Phys. Rev. Fluids 5, 102301 (2020)

We now know that talking in a confined space likely aids the spread of SARS-CoV-2, the virus leading to COVID-19. But the mechanism by which salivary droplets form during speech is far from clear. Case in point: we know that consonants create more droplets than vowels, but we don’t know how. Now, a study by Manouk Abkarian and Howard Stone implicates saliva viscoelasticity and airflow as key factors in the process — and hints at how to stop it.

The pair imaged a speaker pronouncing a series of sounds and found that the action of pressing one’s lips together to form a consonant creates a microscopic lubrication layer. This layer stretches into a thin vertical film, which becomes unstable and breaks up into filaments. The fast airflow that follows — in a ‘pa’ or a ‘ba’ — causes some of the filaments to break and whip outwards towards the unwitting listener.

In good news, the authors found that applying lip balm could impede the mechanism by breaking the filaments before they stretch, and reducing the cumulative number of droplets produced.