From PDMS-based exposure profiling to machine learning-predicted serum concentrations: SVOC exposure disparities...

Occupational exposure to semi-volatile organic compounds (SVOCs) among various population groups has garnered insufficient attention. We investigated the occupational exposures of waste disposal workers and daily exposure of university students to phthalates (PAEs), organophosphate esters (OPEs), and polycyclic aromatic hydrocarbons (PAHs) using polydimethylsiloxane (PDMS)-based passive sampling combined with machine learning-driven serum concentration predictions. The SVOC exposures of students varied depending on their professional activities, e.g., experiments, but dormitories emerge as a significant source. The SVOC exposures among workers varied across different workshops with each acting as the dominant source. The exposure concentrations of PAEs, OPEs, and PAHs among workers were 2.24 times, 6.87 times, and 14.9 times higher than those among students, respectively, whereas the exposure of tris(2,4-di-tert-butylphenyl) phosphate (TDTBPP) among students was 37.8 times higher than that among workers. Sources of PAEs or PAHs for workers and students were relatively similar, while sources of OPEs exhibited greater complexity, especially for TDTBPP. Significant cancer risks were identified for waste disposal workers exposed to di(2-ethylhexyl) phthalate (DEHP), benzo[a]pyrene (BaP), and naphthalene (NAP), and for students subjected to DEHP. Machine learning prediction revealed that despite higher environmental exposures, the predicted serum concentrations of PAEs and OPEs among workers were generally comparable with those of male students but much higher than those of female students, while the predicted serum concentrations of PAHs were comparable across all groups. Risk assessments using Monte Carlo simulations indicated that without protective measures, 99.7% of workers and 55.0% students may face DEHP exposure risk. This emphasized the need for improved ventilation and reduced plasticizer use.