Accuracy of Equilibrium Combustion Models to Predict Mass Fractions of Soot Precursors in Turbulent Jet Flames

Abstract

The primary goal of this thesis is to test a strain-sensitivity parameter’s precision in identifying whether an equilibrium assumption is valid for predicting mass fractions of chemically reactive species in turbulent combustion models. The equilibrium assumption reduces time derivatives of transport functions to zero, greatly reducing the computational time. However, when chemical time scales are longer than transport time scales, the equilibrium assumption is hypothesized to be invalid, and the equilibrium manifold will poorly predict mass fractions of these chemically reactive species. Soot precursors are known to exhibit long chemical time scales and are expected to be poorly predicted by the equilibrium manifold. To identify whether the equilibrium assumption is valid, mass fraction of chemically reactive species predicted by the equilibrium manifold are compared to those predicted by non-equilibrium manifolds. A strain-sensitivity parameter (⇣k), defined as the ratio of the strain rate to the chemical production term1, is hypothesized to describe the ratio of chemical time scales and flow time scales for turbulent flames. Primary heat releasing combustion products along with other small chemically reactive species were observed to be well predicted by the equilibrium manifold and was reflected by ⇣k << 1. However, the strain-sensitivity parameter did not accurately identify chemical species poorly predicted by the equilibrium manifold as having slower chemical time scales because ⇣k relies on instantaneous scalar dissipation rates, and, by definition, flow history affects the production of slow chemically reactive species. An effective scalar dissipation rate was defined where large instantaneous dissipation rates were reflected at future times2 and was observed to better identify slow chemical species. Lastly, a partially non-equilibrium manifold can accurately predict the production of slow chemical species by solving full transport equations when predicting production of slow chemical species but assuming equilibrium when predicting production of fast chemical species.