Theoretical Solid-State Chemistry: theory, modelling, and simulation.
Location: CECAM-ES
Organisers
The extraordinary field of solid-state chemistry borders with solid-state physics, crystallography, quantum theory, inorganic, and organic chemistry, to name a few; also, it is one of the foundations on which the nanoscience and nanotechnology fields rely on. Computer simulation of the electronic structure and properties of solids have become sufficiently reliable such that their application has resulted in an increasingly important impact on solid state chemistry and physics. Although many courses and tutorials on the subject already exist, they are mainly focused on audiences with strong background on solid-state physics, and usually devoted to a particular electronic structure code. Far more unusual are the courses designed to teach the solid-state techniques to chemists, thus contributing to eliminate the cultural barriers that still exist between both groups.
This school is primarily targeted to PhD students and post‑docs who are interested or are starting to learn about the application of theoretical methods and techniques to the study of the physics and chemistry of the solid state. The level of this tutorial corresponds to master or doctorate students in areas of physics and chemistry. After two initial days where the fundamentals of theory of the treatment of the electronic structure of solids will be presented to the students, the remaining of the tutorial will be devoted to the examination of specific and hot areas like characterization of chemical bonding in solids and relationship to macroscopic properties, structure, and reactivity at solid surfaces, including layered systems and highly correlated oxides, and magnetic properties. The afternoons will be dedicated to practical hand-on tutorials using programs like (but not limited to) Quantum Espresso, Vasp, Vesta, jmol, Critic2g or SCALE-UP. Getting familiar with the different codes and their possibilities requires an adequate training that merges theory and practice in substantial amounts.
Standard methods in solid-state simulations allow for relatively large supercells containing up to 1000 atoms at zero temperature. However, predicting or interpreting many experimental results is still a great challenge due to finite temperature effects and defects like impurities, surfaces, grain boundaries, etc. For that reason, there is an ongoing effort in the solid-state community to develop methods that can be applied to larger systems. In this edition we will devote a theoretical session to discuss the factors that limit the application of first-principles simulations to these larger supercells and the kind of techniques that can be applied to overcome these problems. We will discuss the construction of Hamiltonians in localized basis sets and linear scaling techniques with a very brief incursion on the idea of model-Hamiltonians and QM/MM techniques. One practical session will be devoted to discussing these methods with applications on the calculation of optical properties in solids and the simulation of excitons in crystals so that the students get a more detailed look at how some of these complex topics are realized in specific applications.
Regarding the calculation of the vibrational structure of solids, the use of Machine Learning methods in the calculation of interatomic force constants, that drastically reduces up to two orders of magnitude the computational effort required for obtaining the lattice thermal conductivity, will be introduced.
References
Cristina Díaz (Universidad Complutense de Madrid) - Organiser
Pablo Garcia Fernandez (Universidad de Cantabria) - Organiser
Antonio M. Márquez (Universidad de Sevilla) - Organiser