Atomistic modeling of grain boundary segregation in transition metals

Segregation of solute elements to grain boundaries (GBs) is a key factor for production and performance of many technologically relevant materials. It influences fundamental material properties such as formability, crack propagation, grain growth, precipitation, diffusivity or electric conductivity. By controlling the segregation state, a lever for developing materials of superior properties can be obtained.

We will present atomistic simulations based mainly on density functional theory which target the prediction of segregation energies and modification of cohesive properties in transition metals. First the anisotropy of GB properties in pure bcc and fcc metals will be discussed to identify different GB classes and to determine which can be considered as representative GBs. Next, segregation energies in the transition metals W, Mo and Cu will be presented and analyzed to understand whether simple modeling (Miedema model, Friedel model) can reproduce trends or how machine learning techniques can contribute. Then, we will show a comparison with experiment involving tracer diffusion for Cu and atom-probe tomography for a particular Mo-Hf alloy system. In this course also a new model for kinetics of segregation will be discussed. I will conclude by discussing the future challenges of segregation modeling and how databases can be created and used for grain boundary engineering.