standard high-quality surgical mask.

Now in case you’re surprised by that, my reading of it is that anything that stops droplets from flying is good.  You aren’t literally trying to filter out tiny little viruses.  You are trying to filter droplets, most of which you can catch with cloth, some of which fly right through.

Filtering ability of paper towels.  I didn’t find a scholarly article (and got tired of looking), but these people seem to have their act together.  It’s one of those great articles that just gets to the point.  And the short answer is that common household paper towels, do, in fact, have some measurable ability to filter out particles in the size range we are talking about.

For very tiny (0.3 micron) particles, “A single layer of kitchen paper captured just 23% particles. Adding an extra layer only increased particle capture to 33%.”

But for aerosol-sized particles (smaller than “droplets”):  “For larger 2.5 micron particles, paper towel performed better. The single layer of kitchen paper captured 52% of these larger particles”.

(From that, I would infer that two layers would get 75% or so.)

Those same folks provide a nice graphic re-write of the article on home-made masks, at this location.

Near as I can tell, nobody has done the one I want to see, which is a coffee filter.  I use(d) those as a pre-filter when purifying raw water when camping, and I think that’s a pretty common use.   I did test that you could breathe through one (possible, but a lot of resistance).  I will keep looking for that one.

Bottom line:  Wear ’em if you’ve got ’em.  If not, buy one and modify it.  If not, make one from scratch.  Any mask is better than no mask.  Even paper towels have some filtering efficiency. 

Next post is about buying and modifying masks.

Final post will be about making masks.  But you can just go on Amazon and get a free Kindle download on that.