Cloth Masks: Supplemental References to Studies

Ania Mitros, 18 January 2021

This is a supplement to an article on cloth masks. This page notes specifically which scientific publication supports which claim in the article.

Fit is key

Multiple publications have shown this: A leaky mask does about half as much as a snug-fitting mask.

  1. Clapp: A medical procedure mask worn by a volunteer filtered 39% of particles when worn normally, and filtered 80% of particles when nylon hosiery was added on top to eliminate all gaps. Improving fit improved filtering by a factor of 2.
  2. Davies: Evaluation is not relevant to Cub Reporter article
  3. Jang: Not tested
  4. Konda: A gap causes a drop in filtration efficiency from 99.9% to 12% for N95 fabric; 99.6% to 44% for surgical mask; 98.5% to 32% for cotton/silk. Note that the flow rates used in this study were very low, so the way the air moved (diffusion vs convection) may not be representative of mask usage.
  5. Lindsley: Did not disentangle fit from fabric.
  6. Lustig: Not tested.
  7. O'Kelly: Not tested.
  8. Rengasamy: Not tested.

Fabric matters: Cotton woven

  1. Clapp: A cotton bandana filtered at 49% when doubled bandit-style, and at 50% when folded in a multi-layer rectangle as recommended by the U.S. Surgeon General.
  2. Davies: A "Pillowcase" filtered 61.28% of a bacterium and 57.13% of a virus, but the content of the fabric was not stated.
  3. Jang: Not tested
  4. Konda: Flawed study.
  5. Lindsley: Not tested
  6. Lustig: Two layers of Kona cotton filtered 61% as much as an N95; three layers 66%; four layers 110% (slightly better than an N95). Lustig had non-intuitive and surprising results for the thicker and more rigid samples, and I suspect some problem with sealing around ridgid samples, so I'm not sure whether the four layer result is to be trusted.
  7. O'Kelly: Heavyweight woven cotton filtered 68.2% as much as N95 material; quilting cotton 67.7%; shirting cotton 64.0%; lightweight woven cotton 57.6%.
  8. Rengasamy: Unclear whether any tested fabric was a woven cotton.

Fabric matters: Cotton knit

  1. Clapp: A 3-layer cotton knit mask which in the photo appeared to fit well filtered 27% of particles.
  2. Davies: A "100% cotton T-shirt" filtered 69.42% of a bacterium and 50.85% of a virus, with similar pressure drop as a surgical mask. When doubled, the T-shirt filtered 70.66% of the bacterium, nearly the same as a single layer.
  3. Jang: Not tested
  4. Konda: Not tested.
  5. Lindsley: Three-layer cotton face mask (Hanes Defender) blocked 51% (as worn).
  6. Lustig: Heavy t-shirt filtered 52% as much as an N95; two layers of heavy t-shirt 88% .
  7. O'Kelly: A folded sock (cotton/lycra) filtered 67.4% as much as N95 material; heavyweight cotton t-shirt 48.1%; cotton jersey knit 46.8%.
  8. Rengasamy: Single layer cotton t-shirt filtered 14% (very small particles, 20nm to 1um). Cloth masks filtered 10% to 25%. Of three sweatshirts, two filtered 18-30%, and one (Hanes) filtered 60%.

Fabric matters: Non-woven synthetic filters

  1. Clapp: A nonwoven polypropylene mask (not a medical grade mask) filtered 29%, suggesting that a single layer of nonwoven polypropylene does not make a great mask. The addition of a non-woven insert to a 2-layer nylon mask improved filtering from 56.3% to 74.4%.
  2. Davies: Not tested
  3. Jang: Not tested
  4. Konda: Not tested.
  5. Lindsley: Not tested
  6. Lustig: Tested several types of interfacing, both non-woven polyester and non-woven polypropylene. Many of the combinations would make for an unusually stiff mask, and the results for those had high standard deviations, so I question the fidelity of those results. The one result that had a low standard deviation was also intuitively sensible, and that was the combination of woven cotton plus Pellon 931TD midweight non-woven polyester interfacing, which filtered 65% as much as an N95. This result (cotton + 931TD) is very similar filtering to 3 layers of cotton woven, but should be much more breathable.
  7. O'Kelly: HTC fusible interfacing filtered 28.6% as much as N95 material.
  8. Rengasamy: Not tested

Fabric matters: Polyester and nylon

  1. Clapp: A 2-layer knit nylon mask filtered 56% when new and at 79% after washing; how washing improved performance was not understood. A 1-layer knit polyester/nylon mask filtered 39% of particles. A 1-layer knit gaiter filtered 38%.
  2. Davies: Not tested.
  3. Jang: Three 100% different polyester masks filtered at 18.5%, 9.5%, and 23.6%. When doubled, the filtration improved from 18.5% to ~50%, and from 9.5% to ~45%. These filtration numbers are for 0.3-0.5um particles, where the masks perform most poorly. Considering 5-10um particles, the 18.5% mask filters at ~60%; the 9.5% mask filters at ~60%, and the 23.6% mask filters at ~90%. Washing worsened the filtration rate by a factor of 1.04 to 4.
  4. Konda: Not tested.
  5. Lindsley: FKGIONG Sun UV Protection Neck Gaiter (95% polyester, 5% spandex) blocked 47%, and when doubled 60% (as worn).
  6. Lustig: Not tested.
  7. O'Kelly: 100% nylon woven filtered 52.6% as much as N95 material. Lycra (82% nylon, 18% spandex) 41.2%.
  8. Rengasamy: Not tested

Fabric matters: Procedure masks

  1. Clapp: With all gaps sealed by a nylon stocking, a procedure mask filtered 80% of particles. Worn normally, the same mask filtered at 38.5%.
  2. Davies: Surgical mask filtered 96.35% of a bacterium and 89.52% of a virus.
  3. Hossain: Surgical masks filtered 65% ,79%, 98%. N95 masks filtered >95% when new, 75% after washing in washing machine. Paper focuses on recharging discharged masks.
  4. Jang: Not tested
  5. Konda: Flawed study.
  6. Lindsley: Medical procedure mask blocked 59% (as worn).
  7. Lustig: Procedure cone mask (Cardinal Health #AT7509) filtered 82% as much as an N95. Duck bill surgical mask (Halyard #37525) filtered 33% as much as an N95.
  8. O'Kelly: A surgical mask filtered 90.5% as much as N95 material.
  9. Rengasamy: Not tested

Fabric matters: Silk

  1. Davies: Filtered 58.00% of a bacterium and 54.32% of a virus, significantly worse than cotton.
  2. Konda: Flawed study.

Short note on each study

  1. Clapp: Tested filtration of 0.05um NaCl particles by masks worn by a volunteer. Tested 7 cloth masks, plus a medical procedure mask in 6 configurations to modify fit.
  2. Davies: Filtration efficiency of two different aerosolized microorganisms (a bacterium and a virus) was tested across 10 different fabrics, with both filtration and pressure drop across the fabric reported. Table 1 contains the key results.
  3. Jang: Article is mostly in Korean, with only the abstract and figure captions in English. Evaluated filtration as a function of particle size. Cloth masks filter 5-10um particles far better than 0.3-0.5um particles, though error bars are quite large for the larger particle sizes measured. Filtration worsens after washing by factor of 1.04 to 4.0.
  4. Konda: Big oops! They didn't recognize that airflow is significantly reduced by fabric samples and did not monitor air flow rate during testing, only when. Flow rates are about an order of magnitude lower than intended and not known. This renders the data possibly meaningless. Which is disappointing because they studied interesting materials including silk and chiffon.
  5. Lindsley: Tested filtration of 0 to 7um particles by various masks worn by a manequin. The air flow was an emulated cough, with a single rapid expulsion of air and aerosol (KCl and sodium fluorescein) expelled at a rate similar to a cough.
  6. Lustig: Tested filtration of aerosols containing fluorescent virus-sized particles by fabric samples. Large numbers of fabrics tested, including combinations. For many fabrics the results look good. Some of the more rigid fabrics (ex. denim) produced unintuitive results, which may suggest an error such as poor sealing and leakage with the thicker, more rigid samples. Measured how many particles got through but not how many were generated, so best way to compare to other studies is to related the results for fabric of interest to the results for N95 masks.
  7. O'Kelly: Measured filtration of aerosolized NaCl particles of 0.02um to 0.1um by fabric samples at a much higher air flow rate than most other studies. An N95 mask filtered ~53%, so the fast air flow rate seems to affect the results even for this standard baseline material. Looked at dry and damp fabrics.
  8. Rengasamy: Tested filtration of very small particles, 20nm to 1um, through single layer fabric samples. NaCl aerosol. Measured both filtering and pressure drop.
  9. Plana: Tested KN95 masks. Of 15 tested N95 masks, 9 did not filter >95%, 12 filtered >70%.

Particle sizes

Most studies summarized in the article tested with sub-micron particle sizes, Lindsley being the key exception and using particles up to 7um. For comparison: SARS-CoV-2 is about 0.1um, the N95 standard refers to 0.3um particles, and COVID-19 transmission is thought to involve water droplets of at least a few microns. Fabric filters large droplets better than small ones, so the performance of cloth masks for filtering droplets of a size relevant to viral transmission may in practice be better than the results of these studies. Jang [9] shows plots of filtration as a function of particle size, and filtration improves significantly with particle size. For three 100% polyester masks:

Mask Fabric Filtration at 0.3-0.5um Filtration at 5-10um
Mask C100% polyester (cool comfort fabric)18.5% ~60%
Mask D100% polyester (microfibre) 9.5% ~60%
Mask E100% polyester (microfibre) 23.6% ~90%

  1. Clapp: NaCl particles with median diameter of 0.05um (range 0.02-0.60um)
  2. Davies: Bacterium (0.95-1.25um) and a virus (23nm).
  3. Hossain: Measured 0.3um particles.
  4. Jang: Polydisperse NaCl particles, measured 0.3 - 10um (five channels).
  5. Konda: Flawed study due to grossly incorrect air flow rate. Tested with NaCl particles in the range of a few tens of nanometers to approximately 10um. Measured multiple particle sizes from 10nm to 6um.
  6. Lindsley: KCl and sodium fluorescein particles 0-7um.
  7. Lustig: Fluorescent, virus-like nanoparticles emulating the size and surface characteristics of SARS-CoV-2 (10nm to 200nm).
  8. O'Kelly: NaCl particles, measured concentrations of particles between 0.02 and 0.1um.
  9. Rengasamy: NaCl particles in 0.02-1.0um range.
  10. Plana: KCl particles, measured 0.3-10um particles.

References

  1. Clapp: Phillip W. Clapp, PhD; Emily E. Sickbert-Bennett, PhD, MS; James M. Samet, PhD, MPH. Evaluation of Cloth Masks and Modified Procedure Masks as Personal Protective Equipment for the Public During the COVID-19 Pandemic. JAMA Intern Med. Published online December 10, 2020. doi:10.1001/jamainternmed.2020.8168
  2. Clase: Catherine M. Clase, Edouard L. Fu, Aurneen Ashur, Rupert C.L. Beale, Imogen A. Clase, Myrna B. Dolovich, Meg J. Jardine, Meera Joseph, Grace Kansiime, Johannes F.E. Mann, Roberto Pecoits-Filho, Wolfgang C. Winkelmayer, Juan J. Carrero. Forgotten Technology in the COVID-19 Pandemic: Filtration Properties of Cloth and Cloth Masks-A Narrative Review Mayo Clinic Proceedings, VOLUME 95, ISSUE 10, P2204-2224, OCTOBER 01, 2020
  3. Davies: Davies A, Thompson K-A, Giri K, Kafatos G, Walker J, Bennett A. Testing the efficacy of homemade masks: would they protect in an influenza pandemic? Disaster Med Public Health Prep. 2013;7(4):413-418.
  4. Hossain: Emroj Hossain, Satyanu Bhadra, Harsh Jain, Soumen Das, Arnab Bhattacharya, Shankar Ghosh, and Dov Levine. Recharging and rejuvenation of decontaminated N95 masks. Physics of Fluids 32, 093304 (2020); https://doi.org/10.1063/5.0023940
  5. Jang: Jang JY, Kim SW. Evaluation of filtration performance efficiency of commercial cloth masks. J Environ Health Sci. 2015;41(3):203-215.
  6. Konda: Konda A, Prakash A, Moss GA, Schmoldt M, Grant GD, Guha S. Aerosol filtration efficiency of common fabrics used in respiratory cloth masks [published correction appears online ahead of print in ACS Nano, June 18, 2020; doi: 10.1021/acs- nano.0c04676]. ACS Nano. 2020;14(5):6339-6347.
  7. Lindsley: William G. Lindsley and Francoise M. Blachere and Brandon F. Law and Donald H. Beezhold and John D. Noti. Efficacy of face masks, neck gaiters and face shields for reducing the expulsion of simulated cough-generated aerosols. Aerosol Science and Technology Dec 11, 2020. https://doi.org/10.1080/02786826.2020.1862409
  8. Lustig: Steven R. Lustig, John J. H. Biswakarma, Devyesh Rana, Susan H. Tilford, Weike Hu, Ming Su, and Michael S. Rosenblatt. Effectiveness of Common Fabrics to Block Aqueous Aerosols of Virus-like Nanoparticles. ACS Nano 2020, 14, 6, 7651-7658. https://doi.org/10.1021/acsnano.0c03972
  9. Morawska: Morawska, L. Droplet fate in indoor environments, or can we prevent the spread of infection? Indoor Air, Volume 16, Issue 5, October 2006, 335-347
  10. O'Kelly: Eugenia O'Kelly, Sophia Pirog, James Ward, P John Clarkson. Ability of fabric face mask materials to filter ultrafine particles at coughing velocity. BMJ Open 2020;10:e039424. doi:10.1136/bmjopen-2020-039424
  11. Rengasamy: Rengasamy S, Eimer B, Shaffer RE. Simple respiratory protec- tiondevaluation of the filtration performance of cloth masks and common fabric materials against 20-1000 nm size particles. Ann Occup Hyg. 2010;54(7):789-798.
  12. Plana: Deborah Plana, Enze Tian, Avilash K. Cramer, Helen Yang, Mary M. Carmack, Michael S. Sinha, Florence T. Bourgeois, Sherry H. Yu, Peter Masse, Jon Boyer, Minjune Kim, Jinhan Mo, Nicole R. LeBoeuf, Ju Li, Peter K. Sorger. Assessing the quality of nontraditional N95 filtering face-piece respirators available during the COVID-19 pandemic. medRxiv, 27 July 2020, doi: https://doi.org/10.1101/2020.07.25.20161968