Current sampling instruments of respirable dust (RD) overestimate the inhaled dose by more than 400%, depending on the size distribution of the airborne dust. This limitation, together with the practices of assigning a single value for RD to any given job, regardless of the level of activity, i.e., respiratory frequencies (RF) and tidal volumes (TV), are incompatible with the advances in occupational epidemiology. A new dust sampler designed to estimate pulmonary deposition (PD) will be developed to alleviate these limitations. The device consists of a 10 mm cyclone followed by a single-nozzle one-stage impactor. The dust fraction of interest is collected by impaction on a 10 mm diameter microscope coverslip. Estimation of PD is obtained by selecting the appropriate air flow rate and diameter of impactor, so that the combined performance would simulate the bellshaped curves of PD at various RF and TV. This configuration is selected, rather than two impactors in series, because cyclone collection efficiency is not associated with a sharp cut-off curve, as compared to an impactor, which results in better matching to PD. The cyclone is also capable of collecting large amounts of dust without overloading. Performance of the new sampler will be evaluated using monodispersed aerosols ranging from 0.5 um to 10 um at air flow rates within the range of battery-operated personal sampling, i.e., 0.5-3.0 1/min. Preliminary results indicate that PD will be estimated very closely by the new sampler. The advantages of the device include close approximation of pulmonary deposition at various work loads, improved estimation of the inhaled dose in epidemiologic studies of pneumoconiosis, and reduced handling steps for chemical analysis, e.g., a silver membrane filter may be used as the collection stage for silica dust, and the filter is directly submitted for x-ray diffraction analysis. The main limitation of the device is underestimation of PD for particles < 0.5 um, but the net effect is negligible because of the very small contribution of these particles (usually < 5%) to the total weight of the collected sample.
