Development and characterization of samplers whose performances mimic the inhalability of the human head will be an important step forward in exposure assessment for environmental and occupational aerosols. The current method for sampler characterization involves arduous experiments in large wind tunnels, of which only a few exist. Therefore a new, rapid and cost-effective method is needed, and the main objective of this research was to develop a set of scaling laws for aerosol sampling, leading to new methodology for aerosol sampler characterization in a small-scale wind tunnel using a direct-reading aerodynamic particle sizer (APS), which counts and sizes sampled particles. For this purpose, a prototype automated experimental system was designed and built, which included a novel approach to account for particle losses inside the sampler inlet. For this, the sampler entry was filled with a plug of porous plastic foam that smoothed the aspirated air flow and provided well-defined particle penetration into the instrument. The first experiments with this system investigated the aspiration efficiency of a thinwalled probe at 90° to the wind, with the aim of gaining experience with the new apparatus and methodology with an aerosol sampling system that is simple yet still of considerable scientific interest to aerosol scientists. The results were in fair agreement with previously published data, but indicated some internal functional relationships not previously seen. For the main experimental study aimed at practical personal aerosol samplers, the scaling laws we developed enabled full-scale experimental conditions may be simulated by small-scale experiments. A key component was that not only are the sampler dimensions and air flows able to be scaled, but the particle size itself may be scaled as well. This allowed for particles in the range of the APS (up to 20 um) to be used to simulate the behavior of much larger particles. These scaling laws were applied to experiments for testing the performance of three commercially-available personal aerosol samplers: the 10M personal inhalable aerosol sampler, the CIS inhalable aerosol sampler and the Button aerosol sampler. The results from these experiments provided validation of the scaling laws and were broadly consistent with what has been previously observed in large wind tunnels. Again, however, some additional functional relationships were observed that had not previously been seen by others. Despite some differences in sampler performance when compared to full-scale studies, it is concluded that the new testing method developed in this research will be an excellent starting point for the preparation of a standard protocol for the evaluation of practical aerosol samplers. Finally, some experiments were carried out in the field, an aluminum smelter, to compare instruments derived from the new knowledge gained in this research, as well as in a previous NIOSH-funded project. Here, for instruments that compared well in our laboratory studies, agreement was less satisfactory in the field. It is possible that this may be associated with the fact that windspeeds in most workplaces are actually much lower than those in wind tunnel experiments (including not only our own but also most of the aerosol sampling research conducted by others).
Department of Environmental Health Sciences, School of Public Health, University of Michigan, 109 S. Observatory, Ann Arbor, MI 48109-2029