韩国延世大学Aloysius Soon教授学术讲座

Intelligent Surface Science: Revolutionizing Nanostructure Discovery with AI-guided First-Principles Thermodynamics

主讲人 : Prof. Aloysius Soon

Recent advances in experimental and theoretical methods have greatly enhanced our ability to characterize the physical, chemical, and crystallographic properties of well-defined surfaces and interfaces. In particular, the study of low-dimensional nanomaterials on  l substrates has attracted considerable interest, driven by progress in surface spectroscopy and microscopy. Computational simulations have played a complementary role, providing critical insights through data-driven and theory-guided atomistic modeling. However, reconciling experimental and theoretical de ions of supported nanostructures remains a major challenge. This keynote discusses how modern computational approaches and artificial intelligence (AI) are helping to bridge this gap in theoretical surface science. We revisit the application of ab initio atomistic thermodynamics–commonly used in computational catalysis–to predict stable catalyst surfaces under realistic technical conditions. By incorporating AI-driven global optimization, this method addresses the longstanding pressure and temperature divide between ultra-high vacuum experiments and industrial reactor environments. Using a  stable O/Cu surface oxide as a case study, we illustrate how AI-enhanced first-principles simulations reveal novel nanostructures, thereby broadening thematerials search space and reducing dependence on experimental intuition. Looking forward, we highlight the potential of integrating these approaches and extending thermodynamic analysis to aqueous electrochemical systems. This integrated strategy paves the way for intelligent, data-driven surface structure determination and the discovery of industrially relevant interfacial configurations.

主讲人简介：

Aloysius Soon教授是英国物理学会会士（FInstP）、英国皇家化学会会士（FRSC）及英国特许科学家（CSci）。他先后于新加坡国立大学、奥克兰大学及悉尼大学获化学学士、硕士及物理学博士学位，曾任德国马克斯·普朗克学会弗里茨·哈伯研究所洪堡研究员。Soon教授在计算材料科学领域具有广泛影响力，长期致力于发展第一性原理计算与机器学习方法，深刻揭示材料表/界面的物理化学机制。他在二维材料、功能氧化物、催化与能源材料等前沿方向做出了系统性贡献，在Nature Materials、Nature Catalysis、Advanced Materials等顶级期刊发表论文140余篇，引用超6500次，H指数39。他的研究在表面科学、缺陷热力学及机器学习辅助材料发现等方面具有突出影响，并与韩国三星、LG、现代汽车等工业界及国际顶尖科研机构保持广泛合作。