PhD: AI-Driven Microdroplet Chemistry for CO₂ Valorization

Association Bernard Gregory; Paris; Hybrid; EUR 22,000 - 27,000; Full time

Benefits offered by this job

Remote work

Mobility support

CROUS canteen

Health insurance support

Professional development

Job summary

MICADO - Microdroplet catalysis of CO2-amine C–N bond formation seeks to uncover molecular mechanisms of microdroplet catalysis and to build a predictive multiscale model linking droplet properties to reaction rates and product distributions. The ENS-PSL team will develop MLIPs with active learning to enable large-scale simulations at quantum accuracy.

The project is international with collaboration including ETH Zürich.

Qualifications

• Master’s degree in theoretical/ computational chemistry or closely related field.
• Solid background in quantum chemistry and statistical mechanics.
• Experience with MLIPs or AI/ML applied to molecular systems is asset.
• Programming skills expected.

Responsibilities

• Develop MLIPs trained at hybrid-DFT accuracy using active learning.
• Perform large-scale reactive molecular dynamics simulations with machine‑learning potentials.
• Explore reaction mechanisms via MLIP-based MD and path sampling.
• Collaborate with ETH Zürich for experimental benchmarks and validation.

Skills

Machine learning

Python programming

Statistical mechanics

Quantum chemistry

Molecular dynamics

Education

Master’s degree in theoretical chemistry

Tools

ArcaNN

MACE-OMOL

Hybrid-DFT

Job description

MICADO - Microdroplet catalysis of CO2-amine C–N bond formation seeks to uncover molecular mechanisms of microdroplet catalysis and to build a predictive multiscale model linking droplet properties to reaction rates and product distributions. The ENS-PSL team will develop MLIPs with active learning to enable large-scale simulations at quantum accuracy.

The project is international with collaboration including ETH Zürich.