Postdoc Predicting & Mitigating Liquid Copper Infiltration in Steels via Atomistic Simulations

Descrizione

Testo originale dell'annuncio (in inglese)

Join TU Delft and help enable liquid copper infiltration-resistant steels for a circular economy. As a postdoctoral researcher, you will unravel how silicon suppresses liquid copper infiltration at the atomic scale, using density functional theory-accurate machine-learned potentials and molecular dynamics simulations, in close collaboration with leading European research institutes and steel industry partners. Job description At TU Delft, you will contribute to a transformative research initiative focused on sustainable steel production. This project addresses a critical and growing challenge in metallurgy which is the liquid copper infiltration (LCI) in copper-contaminated steels leading to cracking during steel processing. While LCI is a known issue in conventional steel recycling, it becomes especially critical in the context of green steel production, where increased reliance on recycled scrap and electric arc furnace (EAF)-based routes promotes copper accumulation in steel, creating conditions that favour subsequent copper-induced embrittlement. Team Dey within the Computational Materials Science section at TU Delft is actively engaged in developing fundamental understanding for next-generation circular steelmaking. Using advanced atomistic modelling techniques, you will unravel the atomic-scale competition between copper and silicon at grain boundaries and oxide interfaces, delivering atomistic insights directly relevant to improving the recyclability and processability of both conventional and green steels. Where experimental work within the project focuses on process development and validation, this position addresses the underlying governing atomistic mechanisms. Within this position, you will employ molecular dynamics (MD) simulations to investigate the underlying atomistic mechanisms of LCI in steel grain boundaries and the inhibitory role of silicon. Your MD-based approach will elucidate how silicon disrupts copper wetting and diffusion. A central aspect of this project is the development of a Density Functional Theory (DFT)-accurate machine-learned interatomic potential (MLIP) for the multi-component steel system of interest. Ultimately, this simulation-driven framework will allow reliable identification of the new thresholds for copper content and the corresponding optimum silicon concentrations, thereby supporting the development of more robust recyclable steels. Your responsibilities In this role, you will develop fundamental insights into the atomistic mechanisms governing LCI at steel grain boundaries and the inhibitory role of silicon in copper-contaminated steels. As a postdoctoral researcher, you will: Perform molecular dynamics (MD) simulations to obtain atomic-scale insights into copper segregation, wetting and embrittlement at grain boundaries in steels Develop a Density Functional Theory (DFT)-accurate machine-learned interatomic potential (MLIP), enabling predictive MD simulations capable of resolving atomic-scale LCI mechanisms with near-DFT accuracy Investigate how silicon suppresses LCI, including its effects on grain boundary site competition and the formation of copper-silicon intermetallic phases Collaborate closely with researchers within department MSE as well as with the project partners including Leibniz-Institut Für Werkstofforientierte Technologien (IWT), Thyssenkrupp Steel Europe AG, Oulun Yliopisto and Ovako Sweden AB, to connect modelling insights with process development and alloy design Contribute to scientific publications, conference presentations and the development of new research proposals in the field of in the field of sustainable and circular steelmaking Your work environment You will be part of Team Dey within the Computational Materials Science section at TU Delft. This team focuses on atomistic simulations to investigate materials for sustainable energy, with proven expertise in hydrogen embrittlement, hydrogen storage and the behaviour o