Dongguk University · Chemical and Biochemical Engineering
Learn atomic interactions. Explore structure. Predict function.
We combine first-principles theory, machine-learning potentials and data-driven models to discover atomic structures, predict materials behavior and guide the search for new functionality.
Directed by Prof. Yonghyuk Lee
View Google Scholar for current metrics.
About the lab
Materials function emerges from atomic structure and its evolution under operating conditions. At CCML, we learn interatomic interactions, guide structure searches and predict properties in close connection with first-principles theory.
Our research spans dynamic surfaces and interfaces, catalytic activity and stability, and electronic, spin, ionic and optical responses. We seek principles that connect atomic-scale mechanisms to the design of functional materials.
Explore our four research directions →Research at a glance
Four connected directions linking atomic structure, reactivity and materials function through computation and learning.
Learn interatomic potentials and explore atomic structures with active learning.
머신러닝 원자 시뮬레이션
Active learning for atomistic discovery.
Explore surface reconstructions and water chemistry at solid–liquid interfaces.
동적 표면과 계면
ML-guided exploration of working interfaces.
Connect reaction mechanisms and data-driven models to understand catalyst activity and degradation.
촉매 활성과 안정성
Data-driven prediction of catalyst stability.
Connect atomic structure and surface oxidation to optical response and hydrogen evolution.
기능성 소재 설계
Learning structure–property relationships.
Our work
ACS Catalysis, 2026, 16, 10398–10410 [IF=13.1]
A hybrid data-driven model links atomistic calculations and experimental observations to Rh nanoparticle growth.
Angewandte Chemie International Edition, 2025, 64, e202501017 [IF=16.1]
Machine-learning-assisted structure searches explore the reconstructed surfaces of reduced rutile TiO₂.
Journal of the American Chemical Society, 2025, 147, 7799–7808 [IF=14.5]
Machine-learning surface searches are combined with simulations of reconstructed BiVO₄ in contact with water.
Energy & Environmental Science, 2025, 18, 1214–1231 [IF=32.4]
An atomistic model connects oxygen evolution and Ir dissolution in amorphous hydrous iridium oxides.
Join CCML
Interested in graduate study? Prior coding or computational chemistry experience is not required.
