Chlorosilane acute inhalation toxicity and development of an LC50 prediction model.

The acute inhalation toxicity of 10 chlorosilanes was investigated in Fischer 344 rats using a 1-h whole-body vapor inhalation exposure and a 14-day recovery period. The median lethal concentration (LC50(1)) for each material was calculated from the nominal exposure concentrations and mortality. Experimentally derived LC50(1) values for monochlorosilanes (4257-4478 ppm) were greater than those for dichlorosilanes (1785-2092 ppm), which were greater than those for trichlorosilanes (1257-1611 ppm). Apparent was a strong structure-activity relationship (r2 = .97) between chlorine content and LC50(1) value. Estimated LC50(1) values for mono-, di-, and trichlorosilanes were determined to be 3262, 1639, and 1066 ppm, respectively, utilizing this relationship and the lower limit of the 95% prediction interval. The LC50(1) values determined in this series of studies were greater than that reported for hydrogen chloride (3124 ppm), when expressed on a chlorine equivalence basis (3570-5248 ppm), demonstrating that the acute toxicity of these chlorosilanes is similar to or less than that for hydrogen chloride. The good correlation between chlorine content and LC50(1) provides a sound basis for estimation of LC50(1) for chlorosilanes not already evaluated. The use of structure-activity relationships is consistent with the chemical industry and federal agency initiatives to reduce, refine, and/or replace the use of animals in testing without compromising the quality of health and safety assessments.