Stretched laminar flamelet analysis of turbulents H 2 and CO/H 2/N 2 diffusion flames

Abstract

The stretched laminar flamelet approach to turbulent diffusion flames is tested by comparison of instantaneous and conditionally-averaged Raman measurements in turbulent H 2 and CO/H 2 /N 2 jet diffusion flames with stretched laminar opposed-flow diffusion flame model calculations and measurements. The H 2 turbulent jet diffusion flame measurements close to the fuel nozzle show instantaneous temperatures 300K less than equilibrium, peak OH concentrations three times larger than equilibrium, and substantial coexistence of H 2 and O 2 . Quantitatively similar effects are calculated by David, et al., for lightly stretched (α=100 s −1 ) opposed-flow laminar H 2 diffusion flames. The turbulent flame data lie in a narrow band between the highly stretched α=12,000 s −1 laminar flame calculation and the adiabatic equilibrium curve and approach adiabatic equilibrium downstream in the flame. Localized extinction is relatively unimportant in simple H 2 turbulent jet diffusion flames since only a few instantaneous Raman measurements are close to that calculated at extinction (α=12,000 s −1 ). With a fuel consisting of 40%CO/30%H 2 /30%N 2 , experimental data in both the turbulent jet flame and a laminar opposed flow diffusion flame show even larger nonequilibrium effects. The excellent correspondence between stretched laminar and turbulent flame data suggest that the laminar flamelet approach may provide better predictions than the two-scalar pdf approach of finite-rate chemistry in turbulent reaction zones because partial equilibrium assumptions are not required and localized extinction can be accounted for properly. However, differences between stretched laminar and turbulent flame data suggest that the laminar flamelet approach can overestimate the influence of preferential diffusion in flames containing H 2 and may not properly include the slow three body radical recombination reactions which are likely to occur outside of the relatively thin primary reaction zones.