Simple Tool Evaluates Mask Performance at Blocking Droplets

The impetus for the project came when an out-of-work ballet costume designer sent Eric Westman, a specialist in obesity medicine at Duke University Health System, a message on social media. She had heard that Westman was working to source and manufacture masks for vulnerable citizens, and she wanted to offer her team of unemployed costume makers. “These people who are used to using these fabrics for ballroom costumes . . . were now using this beautiful material” to make masks, Westman says.

See “How Face Masks Prevent the Spread of COVID-19”

Westman needed a simple way to test the effectiveness of the different materials, so he picked up the phone and called the physics department at Duke University. Martin Fischer, an expert in biomedical imaging, took up the challenge, bringing in his college-aged daughter so the two could work closely without fear of passing the virus to others.

Together, they built a simple tool out of a cardboard box, a laser, a lens, and a smartphone camera capable of capturing visible evidence of airborne particles. Fischer used a cylindrical lens to stretch the pencil-like beam of the laser into a thin “light sheet,” passing it through slits in the sides of a blackened box. When a person speaks into the box, the particles cross through the light sheet, scattering the light in different directions. A camera on the other end records that scatter, which appears on the video as winking green lights, and a simple computer algorithm tallies the number of droplets and the rate at which they pass through the sheet. The whole setup costs just under $200.

“I think it’s really important what they were able to do in terms of using a laser light scattering method that can be implemented cheaply and easily, so that people can really see the effect,” Christina Bax, a medical student at the University of Pennsylvania who was not involved in the research, tells The Scientist. Earlier this year, Bax had worked with colleagues at NIH who developed a similar technique, and found that when speaking, “everybody has saliva droplets flying everywhere.”

Fischer’s team tested the ability of 14 different masks to dampen the spread of those droplets, including the medical grade N95s (with and without valves) used by frontline healthcare workers, disposable blue surgical masks, hand-sewn cotton masks, neck fleeces, and bandanas, among others. A single person spoke the phrase “stay healthy, people” five times in a row into the box, a process that was repeated 10 times for each mask and for the no-mask control.

See “Scientists Urge Consideration of Airborne SARS-CoV-2 Transmission”

The results, Fisher says, were “an eye opener for how many of those particles are floating around” even when people are just talking. When speaking without masks, Westman and Fischer found that people exhale hundreds of drops per second, many of which are too small to be detected by the naked eye but still flashed green inside the box.

Among the many masks they tested, the N95 with no valve performed best, with only 0.1 percent of the particles making it through the light sheet; the version with the valve may do a good job of protecting the wearer, but it still allows particles to escape when a person breathes out. Hand-sewn cotton masks composed of at least two layers also performed well, as did single-use surgical masks, vastly decreasing the number of particles spreading into the air while talking.

But cotton bandanas and Buff-type neck gaiters made of spandex performed poorly, and in fact  sometimes caused more drops to pass across the beam than had been detected in the control. The authors attribute this finding to a sort of cheese grater effect: larger particles passing through the porous material were broken up into smaller aerosols. “I came in thinking anything is better than nothing,” Westman says. “But the idea that it’s a mesh can change the dynamic.”

This study is only a proof of concept meant to show the utility of the laser setup and not to draw broad conclusions about the efficacy of different masks. The researchers sampled only what was at hand and used only one speaker to test all 14 types. It is likely, the authors say, that variations between individuals based on tone, volume, enunciation, and even language, as well as differences in fabric quality and the fit of a mask, might change their findings. If backed up by further study, the results suggest that scarce N95 masks are best left for frontline workers, as average citizens can make do with masks made at home.

In addition to the relatively small sample size, Shenoy adds, it is possible that not all droplets were captured in the count due to the experimental set up and limitations in the detection sensitivity of the cell phone camera, especially for smaller particles. 

See “How Our Exhalations Help Spread Pathogens Such as SARS-CoV-2”

The authors, who are currently attempting to patent their new product, envision it being used in public community centers or museums as an educational tool to show people why masks are so important and how best to wear them. In addition, it may be useful for companies manufacturing masks to easily test the efficacy of their product, or in resource-poor countries as a cheaper alternative to costly laser particle sorters. According to Bax, “it comes back to this idea of being able to visualize what is happening, visualize the droplets, and understand that everybody spits when they speak.”