In this guest blog post Ania Mitros, Ph.D. discusses the laboratory testing data and science behind her preferred fabric combination for masks: two layers of ~65 gsm (gram/m²) spunbond nonwoven polypropylene (NWPP) plus one layer of ~130 gsm woven cotton. Her recommendation comes after reading dozens of scientific publications on the subject and investing thousands of dollars to test the filtration and performance of widely available mask materials. This post brings you some of her most recent findings.
My quest for an effective fabric mask led to a fabric combination that improves on the ubiquitous 2-layer cotton mask, with both better filtration and better breathability.
|I recommend:||~130gsm cotton woven plus ~130gsm of NWPP|
|for example:||1 layer Spoonflower sateen cotton plus 2 layers OlyFun|
|or:||1 layer Kona cotton plus 2 layers SmartFab Double-Thick|
This washable combination strikes a balance between filtration and breathability. Filtration refers to how many particles are blocked by the fabric, a key performance parameter. Breathability or pressure drop refers to how easily air flows through the fabric. A mask that does not breathe well enough may be uncomfortable and prone to leak around the edges of the mask.
NWPP: NWPP stands for non-woven polypropylene, and in this post refers to spunbond NWPP. Most of the data herein used OlyFun, a 65gsm spunbond non-woven polypropylene (NWPP). Two layers of OlyFun total ~130gsm. SmartFab Double-Thick NWPP (~67 gsm) is another popular option and two layers should be an equivalent substitution (totaling ~134 gsm). Regular SmartFab is about 40gsm so 3 layers, totals about 120 gsm, which should be a reasonable replacement for 2 layers of OlyFun.
I sent five fabric samples for filtration testing at Colorado State University (CSU) in the lab of Professors Volckens and L’Orange. My recommended fabric combination (cotton sateen plus two layers of OlyFun) is shown with yellow triangles (Δ).
My recommended combination is second only to the combination of flannel and OlyFun shown in green squares (□). However, while flannel improves filtration it has poor breathability. Masks with poor breathability are prone to unfiltered leakage around the edges of the mask which can negate any filtration advantage. I will discuss this further in the Breathability section.
In grey are measurements from two different 2-layer cotton samples tested by this same lab. The performance of these 2-layer cottons is poor for particles <2μm, which is a particle size that scientists believe to be significant for SARS-CoV-2 (Covid-19) transmission [Wang 2021]. In blue are samples of OlyFun which filters similarly to cotton but, as discussed later, has excellent breathability.
The combination of cotton and NWPP performed better than NWPP alone, good news for those of us who enjoy a pretty mask.
Why test filtration with 0.5μm to 12μm particles?
Two reasons underlie this choice of particle sizes for filtration testing. For one, particle sizes matter a lot for filtration and this range is where fabrics have the most interesting behavior, transitioning from poor filtration of submicron particles to nearly complete filtration of 10μm and larger particles. Secondly, the science that I’ve read has led me to believe that 1μm to 10μm particles are most likely to dominate transmission of Covid-19, though this is still an active field of research [Mitros 2021, Wang 2021].
For comparison: The SARS-CoV-2 virus that causes COVID-19 is about 0.1μm, but tends to be transmitted within larger particles of other matter (salt and proteins from mucus, mostly). Both the NIOSH (for N95 respirators) and the ASTM F3502 (barrier face coverings) standards require testing with 0.3μm particles, which are the most difficult size to filter. The N95 respirator was originally developed for industrial settings where particles of any size may present a hazard.
To dig deeper into the science, you can read my MakerMask blog post “Fabric Mask Filtration Overview“. To read a similar conclusion from several scientists in the field, also aimed at lay people, check out FAQ-aerosols.
Breathability is NWPP’s superpower.
How breathable a mask feels is a combination of several factors, key among them being how hard we need to inhale and exhale, which can be measured as the pressure drop across a fabric sample, also referred to as air flow resistance. Also important for comfort and breathability are how much the fabric accumulates moisture and the area through which air flows. For example, a mask that collapses onto the face forcing all the air through a small area near the nostrils will feel less breathable than a mask that stays away from the face and provides a larger area for airflow.
Measurements at CTT Groupe (an accredited commercial testing lab) show that NWPP (in this plot, OlyFun) is about 4 times more breathable than quilting cotton (here, Kona brand).
Breathability is why I recommend 1 layer cotton woven plus 2 layers OlyFun (in yellow) over 1 layer cotton flannel plus 2 layers OlyFun (in green), even though incorporating flannel improves filtration. As you can see in the following data, measured at Colorado State University (CSU), 4 layers of OlyFun have less breathing resistance (i.e., are easier to breath through) than 1 layer of sateen woven cotton. And 1 layer of sateen woven cotton has half the resistance of 1 layer of cotton flannel.
Note that CSU and CTT Groupe use different air flow rates, so the two bar graphs cannot be directly compared. If you are nonetheless tempted, note that pressure drop is directly related to air velocity, so the pressure drops measured at 4cm/s should be roughly 4/6 of those measured at 6cm/s. For example, CTT’s data on 2-layer Kona shows a pressure drop of ~6 mmH2O, so 1 layer at 6cm/s would be ~3 mmH2O, and converting that to 4 cm/s we would expect 1-layer Kona to have a 2 mmH2O pressure drop which is actually pretty close to what CSU measured for cotton sateen.
MakerMask’s previous testing also showed the superior breathability of NWPP to cotton. Comparisons of the breathability of fabric samples y of complete masks both showed that 3 layers of NWPP (NWPPx3) were about three times more breathable than a 3 layer stack-up of cotton-NWPP-cotton.
Washing & Filtration
Laundering spunbond NWPP seems to be just fine. In the Colorado State University (CSU) data, OlyFun filtration improved after 13 wash/dry cycles on the default (“Normal”) setting in a front-loading LG washer and dryer with Seventh Generation free & clear laundry detergent. Filtration of identically washed SmartFab Double-Thick is shown in the graph for comparison to washed OlyFun, and both have nearly the same filtration after washing.
This data is consistent with results from a previous study with different particle sizes that showed that unwashed SmartFab Double-Thick and OlyFun filtered nearly identically, and washing did not affect filtration.
For MakerMask’s recommendations on washing and drying masks with NWPP, see ”Washing Fabric Masks: Knowns and Unknowns”.
To allow sewists to buy the same fabrics as tested in this study, following are links to manufacturer websites.
|Kona||Kona cotton (Camellia)||A typical 100% cotton quilting fabric widely used in the sewing community. Roughly 100μm gaps between fibers are visible in microscope photos.||147.5 gsm (measured)
(specified: 4.35 oz per sq yard = 147.5gsm)
|Sateen||Spoonflower Organic Cotton Sateen||100% organic cotton, weave has smaller gaps than Kona as seen in microscope photos.||130 gsm (Specified: 3.8 oz per square yard)
142 gsm (Measured)
|Flannel||Robert Kaufman cotton flannel||100% cotton flannel||169.5 gsm (Specified)
5 oz per sq yard
|OlyFun||OlyFun||Non-woven polypropylene (NWPP) intended for crafting. For published evidence that NWPP makes a good filter see https://makermask.org/masks-for-covid-literature-highlights/.
Note that regular SmartFab is only 40gsm.
|65 gsm (specified)|
|SmartFab||SmartFab DoubleThick (white)||66.5 gsm (measured)|
Sewists have been making masks for over a year and a half now, much of that with limited science-based guidance on mask construction. Multiple labs have tested fabrics, yet many fabrics tested in scientific publications could not be sourced by sewists either because they were described too vaguely or were simply not available to sewists. This data is my contribution to empowering us to choose our fabrics in a more scientifically informed way.
References and Further Reading
- CSU’s test methods are described in the following paper, except the flow rate is changed to 15 L/min. https://pubs.acs.org/doi/full/10.1021/acs.est.0c07291
- My previous data from tests of fabric sample filtration efficiency of submicron NaCl particles, similar to the NIOSH tests for N95 respirators; and of Bacterial Filtration Efficiency (BFE), which is how medical procedure masks are tested. https://ofb.net/~ania/masks/
- Wang 2021: Wang, C., Prather, K., Sznitman, J., Jimenez, J., Lakdawala, S., Tufekci, Z., Marr, L., “Airborne transmission of respiratory viruses”, SCIENCE, 27 Aug 2021, Vol 373, Issue 6558, DOI: 10.1126/science.abd9149
- Colorado State University (CSU), lab of Professors Volckens and L’Orange: jv.colostate.edu/masktesting.
- CTT Groupe: https://gcttg.com/en/