Effects of initial conditions on Rayleigh-Taylor mixing development

Abstract

This thesis describes an experimental study of the effect of an initial disturbance on the development of Rayleigh-Taylor mixing. The experimental apparatus is a flow channel in which a heavy fluid is placed above a light fluid. Both fluids travel with the same free stream velocity so that no shear is present, thus buoyancy drives the mixing between the two fluids. At the unstable interface between the two fluids, either a single or a binary perturbation was introduced via an oscillating flapper. A single perturbation comprises only one wavelength and amplitude, and a binary perturbation is one that comprises two wavelengths, each with a corresponding amplitude. For the single perturbation experiments, a large wavelength (4 cm) perturbation and a small wavelength (2 cm) perturbation were introduced in separate experiments. The binary perturbation experiments introduced a combination of these two wavelengths in a single experiment. The mixing process was studied using dye in one of the fluids, and photographing the mixture. The photographs were then digitized and ensemble averaged to determine the growth of the mixing layer. In the large wavelength single perturbation experiments the mixing layer grew linearly until the growth was impeded by the flow channel walls. The small wavelength single perturbation experiments grew linearly until approximately 4 to 6 exponential turnover times downstream, where they then began to grow nonlinearly and approach self-similarity. Linear growth is characterized by a saturation growth constant (C [] ) of 0.7, while self-similarity is characterized by a quadratic growth constant [ ]of 0.07. For the binary perturbation experiments, it was found that the small wavelength dominates the growth of the mixing layer. This small wavelength when imposed upon the large wavelength was found to suppress the growth of the large wavelength in the early stages.