Establishing aerosol exposure predictive models based on vibration measurements

This paper establishes particulate exposure predictive models based on vibration measurements under various concrete drilling conditions. The whole study was conducted in an exposure chamber using a full-scale mockup of concrete drilling simulator to simulate six drilling conditions. For each drilling condition, the vibration of the three orthogonal axes (i.e., a(x), a(y), and a(z)) was measured from the hand tool. Particulate exposure concentrations to the total suspended particulate (C(TSP)), PM(10) (C(PM10)), and PM(2.5) (C(PM2.5)) were measured at the downwind side of the drilling simulator. Empirical models for predicting C(TSP), C(PM10) and C(PM2.5) were done based on measured a(x), a(y), and a(z) using the generalized additive model. Good agreement between measured aerosol exposures and vibrations was found with R(2)>0.969. Our results also suggest that a(x) was mainly contributed by the abrasive wear. On the other hand, a(y) and a(z) were mainly contributed by both the impact wear and brittle fracture wear. The approach developed from the present study has the potential to provide a cheaper and convenient method for assessing aerosol exposures from various emission sources, particularly when conducting conventional personal aerosol samplings are not possible in the filed.