This is an investigation of the effects of spatial inhomogeneities in both thermally driven convection and Taylor vortex flow (TVF). The investigation will consider both porous media convection (PMC) and Rayleigh-Benard convection (RBC). The experiments on PMC will be a continued exploration of the effects of media spatial structure on both pattern selection and stability-and the consequent, often large, effects on heat and impurity transport. This work will relate pattern selection to pore structure, for both pure fluids and binary mixtures. Regarding RBC, the experiments will involve horizontal variations in both the layer thickness and in the temperature profile. Related experiments will address similar effects in Taylor vortex flow (TVF) where the radial gap, (delta)R, between the inner and outer cylinder varies. The advantage of this flow is that the first pattern-forming instability (when only the inner cylinder rotates) is to rolls aligned along the cylinder axis-therefore, the complications of directional degeneracy are removed. It is envisioned that experiments on different hydrodynamic systems with control over spatial variations in different ways, will yield broad insight into the effects of spatial variations on such flows. In the past, there has been some work focusing on slow spatial variations of parameters or periodic variation of parameters, but very little on the effects of broad-band random spatial variations, a particularly important issue for PMC. There are a number of important applications for this work including heat transfer devices, insulation, and the flow of contaminants in ground water when there are temperature or concentration gradients. In most practical applications involving heat and mass flows, the systems are far from the usually studied idealizations, and the need for more information on imperfect systems is considerable. Recently developed experimental techniques will allow experimenters to non-invasively image patterns in PMC, leading to im portant new insights into the effects of the medium structure on pattern selection. Particle imaging velocimetry (PIV) will be used to study TVF.

Project Start
Project End
Budget Start
1997-09-15
Budget End
2000-08-31
Support Year
Fiscal Year
1996
Total Cost
$225,000
Indirect Cost
Name
Duke University
Department
Type
DUNS #
City
Durham
State
NC
Country
United States
Zip Code
27705