Theoretical Chemistry Methodologies to Generate Thermodynamic Properties for Chemical Process Simulation and Pollutant Behavior Prediction (TSE99-G)

California Institute Of Technology, Pasadena CA

Investigators

Abstract

William Goddard, Mario Blanco, and John Seinfeld of the California Institute of Technology are supported by the Division of Chemistry, the Office of Multidisciplinary Activities, and the Division of Chemical and Transport Systems to develop theoretical methodologies which generate thermodynamic properties for use in chemical process simulation and pollutant behavior prediction. This grant is made under the NSF/EPA Partnership for Environmental Research (Technology for a Sustainable Environment). This research will integrate quantum mechanics, molecular dynamics, and statistical mechanics into the next generation of chemical process simulation and design technology. Needed thermodynamic data, such as excess free energies, activity coefficients, and phase diagrams, will be accurately provided by first-principles molecular simulations of aqueous mixtures of organic solvents. Efficiency in chemical manufacturing is greatly dependent on the quality of the thermodynamic data employed during the chemical process design phase. Due to the enormous number of possible binary and multicomponent mixtures present in a chemical process, such data are often unavailable. Recent advances in atomistic simulations, combined with statistical models for predicting thermodynamic properties of mixtures, will lead to a new approach of achieving pollution prevention through more optimal design of chemical processes.