Density Functional Theory and Long-Range van der Waals Interactions

My primary research interest is to seek fundamental understanding of electronic structures. Based on our understanding, we develop new computational methods to study various aspects of molecules and solids, including ground-state as well as excited-state properties. In electronic structure calculations, the most popular method is Kohn-Sham density functional theory, due to its high computational efficiency and useful accuracy. In this theory, everything is known, except for the exchange-correlation energy component, which has to be approximated as a functional of the electron density. Therefore, the central task of DFT is to develop more reliable exchange-correlation functionals with improved accuracy. In the past thiry years, many density functionals have been proposed and some of them have been widely used in electronic structure calculations.

Density functionals can be divided into two broad categories: Semilocal and nonlocal density functionals. The former is developed using local or semilocal information, such as the electron density, the gradient of the density, and Kohn-Sham orbital kinetic energy density, and thus is computationally more efficient, while the latter is developed by adding nonlocal information, such as exact exchange energy density, and thus is potentially more accurate, because this kind of nonlocal functional can satisfy exact constraints beyond those satisfied by semilocal DFT. The problem is how to build nonlocality into semilocal DFT, without losing the accuracy that semilocal DFT has already achieved.

We are currently developing semilocal DFT using novel methods, aiming to consistently improve the accuracy and applicability of meta-GGA functionals. Then we build nonlocality into our new functionals to construct nonlocal DFT, such as local-hybrid and local-range separation functionals. Finally we incoporate long-range van der Waals interactions into the framework of DFT to study electronic structure of molecules, polymers, clusters, and solids.