The course provides a comprehensive introduction to the theoretical and computational methods used to study and predict the properties of materials at the atomic and electronic scale. It aims to provide students with a solid background in fundamental physical principles, numerical algorithms, and their practical application through modern high-performance computing (HPC) tools.
First, the role of computational science as the third pillar of materials research—alongside theory and experiment—and its necessity in the study of complex systems are presented. The basic categories of methods are analyzed, from classical simulations to quantum mechanical approaches, as well as the capabilities and limitations of modern computing systems.
Most of the course focuses on Molecular Dynamics, examining the description of atomic interactions, the solution of equations of motion, the initialization and equilibration of systems, as well as techniques for the production and analysis of computational data for thermodynamic, mechanical, and structural properties. At the same time, first-principles methods and density functional theory are analyzed, providing a theoretical background for electronic structure, energy states, exchange-correlation functions, and numerical techniques for solving quantum mechanical equations.
Particular emphasis is placed on exploring the potential of Artificial Intelligence and Machine Learning in materials physics and engineering. Contemporary applications are presented, such as the development of machine-learning interatomic potentials, the acceleration of DFT calculations, the prediction of material properties, as well as tools for the automation of high-throughput screening processes. Students are introduced to the concepts of extracted features, neural networks, active learning, and materials databases, understanding how AI is combined with classical computational techniques for the discovery of new materials.
The laboratory part includes applications with professional simulation software. Students perform molecular dynamics simulations to study thermal and mechanical properties, while applying electronic structure methods to optimize geometry and calculate energy states. Through practical exercises, they develop skills in software handling, data analysis, and correct interpretation of results.
Upon completion of the course, students gain a substantial understanding of modern computational tools in materials science, become familiar with the use of HPC, MD, DFT, and AI methodologies, and develop skills necessary for research and applications in materials science and technology.