H/F First principles validation of efficient thermoelectric materials predicted by machine learning.

The intern will first become familiar with the use of VASP, a state-of-the-art computational code widely employed to determine materials properties within the framework of Density Functional Theory (DFT). The main objective of the internship is to validate, through DFT calculations, the thermoelectric properties of half-Heusler compounds that have been predicted using machine learning (ML) techniques during the Ph.D. work of Shoeb Athar. Half-Heusler compounds crystallize in a face-centered cubic structure (space group F-43m), which makes them relatively straightforward to study computationally. These materials are particularly promising for thermoelectric applications: the thermoelectric effect allows the conversion of a temperature gradient across a solid into an electric current (and vice versa), offering potential solutions for energy recovery and efficiency.

The candidate should have a strong background in materials science and quantum mechanics, with a solid understanding of the fundamental concepts underlying electronic structure methods. Prior experience with shell scripting (e.g., Bash) and/or working in a Linux environment would be an advantage but is not strictly required. Most importantly, the intern should be motivated to work extensively with computational tools and enjoy problem-solving in a high-performance computing environment. In addition, this internship will help the student develop valuable transferable skills, including:

• Programming and data analysis through hands-on use of VASP, BoltzTraP, and scripting tools.
• Teamwork and collaboration, by interacting with PhD students and researchers working on related projects.
• Critical thinking and autonomy, by comparing machine learning predictions with first-principles results. This project is therefore well-suited for a motivated Master 2 student who wishes to strengthen both their expertise in computational materials science and their broader research skills.

Langues

English for all the scientific aspect, French for administrative questions.