PhD position, Computations and Machine learning for Nanomaterials, Luleå University of Technology, Sweden
In this project we will extract affordable potentials from expensive DFT results through machine learning (ML) to accelerate nanomaterial growth in MD simulations. Such interactomic potentials are also called force fields (FF). We have developed a ML-FF for carbon and iron that can be used to model the catalytic growth of carbon nanotubes, which we want to extend to include more elements, like hydrogen, oxygen, nitrogen, as well as other metals (copper, nickel, rutenium, platinum, …). This project involves using DFT to construct datasets of labelled atomic configurations, which will be used to train ML-FFs for MD simulations. Your research will for example target growth mechanisms and fundamental properties associated with nanomaterials. Based on these insights you will predict novel catalysts for controlled growth to achieve better quality products.
Deadline 2th May 2024