Influence of initial conditions on turbulence and mixing in Richtmyer-Meshkov flows in presence of re-shock

The study of influence of initial conditions [amplitude ((delta)) and wavelength (λ) of perturbations] on variable-density flows stems from the the recent work done by Dimonte et at. 2004, Miles et al. 2005 and Balakumar et al. 2008a, where it was shown that both Richtmyer-Meshkov (R-M) and Rayleigh-Taylor (R-T) turbulent flows are not truly self similar and have a strong initial conditions dependence on turbulence transport and mixing. However, so far most of the work on this topic has been numerical studies which suggest that for multi-mode systems, the emergence of a regime of self-similar instability growth independent of the initial conditions does not occur. Experimental evidence is needed to verify this theory. Thus, the present work focuses on conducting an experimental study at moderate Mach number (Ma = 1.2) to understand the effects of multi-mode perturbations of the shocked interface on instability growth rate and mixing for R-M flows, which are important mechanisms in inertial confinement fusion reactions, supernovae, combustion and general fluid mixing processes. The ongoing 3-D numerical simulations using ILES will be used for validation of our experimental results. The experiments to study R-M turbulence and mixing are carreid out at the Los Alamos Gas Shock Tube facility shown in Figure I and described in detail in Balakumar et al. 2008b. A heavy gas curtain of SF6, surrounded on both sides by ambient air, representing a light/heavy/light interface is flowed through a varicose nozzle (shown in Figure 1c). This initial interface is then accelerated by a Mach 1.2 shock, generated in the driver section. Simultaneous Particle Image Velocimetry (PIV) and Planar Laser Induced Fluorescence (PLIF) diagnostics are used to characterize the initial conditions and also image the evolving flow to measure instantaneous velocity and density fields. The evolving structures are re-shocked at various times using a moveable reflecting wall to study the initial condition effects on turbulence and mixing. Mean flow fields are averaged from an ensemble of experiments whose initial density fields correlate to within 97% of each other. From the mean field, the fluctuating quantities are determined, and the density self-correlations and density-velocity correlations are calculated.