Welcome to the Computational Transport Phenomena Laboratory. Our research in the areas of transport phenomena encompasses physical, analytical, and numerical modeling. We currently focus on two areas: the modeling and simulation of turbulent reacting flows and nanoparticle dynamics (formation and growth) in turbulent flows. In these endeavors our goal is two-fold: (1) To utilize the latest mathematical and numerical tools to investigate and elucidate the underlying physico-chemical processes, and (2) to develop models and numerical algorithms which accurately represent the phenomena in a computationally-affordable manner, facilitating their use in industrial applications.

Methodologies utilized in performing computations of turbulent flows include direct numerical simulation (DNS) and large eddy simulation (LES). DNS involves the solution of the "exact" governing equations on a computational grid which is fine enough to resolve all scales of motion. LES is the solution of the filtered governing equations. In performing LES, the behavior of the large scale field is solved for explicitly while the effects of the small or sub-grid scales (SGS) are modeled. Accurate SGS models facilitate simulations with a greater ranges of physical and chemical parameters. We utilize DNS primarily to elucidate the underlying fluid-scalar/particle interactions in turbulent flows. Additionally, the results of DNS simulations are often used to assess the performance of SGS models used in LES.

Applications of these techniques include high-speed propulsion, gas turbine and internal combustion engines, to name a few. Recent methodologies have proven extremely effective for a variety of turbulent reacting flows, and have enormous potential for performing accurate simulations of turbulent combustion, combustion synthesis, and other vapor-phase syntheses. One such application is soot formation. Soot formation is the result of very complex thermo-fluid-chemical interactions. The inherent unsteadiness and limited spatial resolution has rendered the measurement of soot properties in turbulent flames impossible. However, knowledge of the spatio-temporal evolution of the chemical species - obtained via simulatin - can help to reveal the underlying structure of such complex phenomena.

A large number of our simulations are performed at the Minnesota Supercomputing Institute (MSI). The MSI provides supercomputing resources and user support to University of Minnesota researchers. This includes all aspects of high-performance computing and scientific modeling and simulation as well as graphics, visualization, and high-performance network communications. Additionally, the Computational Transport Phenomena Laboratory has a number of Apple PowerMacintosh & PowerBooks, as well as Linux and Silicon Graphics workstations for local analysis and visualization. Recently added to our computational ensemble is an Apple Xserve G5 cluster, which is described in detail here.

The Computational Transport Phenomena Laboratory is an environment populated by Undergraduate, Masters, PhD students as well as Post-doctoral scientists. The Lab is continually looking for motivated undergraduate and graduate students who are interested in performing research at a high-level. If interested, please make all inquiries to Professor Garrick.