We develop and apply a variety of computational methods, including: atomistic simulation, density functional theory and structure prediction to investigate the materials’ structure-property relation.

Our goal is to discover and design new materials in the following aspects:

Materials Informatics

Materials Informatics

Big data and machine learning in materials sciences.

Learn more

organic crystal polymorphism

Organic Crystal Polymorphism

Predicting the crystal structures for a given molecule could help determine the existence of different forms and suggest as yet unseen polymorphs of currently known structures.

Learn more

defects in materials

Defects in Materials

Understanding the structural and functional properties influenced by defects is key to optimizing next-gen materials needed for advanced energy applications.

Learn more

Topological Phonon database

Topological phonon database!

The first online topological phonon database with over 5000 materials is in live!

Learn more

Machine learning Force Field

Latest paper (2022/10/28)

Simulation of GaN’s phase transition using the machine learning metadynamics

Learn more

Crystal packing

Latest paper (2022/10/27)

Quantification of Crystal Packing Similarity from Spherical Harmonic Transform

Learn more