Overview
Researchers at the Computational Transport Phenomena Laboratory use numerical simulation to answer fundamental questions in the fluid and thermal sciences as they pertain to industrially-based problems. This work covers a range of areas from reacting flows, to turbulent gas dynamics, to nanoparticle-laden flows. Industrial applications include in-cylinder turbulent combustion, plasma processing, chemical vapor deposition as well as nanoparticle-laden flows include pollutant formation & growth, synthesis of nanostructured materials (nanoparticles, nanotubes, fractal-like agglomerates), and the processing and delivery of pharmaceutical powders.Nanoparticle Synthesis
Nanoparticles play an integral role in a wide variety of physical/chemical phenomena and processes. The market for nano-structured materials has dramatically increased in recent years with the growth of applications in industries such as microelectronics, cosmetics, chemical gas sensors, capacitors, heat & mass transfer, and others. These materials are expected to play an increasingly significant role in many major industries as we enter the new millennium. Gas-phase synthesis is a well proven method for the bulk production of nanopowders including silica, titania and carbon black which are produced in quantities in excess of 2 million tons per year. Our interest is in the development of a more thorough understanding of how to control particle formation (and growth) in order to enable the economical production of materials with unique and useful properties.Analysis of Turbulent Flows
Simple flows contain much of the physics and chemistry observed in more complex flows. Through statistical analysis and visualization we may be able to develop mathematical models which can be used in simulating more complex, or realistic flows. One of the greatest challenges we face in visualizing the results of these simulations is posed by the sheer quantity of data produced.Typical large-scale simulations of turbulent reacting flows produce gigabytes of data per time step, and are performed for tens of thousands of time-steps. Analyzing all of the data from a single simulation would take months if not years! Additionally, only a fraction of the data produced at each time step - the location of the flame for example - is of interest; the remainder only provides the global context that supports the evolution of the simulation. Our work has been performed with computer scientists and mathematicians and has been featured in the popular press (Computer Graphics World magazine, November 2001) as well as receiving acclaim from the visualization community.
Short overviews of some of our recent activities are below.
