Multi-Configuration Pair-Density Functional Theory

点击翻译按钮获取中文摘要

With this award, the Chemical Theory, Models and Computational Methods Program in the Chemistry Division is supporting Professors Laura Gagliardi and Donald G. Truhlar of the University of Minnesota to develop theoretical methods and tools to describe the electronic structure of very important classes of molecules that are difficult to study by most widely used, established approaches. In particular, density functional theory (DFT) methods have become the "workhorse" of modern quantum mechanical methods for large and complex molecular systems. The planned research combines the particular expertise of the Gagliardi and Truhlar research gruops to develop a method that draws on aspects of each of their individual research. The techniques developed as part of this research will be applicable to molecular systems that are expected to have broad impacts in, among others, the following application areas: chemical dynamics, thermochemistry, environmental chemistry, including atmospheric chemistry and aqueous environments, electrochemistry, catalysis, materials, supramolecular chemistry, biochemistry, and nanotechnology. Thus the work is poised to have long-term implications for health, materials, environmental, and industrial applications. Undergraduates, graduate students and postdoctoral fellows will be trained in modern computational chemistry methods as an integral part of this research. Software developed in the project is distributed free to the community in portable well-documented code.Under this award, the Gagliardi and Truhlar research groups are developing and implementing a new quantum chemical theory, called multiconfiguration pair-density functional theory (MC-PDFT). The project will explore new ways to combine density functional theory with multiconfiguration wave functions for multireference systems. MC-PDFT maintains the correct spin and spatial symmetry of the systems under consideration and, at the same time, avoids any possibility of double counting the electron correlation energy. The scaling of the computational costs of MC-PDFT is seen to hold promise. The method is employed for the determination of chemical properties such as bond energies and electronic excitation energies and for the generation of potential energy surfaces to be used in photochemical applications.