-----------------------------------------------------------
Fluid Dynamics Seminar
Monday, Dec. 5, 2011,
4:00 PM
Cullimore, Room 611
New Jersey Institute of
Technology
Numerical simulations of the Nonlinear Rayleigh-Taylor Instability
Praveen Ramaprabhu
Mechanical Engineering and Engineering Science,
University of North Carolina at Charlotte
Abstract
The Rayleigh–Taylor (RT) instability (after Lord Rayleigh and G. I. Taylor) occurs when a light fluid is separated from a heavy fluid by a sharp interface and an acceleration is directed from the light fluid to the heavy. The classic Rayleigh-Taylor problem consists of a dense fluid on top of a light fluid in the presence of a gravitational acceleration. Understanding Rayleigh-Taylor instability is particularly important to Inertial Confinement Fusion (ICF), a process where nuclear fusion reactions release large quantities of energy by heating and igniting a target fuel. Mixing due to RT and other hydrodynamic instabilities reduce the energy yield from the ICF process. RT instability and mixing also plays a key role in nuclear stockpile management, Type 1a supernovae detonations, atmospheric inversions, extraction of oil, etc. The nonlinear phase of RT evolution has traditionally been described by potential flow-type models. I will describe recent numerical simulations that challenge the validity of well-established potential flow models that have stood for over fifty years. In contrast to the terminal velocity predicted by the potential flow theory, RT bubbles instead experience transients in the form of an acceleration before saturating at a higher Froude number (velocity non-dimensionalized). This transient behavior is explained by the formation of secondary instabilities such as Kelvin-Helmholtz vortex rings that drive the bubbles to acceleration through an induced velocity. Other factors affecting the secondary Kelvin Helmholtz structures such as density contrast and viscosity were studied using numerical simulations and will be presented. These results will significantly impact our understanding of the turbulent flow counterpart in a variety of applications.