Soot is the main contributor to thermal radiation, the dominant heat transfer mode in fire spread. Additionally, soot concentrations strongly affect visibility (and the consequences on evacuation and search and rescue operations) in the case of enclosure fires.

Currently, engineering calculations using Computational Fluid Dynamics (CFD) rely on the so-called ‘fuel conversion model’, which consists of specifying a constant soot yield at the fire source, in conjunction with a soot transport equation. The accuracy of such approach is limited to overventilated fires and far from the flame region. Predictions of local soot concentrations in the reacting region (i.e., the flame) and the subsequent thermal radiation therein remains very challenging.  

The main objective of this project is to improve the state-of-the-art modelling of soot by developing and validating a model which incorporates ‘essential physics and chemistry’, without substantially increasing the computational times for CFD simulations. More specifically, chemistry will be treated using the laminar smoke point (LSP) concept. The interaction of chemistry with turbulence will be addressed in the framework of the Eddy Dissipation Concept (EDC). Although promising results have been previously published in the literature on the potential of such approach (along with a specific treatment of thermal radiation), the quality of the predictions remains to be improved and further scrutinized for a wide range of scenarios (e.g., in terms of fuels and fire size).

The methodology that will be followed in this project is based on a step-wise approach by starting with laminar flames, before moving to turbulent buoyant flames in open atmosphere conditions and then, eventually, under-ventilated conditions.