Experimental study of shock-driven, variable-density turbulence using a complex interface

Abstract

The overarching goal of this work is to advance the current knowledge of hydrodynamic instabilities (namely, Richtmyer-Meshkov and Kelvin-Helmholtz instabilities) and associated turbulent mixing phenomena which is important for several emerging technologies and verification/validation of numerical models being developed to study these phenomena. Three experimental campaigns were designed to focus on understanding the evolution of the instability under different impulsive acceleration histories and highlight the impact of initial conditions on the developing turbulent flow environment. The first campaign highlights the importance of initial baroclinic torque distribution along the developing shocked interface in a twice-shocked variable-density flow environment. The second campaign is a parametric study which aims at providing a large dataset for validating models in literature as well as simulations. In the last study, a new type of initial condition was designed to study the effect of initial conditions on late time turbulent flows. A description of the optical diagnostic techniques developed in our laboratory in order to complete these studies will be given. Now each campaign will be introduced.

In the first campaign, an inclined interface perturbation is used as the initial condition. The Mach number (1.55), angle of inclination (60 degrees), and gas pair (N2/CO2) were held constant. The parameter which changed was the distance that the initial condition was placed relative to the end of the shock tube (i.e., the end of the test section). Three distances were used. The vorticity distribution was found to be very different for the most developed case after reshock. Furthermore, the most developed case started to develop an inertial range before reshock.

The second campaign is parametric and seeks to test a proposed inclined interface scaling technique. The data is also useful for comparing to Ares simulation results. The parameter space covered Mach number (1.55 and 2.01), inclination angle (60 degrees and 80 degrees), and Atwood number (0.23 and 0.67). PLIF was developed and used to collect data for four cases before and after reshock. Linear and nonlinear cases developed very differently before reshock, but their mixing widths converged after reshock.

The last campaign involves a new perturbation technique which generates what will be referred to as a complex interface. Counter-flowing jets were placed near the interface exit ports to create shear. The perturbation was made more complex by also injecting light (heavy) gas into the heavy (light) one. Density and velocity statistics were collected simultaneously. The complex case retained a signature of the inclined interface perturbation at late time before reshock and developed a larger inertial range than its inclined interface counterpart. Important parameters for a variable-density turbulence model are also presented.