Education and Background

B. S. '82, Peking Univ.; Ph.D. '86, Postdoctoral Fellow '86-'87, Univ. of North Carolina-Chapel Hill; Postdoctoral Fellow '88-'89, Univ. of California-Berkeley; Asst. Prof. '89 -'95, Duke Univ.; Assoc. Prof. '95-'99, Prof. '99-, Duke University; Visiting Associate Professor, '96-'97, Hong Kong Univ. of Science & Technology; Fellow of the NC Supercomputing Center Research Institute '92-; Alfred P. Sloan Research Fellow '93-'97, 1997 Annual Medal of the International Academy of Quantum Molecular Science.
 

Research Interests

Theoretical Chemistry

Our interests are in the application of quantum mechanical methods to understanding and determining molecular structure and properties for chemical and biological systems.

We particularly focus on density-functional theory, which is a very promising alternative approach to electronic structure problems. Conventional quantum chemistry describes electrons in molecules in terms of the wavefunction and is successful for small molecules only. The difficulty associated with large molecules originates from the use of molecular orbitals as building blocks. With an increasing number of electrons, the computational requirement soon becomes prohibitive. Instead of wavefunctions, density functional theory uses electron density to describe all the electrons. Electron density is a probability distribution function in three-dimensional space; much simpler than the wavefunction.

We have constructed the divide-and conquer theory for electronic structure calculations of large molecules and established the validity of the theory through various molecular computations. The approach solves the long-standing problem of using the electron density or the density matrix as the only basic computational variable, instead of the conventional molecular orbitals. It divides a molecule into subsystems and determines the electron density of each subsystem separately. The computational effort of the theory scales linearly, in contrast to the N3 scaling in the conventional methods. The method has been applied to the studies of the electronic structure, the thermodynamic properties, and the effects of solvent polarization in biological macromolecules at the semiempirical level. We are extending and applying our linear-scaling semiempirical methods to study mechanisms of enzyme reaction systems.

We plan to develop methods for efficient and accurate determination of electronic structure of large molecules based on density functional theory. Specifically, we plan: to make the calculations very efficient by the development of optimal linear scaling approaches based on the newly developed absolute minimum principles with nonorthogonal localized orbitals for the diagonalization problem and the recursive bisection method for the classical Coulomb interaction of electrons. We plan to enhance the accuracy of DFT by the construction of new and improved density functionals for electron exchange and correlation based on the adiabatic connection, and the formulation of functionals from wave function approach with localized orbitals.

Related Links

For more information, please click here, or here.