While fitted N95 masks allowed zero droplets to pass through, the three-layered surgical masks too performed very well with 0-10% droplet transmission. N95 masks with exhalation valve fared poorly. Double-layered cotton masks were better that single-layered ones.

Finally, the efficiency of N95, surgical masks and 12 other masks made of different materials to reduce droplet emission while speaking has been scientifically validated. In a study published in Science Advances, researchers from Duke University found that the fitted N95 masks without the exhalation valve, which health workers generally use to prevent viruses and bacteria from passing out of the mask or entering it, was the best with almost zero droplets passing through the mask.

The three-layered surgical masks too performed very well with droplet transmission fraction varying between zero and 10%. These are medical grade masks that health workers typically use.

A two-layered mask made of polypropylene and a three-layered mask with polypropylene sandwiched by cotton layer on either side too performed well with droplet transmission fraction of about 10%, which is comparable with surgical masks.

Two-layered cotton masks of different types had better efficiency in cutting out droplet transmission than a single-layered cotton mask. The two-layered cotton masks had droplet transmission fraction between 10 to 20%, while the single-layered cotton mask had droplet transmission fraction 20%. Clearly, the singe-layered cotton mask was a bit inferior compared with the two-layered cotton mask.

Mask made of knitted material fared poorly compared with either single- or two-layered cotton masks, with droplet transmission fraction over 30%, while bandana fared even worse with droplet transmission fraction of 50%.

While the control with no mask had a relative droplet count of 100%, neck fleece with droplet transmission fraction of 110% performed worse than even not wearing a mask. Researchers found that when the volunteer was wearing the neck fleece and speaking more droplets were produced than when not wearing a mask.

“Some mask alternatives, such as neck fleece or bandanas, offer very little protection,” they write.

Why neck fleece performed poorly

Explaining why the neck fleece performed worse than even not wearing a mask, the researchers say that when the volunteer speak through the fleece, the largest droplets are dispersed into smaller droplets, thus increasing the number of droplets compared with no masks. It is not just the increased number of droplets but the generation of smaller droplets that make the fleece counterproductive. Smaller particles tend to be airborne for longer time than large droplets and if such small droplets contain coronaviruses then it greatly increases the risk of others inhaling it even at farther distances. Also, smaller droplets tend to reach deep inside the lungs than larger droplets.

The N95 masks with exhalation valves had a droplet transmission fraction less than 20%. N95 masks protects the person wearing the mask but by allowing the droplets to cross the mask through the valve, others standing close to the person wearing N95 mask with a valve are at higher risk.

“In summary, our measurements provide a quick and cost-effective way to estimate the efficacy of masks for retaining droplets emitted during speech for droplet sizes larger than 0.5 micron,” they write.

“We want to emphasise that we really encourage people to wear masks, but we want them to wear masks that actually work,” Martin Fischer, one of the authors of the study, told CNN. “This is a very powerful visual tool to raise awareness that a very simple masks, like these homemade cotton masks, do really well to stop the majority of these respiratory droplets.”

How they performed the experiment

To evaluate the performance of each mask, a volunteer wearing a face mask was made to speak in the direction of an expanded laser beam inside a dark enclosure. The person wearing a mask was made to repeat the phrase, “stay healthy, people,” five times. Any droplet that passes the mask spreads and comes in line with the laser beam. The droplets scatter the light, which is captured by a mobile phone camera. A simple computer algorithm was used to count the droplets in the video. Each volunteer spoke the phrase wearing each of the 14 masks tested. Only one mask was worn at a time. For reference, the researchers conducted control trials where the speakers did not wear wore any mask.

“The experimental setup is simple and can easily be built and operated by non-experts,” they write.