Scrutable

https://medicalxpress.com/news/2021-12- ... masks.html

Although the detailed analysis by the Max Planck researchers in Göttingen shows that tight-fitting FFP2 masks provide 75 times better protection compared to well-fitting surgical masks and that the way a mask is worn makes a huge difference; even medical masks significantly reduce the risk of infection compared to a situation without any mouth-nose protection at all. "That's why it's so important for people to wear a mask during the pandemic," says Gholamhossein Bagheri. And Eberhard Bodenschatz adds, "Our results show once again that mask-wearing in schools and also in general is a very good idea."

https://www.pnas.org/content/118/49/e2110117118

This study combines interesting results with useless theory. The interesting measurements are in the appendix (https://www.pnas.org/content/suppl/2021 ... pplemental) and were measurements of particle penetration of masks as worn by real people. Unfortunately, only seven people were measured. It found that masks may substantially reduce the quantity of particles in inhaled air. However it also found:

D. Inward leakage variability for the same subject. The cases (i), (ii) and (iv) were tested two separate times for two out of the
seven subjects (Fig. S10), the mask was taken off in between measurements. We found, that the total inward leakage of each
mask case and therefore the fit varies from time to time even if the mask is donned by the same subject. As mentioned in the
main paper, the results vary by 40% for the FFP2 mask without adjustment and up to factor 16 (factor 16 for one subject
that also stated an unusually high fit of the taped mask (always below 7%), and factor 2 for the other subject (always below
2%)) for the FFP2 mask with adhesive tape on the nosepiece from time to time for the same mask/subject combination. The
relative variability increases with better fitting mask cases.

Such variability means that any conclusions drawn from only seven subjects are going to be very weak. I would like to see the experiments repeated at a larger scale. The scenario of using 1× 12 cm double-sided 3M Medical Tape 1509 to improve the seal is interesting for people who want to get the best possible protection for themselves, but if transferred to the general context of telling the public to wear a mask should be considered merely as a cause for hilarity, given the number of people you see in public whose masks have huge gaps around the edges or not even covering the nose.

The main paper is essentially a theoretical analysis. It is also seriously flawed in that it assumes that infectious material travels in a straight line directly from an infected person to a susceptible person. Using their model, a (presumably transparent) screen between the two people would be perfectly effective, even though in reality it is only useful where ventilation is good enough to prevent infectious material being able to get around the screen.