Passive scalar mixing in a planar shear layer with laminar and turbulent inlet conditions

Abstract

The effect of inlet conditions on the mixing of a passive scalar was investigated in a planar shear layer with inlet boundary layers that were laminar, tripped and naturally turbulent-transitional. Planar laser-induced fluorescence measurements of acetone were used to directly evaluate the shear layer structure, and were processed to determine probability density functions ͑PDFs͒ of the mixture fraction. The results agree well with previous studies in aqueous and gaseous systems for laminar inlet conditions. Large-scale structures of a nearly homogeneous composition were found, and the structures spanned the mixing layer width giving rise to a nonmarching style PDF. A high-speed boundary layer that differed from the laminar state ͑produced by tripping a laminar boundary layer, from the naturally turbulent-transitional state at high flow rates, or by tripping the turbulenttransitional condition͒ gave rise to a hybrid-style PDF that was markedly different from either the laminar nonmarching case, or the marching shape that has been found by other investigators. The hybrid PDF shape had a marching character on the high-speed side of the mixing layer, and a distributed nature not favoring any specific composition on the low-speed side of the mixing layer. Tripping the low-speed boundary layer produced no change in the hybrid PDF shape, confirming that the difference observed between the high-and low-speed sides was the result of the shear layer development with turbulent inlet conditions, not an inlet condition effect. In addition, with turbulent and transitional inlet conditions all turbulent passive scalar profiles were found to be self-similar and the velocity power spectra displayed a Ϫ5/3 slope indicating well-developed turbulent conditions prevailed at a relatively low (10 4 ) Reynolds number. Secondary structures were observed in images with turbulent-transitional inlet conditions, but not with tripped inlet conditions.

On streamwise vortices in Large Eddy Simulations of initially laminar plane mixing layers

“…As such, it is impossible to precisely replicate the initial conditions of the experiment, and the inflow conditions reported here should be considered as representative of the conditions that may have existed in the experiment. The fluctuations reported here are similar to those in both magnitude and distribution to those found in other experiments [44], [45], and numerical simulations [46]. It should also be noted that a study of the Details specific to the setup of each simulation are described in turn below.…”

Passive Scalar Transport in a Turbulent Mixing Layer

“…In summary, the scalar transport octants that are compatible with mean gradient transport are the dominant contributors to the concentration fluxes on the high-speed side of the mixing layer, but counter-gradient transport is almost equally as important on the low-speed side. The difference in how mixing of the scalar occurs between the high-speed and low-speed sides of the layer has been referred as 'mixing imbalance' (see for example [21]) and has other consequences, e.g. the asymmetric shape of the concentration rms profile.…”

Initial condition effects on large scale structure in numerical simulations of plane mixing layers

“…Similar profiles are also imposed in the low-speed stream. As noted above, there is insu cient experimental data to precisely recreate the initial disturbance environment; the profiles imposed here are representative of the fluctuation environment found in both experiments 31,37,38 and numerical simulations 39 of free shear flows. In Case WN, these fluctuations are generated using a random number generator, and as such are both spatially-and temporally-uncorrelated.…”