ANR project
Design of new materials for SOLAR fuel production from machine learning and DFT calculations
Dates:
March 2026 – March 2030
Project coordinator:
Unité de Catalyse et de Chimie du Solide (UCCS), Lille
Partner laboratories :
- Catalysis and Solid Chemistry Unit (UCCS, Lille)
- IFP Energies Nouvelles (IFPEN)
IMN staff involved:
Houria KABBOUR, Romain GAUTIER, David BERTHEBAUD, Shunsuke SASAKI
Designing new materials for solar fuel production using machine learning and DFT calculations
The project presented proposes to design new materials for solar fuel production. In particular, we want to find new chemical compositions with stable bandgap crystal structures. These will improve the conversion of CO2 to CH4 and H2O to H2.
To this end, a new machine learning (ML) model developed by one of the project partners will be used. All the data used will be high-quality public data from the materialproject.org website.
Once the desired structures have been identified on the basis of their optoelectronic properties and their stability confirmed by DFT, they will be synthesized. They will also have been characterized and tested in water photoelectrocatalysis and CO2 photoconversion. We’ll be focusing on the generation of new materials from the family of heteroanionic photocatalysts, such as metal oxysulfides (MxOySz) and oxyhalides (MOX).
