Missions

The aim of this work is to predict new compounds and phase diagrams for as not yet explored chemical systems in the field of molten salts. This prediction will rely on generative techniques to accelerate the discovery of new compounds, as well as on high-throughput ab initio calculations to map phase diagrams and identify stable and metastable solids. This approach will also enable the extraction of features related to chemical compositions, crystal structures and electronic structures. Finally, data mining will be used to identify simple descriptors for predicting electrocatalytic properties, to be measured later by our project partners.

Activities

- Generate crystallographic cells using various methods (scored-based denoising, evolutionary, etc.)
- Estimate their stability (DFT calculation, neural network estimator, etc.).
- Explore the chemical space of molten salts, accelerating the process through various approaches (combinatorial, machine learning, etc.).

Skills

- Technical skills: Python programming, DFT calculation
- Solid state chemistry skills: crystallography, thermodynamics
- Ability to interact with experimentalists

Work Context

The host laboratory will be the LINK (IRL3629), an international unit with three supervision units (CNRS, Saint-Gobain, and NIMS), located in Tsukuba (https://link.cnrs.fr/).
This project is part of the PEPR DIADEM MADNESS program: Artificial intelligence-assisted methodology for the discovery of new molten salt materials.

The position is located in a sector under the protection of scientific and technical potential (PPST), and therefore requires, in accordance with the regulations, that your arrival is authorized by the competent authority of the MESR.

Constraints and risks

The researcher must already have been living in Japan for more than three months prior to recruitment and hold a “research” visa valid for the entire contract period. Travel within Japan and/or abroad possible.