Turbulent Multi-Material Mixing in the Richtmyer-Meshkov Instability

PI Name: 
Sanjiva Lele
PI Email: 
lele@stanford.edu
Institution: 
Stanford University
Allocation Program: 
INCITE
Allocation Hours at ALCF: 
20,000,000
Year: 
2012
Research Domain: 
Engineering

Richtmyer-Meshkov instability (RMI) occurs when a shock wave interacts with a perturbed interface, separating fluids of different densities. After the shock refracts through the interface, perturbations grow; if the incoming shock is strong enough, or if the interface is sufficiently perturbed, the instability evolves into a turbulent mixing region. Even though Richtmyer-Meshkov instability occurs in a wide range of flows (e.g., supernovae explosions, inertial confinement fusion, and hypersonic propulsion systems), the turbulent mixing is not well understood. Few investigations of this phenomenon have been carried out, chiefly because of the unavailability of experimental data to validate numerical results, inadequacy of numerical algorithms, and the unknown (and possibly prohibitive) computational cost. Leveraging the IBM Blue Gene/P’s capabilities, researchers are studying the fundamental physics governing this phenomena, in particular the mechanisms at play in turbulent multi-material mixing in shock accelerated flows. The study employs a novel solution-adaptive numerical framework that scales well and will enable scientific discovery through high-performance computing.

This study enables robust, high-fidelity simulations of the turbulent multi-material mixing generated after RMI. Computing these turbulence statistics, researchers will be able to answer fundamental questions such as: Is the classical Kolmogorov theory for turbulence valid in a transient non-stationary flow? How anisotropic is the turbulence generated in such problems? Does it relax toward isotropy? Is the inertial scaling for the decay rate in the energy-cascade a central element of turbulence modeling even valid in turbulent flows generated by the RMI?

High-resolution simulations of this phenomenon will capture the scales at which viscous dissipation and molecular mixing occur while representing the nonlinear dynamics of the energy-containing scales. The resulting database will enable a fundamental study of the mechanisms at play in turbulent multi-material mixing in shock-accelerated flows and help develop improved models for engineering.