Aerosol and droplet generation from performing with woodwind and brass instruments

Abstract

The performing arts have been significantly restricted due to the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. We report measurements of aerosol and droplet concentrations generated when playing woodwind and brass instruments and comparisons with breathing, speaking, and singing. These measurements were conducted in a room with zero number concentration aerosol background in the 0.5-20 μm diameter size range, allowing clear attribution of detected particles to specific activities. A total of 13 instruments were examined across 9 participants. Respirable particle number concentrations and size distributions for playing instruments are consistent with those from the participant when breathing, based on measurements with multiple participants playing the flute and piccolo as well as measurements across the entire cohort. Due to substantial interparticipant variability, we do not provide a comparative assessment of the aerosol generated by playing different instruments, instead considering only the variation in aerosol yield across all instruments studied. Both particle number and mass concentrations from playing instruments are lower than those from speaking and singing at high volume, and no large droplets >20 μm diameter are detected. Combined, these observations suggest that playing instruments generates less aerosol than speaking or singing at high volumes. Moreover, there is no difference between the aerosol concentrations generated by professional and amateur performers while breathing, speaking, or singing, suggesting conclusions for professional singers may also apply to amateurs.

Comparing aerosol number and mass exhalation rates from children and adults during breathing, speaking and singing

“…while breathing and speaking at 70-80 dBA. For both breathing and speaking and across both cohorts, differences in aerosol generation among individuals are lognormally distributed, consistent with previous comparisons of aerosol generation across smaller cohorts [4,31]. Lognormal distribution parameters are provided in electronic supplementary material, table S1.…”

“…We note the same observation for children, with statistically significant differences between speaking at 60-70 dBA and 80-90 dBA ( p < 0.001) and between singing at 60-70 dBA and 80-90 dBA ( p < 0.001) for children. This observation is consistent with previous studies noting a volume dependence for vocalization and instrument playing [1,4,31].…”

“…A plot of particle number exhalation rate against the concurrently recorded sound volume levels for a range of vocalization exercises performed by the same cohort of participants is shown in figure 10b. Data are fit (R 2 = 0.40) by a linear relationship between sound level and number exhalation rate on a logarithmic scale, in line with our previous study which determined the same relationship for instrument playing [31]. Plots of number concentration against both minute ventilation and sound volume are included in the electronic supplementary material, figure S6.…”

“…Previous studies by ourselves and others have quantified the absolute particle number and mass concentrations of aerosols in exhaled air, an intensive property, or the relative particle exhalation rates from different activities through comparing detected particle numbers [1][2][3][4]13,20,[31][32][33]. In this paper, we quantify and compare absolute source-specific respiratory aerosol particle exhalation rates from children and adults during expiratory activities, an extensive property.…”

Measurements and simulations of aerosol released while singing and playing wind instruments

“…20 The data from this study, however, showed that the aerosol concentration measured at the clarinet's bell to be upwards of 2.4 particles/cm particles/cm 3 in the high note range. 39 The plume generated by playing the clarinet was highly directional, had high velocity, and dispersed quickly, which may lead to discrepancies between the studies. Our results that singing produced more aerosol compared to normal speaking levels are in agreement with Alsved et al (2020).…”