General Research Interests

This is a research group working on computational molecular and materials sciences. Our current research interests include: quantum computing and Al-based electronic structure methods, computational exploration of mechanisms in physical/chemical processes, and rational design of advanced materials.

• Quantum Computing and AI-Based Electronic Structure Methods

  In electronic structure theory, it is well known that computational cost increases exponentially with system size, a challenge
  often referred to as the exponential wall problem. Quantum computing operates naturally in an exponentially large qubit
  space, making it a promising approach to overcome this limitation. On the other hand, although the Hilbert space of
  the electronic structure problem is exponentially large, the solution to the Schrödinger equation may lie in a relatively small
  subspace if electron correlation is properly understood. Artificial intelligence (AI) offers a powerful tool for capturing
  correlations in high-dimensional spaces. As a result, AI is also expected to play a significant role in advancing modern
  electronic structure calculations.

• Computational Exploration of Mechanisms in Physical/Chemical Processes

  By integrating ab initio modeling with machine learning-accelerated simulations, we investigate the mechanisms of
  physical and chemical processes across multiple scales. Methods such as density functional theory (DFT), neural network
  potentials (NNP), molecular dynamics (MD), and kinetic Monte Carlo (KMC) are routinely employed to study phenomena
  spanning from nucleation and growth to electrochemical reactions. Our ultimate aim is to obtain electronic and atomistic
  insights into the mechanisms that bridge microscopic interactions and macroscopic behavior.

• Rational Design of Advanced Materials

  By elucidating the underlying mechanisms, we design advanced materials for energy and information technologies
  —including solid-state electrolytes, solar cells, photocatalysts, and spintronic devices. Our design strategies incorporate
  defect engineering, passivation strategies, energy level alignment, stress modulation, among other principles.
  Computational methods such as DFT calculations and non-adiabatic molecular dynamics simulations are employed
  to validate these theoretical designs. Furthermore, modern AI techniques, including generative models, can be leveraged
  to propose novel candidate materials, thereby accelerating the entire materials design process.