Research

We develop AI-powered electron microscopy to decode atomic structures of amorphous materials. By integrating machine learning with advanced imaging, we reveal hidden patterns in non-crystalline systems, advancing applications in renewable engineering system.

1. Metallic Glass

We aim to establish a predictive model to explain the atomic configuration and its connection to physical phenomena in metallic glass.

2. Fission and Fusion Nuclear Reactor

We investigate structural degradation mechanisms in reactor materials, particularly those involving crystallinity-to-amorphous phase transitions, under extreme operational conditions—intense radiation fields, corrosive coolant interactions, cyclic thermal stresses, and multiaxial mechanical loads.

3. Next-generation Lithium-ion Battery

We resolve crystalline-amorphous phase coexistence in reactive Li metal and SEI layers during cycling using cryo-TEM, addressing beam-sensitive interfaces. These insights inform robust solid-state battery designs for EVs and grid-storage systems.

4. AI for Electron Microscope

We develop AI-driven automated microscopy to enhance electron microscopy precision for amorphous materials. Machine learning autonomously optimizes experimental parameters, enabling real-time beam control, sample positioning, and low-dose data acquisition.