Using a Powerful Subwoofer as an Ex Vivo Model for Droplet Formation in the Human Larynx

Micron-scale aerosol droplets formed in the larynx and emitted while speaking are a primary vector for transmission of lower respiratory infections, but very little is known about the fluid mechanics underlying the formation of these droplets. A key challenge is the difficulty in directly visualizing the droplet formation within the larynx, so it is desirable to construct an artificial model that mimics how the vocal folds open and close rapidly (at 100 to 200 Hz) with large (millimeter-scale) amplitudes. Here we describe an experimental apparatus featuring a powerful subwoofer, of the type commonly used in ‘monster trucks,’ that provides the requisite operating conditions. High speed video reveals that saliva forms salivary filaments or ‘strings,’ that stretch and pinch off. We quantified their dynamics as a function of frequency and amplitude, and we identified regimes where so called ‘beads-on-a-string’ structures yielded droplet formation. The results provide insight into physiological differences that causes a subset of the human population to emit significantly more aerosols during speech.