Human occupants themselves constitute an important source of volatile organic compounds (VOCs) in indoor environments through breath and dermal emissions. In order to quantify VOC emissions from occupants under real-world settings, previous indoor observational studies often determined emission factors (i.e., average emission rates per person). However, the values obtained across these studies exhibited large variability, and the causes of this variability still need to be understood. Herein we report 10-day real-time VOC measurements in a university student office, using a proton transfer reaction-quadrupole interface-time-of-flight mass spectrometer. A method was developed to identify VOCs of primary human origin and to quantify the corresponding emission factors, accounting for the dynamically changing occupancy level and ventilation rate in the assessed office. We found that the emission factors of many dermally emitted VOCs strongly increased as the ozone concentration increased from <3 to 10-15 ppb. These VOCs include geranyl acetone, 6-methyl-5-hepten-2-one (6-MHO), and C-C saturated aldehydes, which align with characteristic first-generation ozonolysis products of skin oil. The strongest increase occurred for 6-MHO, from 113 to 337 μg/h/p. In comparison, acetone and isoprene, which are primarily emitted from human breath, varied little with the ozone level. In light of this finding, we conducted an integrated analysis of emission factors reported in the literature for two frequently reported species, namely, 6-MHO and decanal. Ozone concentration alone can explain 94-97% of the variation in their emission factors across previous studies, and the best-estimated ozone dependence obtained using the literature data is consistent with those obtained in the current study. These results suggest that the ozone concentration is a key factor regulating emission factors of many dermally emitted VOCs in real indoor environments, which has to be considered when reporting or using the emission factors.
