On the use of surrogate gases in fire toxicity calculations

Abstract

Engineering analyses of fire toxicity are often limited to the production, transport, and incapacitating effects of carbon monoxide and hydrogen cyanide, despite the availability of a broader range of gases in the Purser's formulation for the fractional effective dose (FED). In this work, we show how the effect of a wide set of gases can be assigned to two alternative surrogate gases, CO or CO₂, combined with an effective yield of HCN. We validated the methods by measuring the yields of asphyxiant and irritant gases for five fuels, and by carrying out CFD predictions of FED using full and reduced sets of gas species. Results indicate that ignoring the full list of gases can lead to a severe underestimation of FED, with errors ranging between −12 % and −77 %. Two surrogate methods provided much more accurate predictions, with relative uncertainties between −2.2 % and +11 % for the CO + HCN -method, and −6.6 % and +1.5 % for the CO₂+HCN -method. In comparison to a simulation using the full set of toxic gases, the proposed surrogate methods provide significantly simplified user input and computational savings.