Combustion Product Toxic Potency Measurements: Comparison of a Small Scale Test and "Real-World" Fires

A long range goal of fire science is to be able to predict "real- world" fire performance from a small set of laboratory scale fire measurements. One material property of primary concern is the toxicity of decomposition prod ucts. Several small scale toxicity protocols that measure toxic potency of the smoke from burning materials have been developed. While several attempts have been made to correlate individual protocols with large scale results, no systematic set of criteria for determining the extent of similarity has been defined and tested. In this paper, three criteria have been proposed and tested using one set of materials (cotton fabric and polyurethane foam) in two dif ferent large-scale enclosures and three different fire scenarios. Tests were con ducted to determine the extent of similarity between the NBS Toxicity Test Protocol and large-scale "real-world" fire performance. The similarity criteria were: LC50 values, time to death of the animals, and toxicant(s) causing the deaths (i.e., the ratios of yields of known toxicants and the contribution of the unknown toxicants). Comparison of test results showed that similar LC50 values were observed. However, toxicant yields varied and animal deaths occurred at different times, implying a difference in physiological syn drome(s).