Performance of fabrics for home-made masks against the spread of COVID-19 through droplets: A quantitative mechanistic study

Abstract

Coronavirus Disease 2019 (COVID-19) may spread through respiratory droplets released by infected individuals during coughing, sneezing, or speaking. Given the limited supply of professional respirators and face masks, the U.S. Centers for Disease Control and Prevention (CDC) has recommended home-made cloth face coverings for use by the general public. While there have been several studies on aerosol filtration performance of household fabrics, their effectiveness at blocking larger droplets has not been investigated. Here, we ascertained the performance of 11 common household fabrics at blocking large, high-velocity droplets, using a commercial medical mask as a benchmark. We also assessed the breathability (air permeability), texture, fiber composition, and water absorption properties of the fabrics. We found that most fabrics have substantial blocking efficiency (median values >70%). In particular, two layers of highly permeable fabric, such as T-shirt cloth, blocks droplets with an efficiency (>94%) similar to that of medical masks, while being approximately twice as breathable. The first layer allows about 17% of the droplet volume to transmit, but it significantly reduces their velocity. This allows the second layer to trap the transmitted droplets resulting in high blocking efficacy. Overall, our study suggests that cloth face coverings, especially with multiple layers, may help reduce droplet transmission of respiratory infections. Furthermore, face coverings made from materials such as cotton fabrics allow washing and reusing, and can help reduce the adverse environmental effects of widespread use of commercial disposable and non-biodegradable facemasks.

Keywords: Breathability; COVID-19; Cloth face covering; Droplet blocking; Face mask; Respiratory droplets; SARS-CoV-2.

Fig. 1

Distinction between aerosol filtration and large droplet blocking by fabrics. (A) Typical mechanisms of particle capture and transport during aerosol filtration: Particles 1, 2, and 3 are captured by the fiber via interception, impaction, and diffusion, respectively. Particle 4 is smaller than the inter-fiber spacing and is transmitted through the fabric, carried by air flow. Particle 5, being larger than the inter-fiber spacing, is captured by straining. Particle 6 is subsequently captured by settling/caking. (B) Blocking of nanoparticles carried by large droplets. Top and bottom rows represent transmission through hydrophilic and hydrophobic fabrics, respectively. Droplets impact the fabric with high velocity, squeeze through the pores, and part of the volume can transmit. This process involves energy costs associated with interfacial energies and shear stresses, which may be influenced by fabric porosity, fabric type, and viscosity of the droplet. Energy barriers for transmission increase with decreasing pore size, increasing droplet viscosity, as well as hydrophobicity of the fabric. For example, interfacial energy barrier for transmission through hydrophobic fabric is much higher than that for hydrophilic one.

Fig. 2

Samples used in this study. (A) Image of a medical/dental quality FM-EL style face mask with (B) 3-layer construction, which was used as a benchmark. (C) Microscopic texture of the outer surface of the medical mask and the 11 different home fabric samples. All scale bars: 1 mm.

Fig. 3

Droplet challenge tests. (A) Schematic of the experimental method and (B) image of lab set-up. (C) Box plots showing incident droplet velocities at various distances from the inhaler nozzle, and exit velocities of droplets that penetrate medical mask and single and double-layered T-shirt fabric (fabric 6). *p<0.001, two-sample t-test comparing exit velocities of penetrating droplets to incident velocity (measured within 25 mm from the nozzle, i.e., leftmost box plot). (D) High-speed snapshots of droplets hitting and penetrating the medical mask. Scale bars: 10 mm. (E) Brightfield and fluorescence images of droplets collected on a petri dish. Scale bars: $100\mathrm{\mu }\mathrm{m}$. (F) Confocal images of homogenized bead collection; representative images from samples with high and low bead density. Scale bars: $100\mathrm{\mu }\mathrm{m}$.

Fig. 4

Droplet blocking efficiency. Box plots showing the distribution of droplet blocking efficiencies for different fabrics at (A) 25 mm and (B) 300 mm from the nozzle, representing efficiencies against high velocity and low velocity droplets, respectively. For 3 layers of fabric 6 at 25 mm and 2 layers of fabric 6 at 300 mm, measurements were below detection limit, therefore, only a line corresponding to the estimated lower bound is presented for these samples.

Fig. 5

Fabric breathability measurement. (A) Schematic of the experimental method and (B) image of plug flow tube set-up. (C) Flow velocity vs. pressure plots for selected samples. (D) Breathability measurement results for all samples. (E) Fabric porosity vs. breathability for woven and knit fabrics, and (F) droplet blocking efficiency vs. breathability plots with regression lines.

Fig. 6

Analysis of high impact droplet resistance by T-shirt fabric and medical mask. (A) Schematic representation of the processes during high-speed impact of droplets on medical mask and cotton T-shirt fabric (fabric 6). Note that the gap between the 2-layered fabric is exaggerated to highlight the droplets between the layers. Images show water droplets on medical mask and T-shirt fabric. While the mask is highly hydrophobic, T-shirt fabric is hydrophilic. (B) Impact response of various samples. Top row: medical mask, middle row: 1 layer of T-shirt fabric, bottom row: 2 layers of T-shirt fabric. It is apparent that mask material does not bend much, compared to the T-shirt fabric samples that undergo extensive bending deformation due to impact. Scale bars: 10 mm.