Our research focuses on the computational study of electronic structure of molecules in excited, ionized, and electron-attached states, in which static correlation needs to be effectively handled in order to the dynamical electron correlation, in order to address the strong degeneracies in the electronic states. A major part of our work is based on equation-of-motion coupled-cluster (EOM-CC) methods, which provide a reliable and systematically improvable framework for achieving this. We are particularly interested in understanding how these methods perform for complex electronic manifolds and how they can be realized efficiently in practical computations.
Our primary effort lies in computational implementation of advanced electronic structure methods and making them work reliably in practice. This involves translating many-body theory into efficient, stable, and scalable algorithms, carefully managing computational cost and memory requirements, and testing the implementations extensively to ensure robustness. While our current work is mainly focused on computational implementation, it is closely guided by the underlying theoretical structure of the methods and remains open to future methodological developments.
A strong emphasis is placed on scientific software development. The group is actively involved in the implementation, testing, and long-term maintenance of advanced electronic structure methods within the Q-CHEM electronic structure package. Particular attention is given to benchmarking, validation, and smooth integration with existing workflows, so that these methods can be used reliably by the wider computational chemistry community.
The methods and software developed by our group are used for high-accuracy computation of excited-state electronic structure, bonding patterns, spectroscopic parameters and molecular properties and enable numerically reliable analysis of challenging phenomena such as bond-dissociation, conical intersections and intersystem crossings.
