The PI will undertake a computational study of scalar transport (e.g., contaminants or thermal energy) in wall turbulence and the influence of Prandtl (or Schmidt) number on near-wall and dispersion behavior. Central to this study will be the development of Direct Numerical Simulation (DNS) in conjunction with Langrangian Scalar Tracking (LST) for the generation of numerical data and the development of descriptive models. The program will incorporate advances in High Performance Computing to allow the detailed investigation of heat and mass transfer in turbulent flow for a variety of fluid systems. Central concerns are the prediction of the spatial variation of turbulent transport properties, the effect of molecular Prandtl number on turbulent transport and the effect of boundary conditions on the scalar property transport. The innovations of the present work are (a) the use of hydrodynamics generated by a Direct Numerical Simulation (DNS) to generate Lagrangian data for turbulent dispersion; (b) the development of plane Couette flow simulations for the study of transport in the constant stress region; and (c) the study of chemical reaction in turbulence in the Lagrangian framework. The results, if successful, will have impact on a variety of practical heat and mass transfer problems (heat exchange, mixing, reactor flows, turbine blade cooling, pollutant dispersion in the atmosphere). The work will also foster the application of HPC technology in ways that will make DNS more assessable and cost effective, and potentially will lead to the ability to study scalar transport in other important physical problems, such as transport in very small-scale flows. Specific plans have been laid out to introduce HPC to undergraduate students who will participate in the research at different stages of the work and through the introduction of HPC into the curriculum of Chemical Engineering at the University of Oklahoma.
