Effects of metallic coating on the combustion toxicity of engineering plastics

Experiments were conducted on 34 plastic materials having a variety of metallic coatings to determine the toxicity of their thermal decomposition products. Mice were exposed for 30 min in a dome exposure chamber to the products obtained by ramp-heating the samples from 200°C to 800°C. An LC₅₀ value was obtained for each material. Postmortem examinations were conducted on all dead mice, and on survivors after 14 days, to determine the gross pathological effects of exposure; particular attention was devoted to pulmonary pathology. The exposure protocol chosen has been extensively criticized, but it is very useful to study the effects of stress on mice, which was the most important part of this work. Experiments were made involving unrestrained mice in groups of four, restrained mice in groups of four and unrestrained single mice. The LC₅₀ values for single unrestrained mice were greater, by factors of 2–3, than those for four restrained mice, with the differences being shown to be statistically significant. This suggests that stress on the test animals will tend to reduce the LC₅₀ values in bench-scale smoke toxicity tests. The LC₅₀ values for all of the materials tested were equal to or higher than the value of 8 mg1^?1 representative of the contribution of carbon monoxide to post-flashover fires. Moreover, no ‘supertoxicants’ were found in the smoke of any of the materials tested. Finally, the coatings did not adversely affect the smoke toxicity of the substrate materials by a factor higher than 2–3 in any of the cases investigated. Uncoated polyethylene was the most toxic substrate material tested (LC₅₀ = 16 mgl^?1) and uncoated NORYL® resin was the least toxic (LC₅₀ = 91mgl^?1). Metallic coatings involving Cu, Ni, graphite, and Zn typically had no statistically significant effect on the smoke toxicity of the substrate materials, although Ni coatings increased the smoke toxicity of ABS I and of white polycarbonate structural foam, by factors of 2–3. Overall smoke toxicities were well correlated with production of carbon monoxide (r=0.84) and carbon dioxide (r=0.82); oxygen levels and chamber temperature did not vary beyond acceptable limits. The materials tested generating the more toxic smokes (including polyethlene, polystyrene, and several polycarbonates) produced severe lung damage at low concentrations. The LC₅₀ of these materials was also typically greater than predicted on the basis of CO production. Other materials (including several coating on NORYL® resin and Lexan® polycarbonate) produced pulmonary damage at higher concentrations amd had LC₅₀ values more closely correlated with CO production. None of the polyurethane materials tested produced severe lung damage at the concentrations employed.