Passer au contenu

Machine-learning approaches for simulations of phase transitions in nanoparticles




Duration: NanoX master Internship (8 months part-time in-lab immersion)
5 months full-time internship
6 months full-time internship

Latest starting date: 01/04/2022

Localisation: CEMES, CNRS, and Université de Toulouse, 29 rue Jeanne Marvig, 31055 Toulouse Cedex, France

Julien Lam
Magali Benoit

This research master's degree project could be followed by a PhD

Work package:
A key challenge in today’s nanotechnologies is the control of the structural properties during the nanoparticle synthesis. Reaching a targeted synthesis of nanoparticles requires a much better understanding of the involved complex mechanisms and in particular of crystal nucleation which corresponds to the initial structure formation. However, with the current state-of-the-art both in terms of experiments and simulations, controlling nucleation during nanoparticle synthesis remains a glass ceiling that needs to be overcome. In this project, we first introduce an original simulation approach based on machine-learning that will allow us to perform large scale simulations while retaining the accuracy of quantum calculations. Then, prompted by the proposed numerical development, we will study the example of iron oxides nanoparticles which offers a rich playground for fundamental understanding while also being considered in numerous technological applications.

-”Perspective: Machine learning potentials for atomistic simulations” J. Behler, J. Chem. Phys. 145, 17, 170901 (2016) -”Measuring transferability issues in machine-learning force fields: The example of Gold-Iron inter- actions with linearized potentials” M. Benoit, J. Amodeo, S. Combettes, A. Roux, I. Khaled, J. Lam, Mach. Learn.: Sci. Technol. 2 025003 (2021) -”Combining quantum mechanics and machine-learning calculations for anharmonic corrections to vibrational frequencies” J. Lam*, Saleh Abdul-Al, A-R Allouche*, J. Chem. Theory Comput. 13,3 (2020) -“Out of equilibrium polymorph selection in nanoparticle freezing” J. Amodeo, F. Pietrucci, J. Lam*, J. Phys. Chem. Lett. 11, 8060 (2020)

Areas of expertise:
Molecular dynamics, DFT, machine-learning, nucleation, crystallization

Required skills for the internship:
- Master in physics, chemistry or materials science - Good knowledge of statistical mechanics and computational physics/chemistry - Programming in C++/Python/Bash