Beryllium aerosol characteristics in the magnesium and aluminum transformation industry in Quebec: a comparison of four different sampling methodologies.

To examine the influence of the sampling method on beryllium (Be) exposure assessment, a study was conducted in foundries and smelters to contrast the performance of five different dust sampling devices. Six sampling surveys were conducted in four different settings, and both personal and fixed station samples were collected using the following sampling heads: IOM samplers (inhalable dust), 35-mm plastic cassettes (total dust), aluminum SKC cyclones (respirable dust), 8-stage Sierra cascade impactors, and 12-stage MOUDI impactors. In total, beryllium concentrations were determined for 66/68 inhalable dust samples, 62/62 total dust samples, 56/57 respirable dust samples, 54/64 8-stage Sierra samples, and 19/25 12-stage MOUDI samples. In the magnesium foundry and aluminum smelters, the concentrations obtained during specific tasks could exceed the actual permissible exposure limit of the province of Quebec (0.15 microg/m(3)) or of the ACGIH threshold limit value (TLV) (0.05 microg/m(3)). The median of median dust concentration ratios computed from the sampling heads at the fixed station decreased as follows: IOM (1.00) > Sierra (0.76) > 37-mm cassette (0.61) > MOUDI (0.48) > respirable (0.12). The same trends were observed with the ratios of the median of median Be concentrations at the fixed station but with a larger scattering within sampling heads as follows: IOM (1.00) > Sierra (0.69) > 37-mm cassette (0.64) > MOUDI (0.54) > respirable (0.19). The median of median ratios of dust (IOM (1.00) > Sierra (0.56) > 37-mm cassette (0.35) > respirable (0.06)) and Be (IOM (1.00) > Sierra (0.66) > 37-mm cassette (0.48) > respirable (0.11)) in dust were lower, and there was less scattering for the 37-mm cassette and SKC cyclone used during breathing zone sampling than for the same sampling heads at the fixed station. Inhalable aerosol measurements should remain the tool for estimating the risk of exposure to beryllium in these settings until a clear dose response is established for these sampling heads.