Comparisons among various theories for turbulent, reacting, and planar mixing layers

Ignition in the planar, turbulent, reacting mixing layer was analyzed via several model approaches and comparisons were made. Both Favre and time-averaged forms of the equations were employed with various forms of the mixing length theory; k, epsilon theory; and k, omega theory. A coordinate transform was employed to remove the singularity at the mixing layer edge where the turbulent diffusivity goes to zero. An iterative quasi-linearized implicit finite-difference scheme was used to solve the governing set of parabolic partial differential equations. The nonreacting limit with temperature and density gradients was fully examined. The major differences between the models and types of averaging were in the predicted spreading parameter. The comparison between theory and data was better for velocity profiles than for density profiles. Additional nonreacting variable density data are needed, particularly for high Reynolds numbers. For reacting flows, two different approaches to the calculation of the ignition length were analyzed. Comparisons among the models indicated significant sensitivity to the form of averaging and the choice of kinetics with a lesser sensitivity to the choice of turbulence modeling.