Computational Chemistry Software
Introduction
This series of courses covers hands-on operation of mainstream computational tools such as VASP, QE, and CP2K, as well as the fundamentals of ASE and Packmol modeling and efficient application of ai²kit. It systematically guides software usage, optimization, and molecular dynamics simulation.
Course List
- ASE&Packmol Modeling Fundamentals
This course briefly introduces ASE and Packmol, two modeling tools. It covers how to use ASE modules for atomic operations, visualization, interface and bulk phase modeling, as well as Packmol modeling setup formats and keywords, such as inside cube and outside cube. - Quantum EspressO/VASP Practice
This course introduces the operation of VASP and QE software, including structure optimization, static self-consistent calculations, and band calculations. It details input and output file parameter settings, task submission and runtime monitoring, as well as result analysis and visualization methods, and includes relevant operation examples. - Molecular Dynamics/Machine Learning Molecular Dynamics Practice
This course aims to introduce the basic principles of classical molecular dynamics (Classical Molecular Dynamics) and machine learning molecular dynamics (Machine Learning Molecular Dynamics). The course introduces how to use molecular dynamics simulations to study the physicochemical properties of materials and molecules, and understand how to use machine learning methods to generate potential functions to improve the efficiency and accuracy of molecular dynamics simulations. The classical molecular dynamics section covers the basic principles of molecular dynamics, atomic interatomic potentials and classical force fields, numerical solutions of Newton's equations of motion, concepts of statistical ensembles, temperature and pressure control methods, and the implementation process of molecular dynamics simulations. The machine learning molecular dynamics section covers the relationship between machine learning potentials and classical mechanics and quantum mechanics, the principles of machine learning molecular dynamics, the concepts of atomic energy and deep potential smooth models, and the structure and optimization methods of machine learning models. The course is suitable for beginners in materials science, chemistry-related fields, and those interested in molecular simulation and machine learning. - CP2K Practice
This course focuses on efficient computational simulations using the CP2K software. The course content is divided into five parts, designed to guide newcomers in mastering computational skills from basic to advanced levels:
①Single-point energy calculation. Single-point energy is the total energy of a molecule at a specific geometric configuration, which is the basis for understanding molecular properties.
②Static structure optimization: By adjusting the geometric configuration of a molecule to achieve a stable state with minimum energy, the true structure of the molecule is revealed.
③NEB transition state search: Constructing NEB pathways and analyzing the structure and energy characteristics of transition states to gain in-depth understanding of chemical reaction mechanisms.
④Frequency calculation: Revealing the vibrational modes of molecules, including normal modes and imaginary frequencies, to understand molecular stability and reactivity.
⑤Molecular dynamics simulation. Simulating the motion and interactions of molecules over long time scales. - Introduction to ai²-kit
The efficient training of machine learning potential functions relies on efficient data set collection. Through the ai²-kit synchronous learning workflow, users can efficiently explore unknown chemical spaces and expand data sets by combining molecular dynamics methods. This approach can also be coupled with enhanced sampling methods to further improve sampling efficiency.