Context

Europe is currently spearheading the development of Lead-cooled Fast Reactors (LFRs)—a next-generation nuclear technology that uses molten lead as a coolant instead of water. Because lead has a very high boiling point and excellent natural cooling properties, LFRs offer enhanced safety, sustainability, and the potential for modular, cost-effective energy production. SCK CEN is a key member of the European consortium dedicated to bringing this technology to market, focusing on the design and licensing of these innovative systems in the form of a small modular reactor (SMR-LFR).

A critical issue in LFRs is the corrosion of the structural steel exposed to liquid lead. Understanding and mitigating this phenomenon is a crucial aspect of the design and development of a viable reactor. In pure lead, the corrosion process is controlled by both thermodynamic driving forces (solubility of metals in lead) as well as kinetic factors (the rate by which metals traverse the steel/coolant interface). In addition, corrosion is affected by changes in the coolant chemistry (such as controlled dissolution of oxygen) and flow conditions.

Recent advances in quantum mechanics (QM) and artificial intelligence (AI) have enabled the simulation of materials and interfaces at scales and accuracies that were unachievable years ago. Such simulations can also provide valuable insights into the corrosion process. At SCK CEN, progress has been made in developing such AI-enabled atomistic simulation techniques specifically tailored to the lead coolant chemistry.

Thesis Objectives

In this thesis, the student will apply state-of-the-art atomistic simulations to study phenomena relevant to corrosion in Pb. Possible aspects of this project include

• Benchmarking internal QM/AI models against publicly available foundation models
• Simulating oxygen dissolution and calculating its diffusion rate in Pb
• Quantifying diffusion rates of corrosion products such as Fe, Cr, and Ni in Pb
• Linking obtained transport quantities to corrosion rates through kinetic modeling
• Interpreting model predictions in the context of experimental results obtained in SCK CEN facilities and found in scientific literature

Candidate Profile

The ideal candidate should possess several of the following skills.

• A notion of computational chemistry (molecular dynamics, force fields, quantum chemistry)
• Basic programming in Python
• Familiarity with the Linux command line and HPC systems
• Strong interest in numerical techniques and machine learning