演講公告

時間：3/16，10：00-12：00

地點：材料系E1-102會議室

講員：日本九州大學飯久保智(Satoshi Iikubo)教授

Lecture Title

Computational Approaches to Materials Development

Abstract

Computational science has become an essential component of modern materials development, providing powerful tools to understand, predict, and design materials properties beyond experimental limitations. Among various approaches, the CALPHAD method has long served as a practical framework for describing phase equilibria and thermodynamic properties by integrating experimental and theoretical information.In parallel, first-principles calculations based on electronic structure theory have advanced significantly and now play a central role in providing physically grounded thermodynamic data, such as formation energies and temperature-dependent properties through lattice vibration and cluster expansion methods. More recently, data-driven and machine-learning approaches have emerged as complementary tools for efficiently exploring complex compositional and structural spaces. In this lecture, I introduce a broad overview of computational approaches for materials development, focusing on how first-principles calculations, CALPHAD modeling, and machine-learning techniques can be combined in a unified manner. Recent research examples on thermodynamic property calculations for binary and ternary alloy systems, as well as their integration into thermodynamic databases, will be presented to illustrate the synergy among these computational methods.

Keywords: modern materials development; CALPHAD method; phase equilibria; thermodynamic properties; machine-learning techniques;