QUANTUM SIMULATION OF CHEMICAL DYNAMICS IN CONDENSED PHASES

University Of Texas At Austin, Austin TX

Investigators

Abstract

Peter Rossky from the University of Texas at Austin is supported by the Chemical Theory, Models and Computational Methods Program in research focused on understanding, at the level of atoms, the molecular arrangements and time-dependent processes that occur when chemical changes take place in liquid solutions or in unstructured solid materials. These include the changes that occur in biological materials, in semiconductor materials, and in plastics. Although computational modeling is increasingly able to describe these processes, serious challenges remain. For example, it is very important to be able to understand what happens to the energy that is deposited in molecules through the absorption of light. Such knowledge is a critical element for our understanding of the degradation of materials, of the functioning of solar cells, and of biological processes such as photosynthesis. The specific objective of the present studies is the development of new, generally applicable, theoretical and computational models and methods that can fully describe chemistry involving very light atoms and electrons, where quantum mechanics must be invoked. The algorithms being developed can be directly applied as computational tools for the design of advanced materials, pharmaceutical discovery, and nanoscale engineering. They can also impact critical practical issues, including affordable solar energy capture for sustainable energy and a sustainable climate. Outreach activities stemming from this work contribute to the future stability of the scientific workforce, as well as to the public's ability to participate in science policy. The research efforts are developing new simulation algorithms for the quantum mechanical description of nuclear motion that go beyond presently available practical methods in key respects. The novel method is based on the dynamics of Wigner transform variables and is a rather direct generalization of an earlier linearized path integral ("classical Wigner") approach. The new approach retains the ability to directly evaluate the dynamics of non-linear operators (e.g., in inelastic X-ray scattering) and describes vibrational dynamics without unphysical components associated with the simulation algorithm. Further, since the new method directly propagates the phase space points (in contrast to path-integral based approaches), there is good reason to believe that it can be integrated into a non-adiabatic electronic dynamics protocol. Applications to relatively well-characterized molecular systems are being used to test the capability of new methods. The molecular-scale insight needed to guide us to new materials with desired functions or insight into Nature's solutions will follow from the implementation of advanced computational science, coordinated with revealing experimental probes. The resulting capabilities form the basis for what may be the routine modeling methods of the future, just as the basic research of the last 50 years routinely provides for today's experimental researcher. Project efforts enhance education at all levels, including the professional development of graduate and postdoctoral students as researchers and educators with timely skills in the computational solution of physical problems.