The effect of chemical composition on the charring of wood across scales

Abstract

Listen

Translate article

Structural softwood (timber) recently gained attention by architects and engineers as a construction material for high-rise buildings. Regulations restrict the height of these buildings due to safety concerns as their fire behaviour is poorly understood. The fire behaviour and loss of loadbearing capacity of timber is controlled by charring, whose chemical kinetics has rarely been studied. Current models of charring assume, without proof, the same reaction scheme and kinetic parameters apply to all wood species, which potentially introduces a large uncertainty. Here, the hypothesis is tested that the kinetics of different wood species insignificantly affects their charring behaviour. The kinetics is modelled by a microscale kinetic model—including pyrolysis and char oxidation reactions—which assumes that the three main components (cellulose, hemicellulose, and lignin) of wood degrade independently. Variation in the kinetics between different wood species is captured by their different chemical compositions within a wood group (softwood or hardwood). Hardwood is included for comparison. A database of over 600 compositions was compiled from literature, and studied across scales using a microscale (mg-samples) and mesoscale (kg-samples) model. All reactions, kinetic parameters, and physical properties were selected from literature. Both models were validated using blind predictions of high-fidelity experiments from literature. Variation in kinetics were found to have a small effect on the predicted mass loss rate at both scales (±1 g/m2-s) and a negligible effect on the predicted temperatures (±16 K) across different depths, heat fluxes, and oxygen concentrations at the mesoscale. These results prove, for the first time, that the variation in kinetics is negligible for predicting charring across scales. A kinetic model of charring derived for one wood species should be valid for all wood species within softwood or hardwood. Modellers should, therefore, focus on the difference in the physical properties instead of the kinetics between wood species.