Experimental characterization of respiratory droplet emission

The droplet-laden air cloud exhaled by humans during different respiratory activities plays a major role in infectious disease transmission. That exhaled droplets contain pathogen is a well-known fact in the scientific community since the 19th Century. Unfortunately, pandemics as COVID-19, SARS, and MERS, have recently brought back the attention to this issue, which is rather complex since multiple-scale phenomena and different disciplines (epidemiology, biology, fluid mechanics) are involved. Fluid mechanics plays a major role in the comprehension of droplet-laden air cloud dynamics and mitigation of the related risks. Indeed, the pathogens interact with fluids from their encapsulation within the droplets in the airways to their inhalation by susceptible individuals. The prediction of the fate of the droplets after their emission have widely been improved, especially in the past three years, by means of experiments and models. However, a lack of knowledge of the air and droplet properties at the emission (mouth) emerges from the literature. Providing precise information on emission characteristics to numerical or theoretical models that predict droplet dispersion is of striking importance to obtain reliable results. The present thesis aims to contribute to this field by improving the characterization of droplet emission, namely, their size and velocity distribution. A series of laboratory experiments have been conducted considering different respiratory activities, namely, speaking, coughing and breathing. The Interferometric Laser Imaging for Droplet Sizing (ILIDS) technique has been used for data collection. Both the setup and the related data processing have been improved with respect to ILIDS standard applications in order to detect droplets with size down to 2 μm and to measure all their three velocity components. Two experimental campaigns involving twenty-three volunteers have been carried out. The effects of protection masks and the variability in the results obtained for the same volunteer repeating the tests are also assessed. Finally, droplet size and velocity distributions have been used as input data for Computational Fluid Dynamics simulations in order to analyse their role in the dispersion process following their emission.