Quantum Chemistry Methods for Condensed Matter
Predictive computation of correlated condensed matter systems has been a long-standing challenge due to strong electronic interactions and the need to simulate in thermodynamic limit. Our group aims at developing systematically improvable many-body electronic structure methods for describing quantum properties of solid-state materials and molecule-solid interfaces. Our main framework is quantum embedding theory, such as dynamical mean-field theory (DMFT), which enables utilizing molecular quantum chemistry tools for extended systems. In this area, we work on new ab initio quantum embedding formulations and efficient many-body Green's function approaches (e.g., GW, coupled-cluster, selected CI). In addition, we are also interested in exploring machine learning techniques for accelerating expensive many-body calculations.
Materials exhibiting emergent quantum phenomena provide platforms for understanding new physics and realizing more advanced technology. Our group focuses on establishing many-body electronic structure protocols for simulating and exploring point defects in semiconductors and insulators, which are promising for quantum information science applications. We are also interested in investigating unconventional superconductors such as cuprate and nickelate materials, through simulating their phase diagrams and spectroscopy using ab initio computational techniques.
We are interested in investigating molecular-level mechanisms and explore the vast design space of heterogeneous catalysis. Specifically, we focus on developing quantitative description of strongly correlated transition metal catalysts and understanding how to break limitations of simple scaling relationships in electrochemical and single-atom catalyzed reactions, using efficient and reliable machine learning electronic structure approaches.