EUROPEAN CONGRESS AND EXHIBITION ON ADVANCED MATERIALS AND PROCESSES - EUROMAT 2021

Several studies emphasise the impact of vacancies and impurities leading to severe degradation effects that have been observed experimentally in metals. To fully understand these effects first principle approaches are usually employed. However, such methods are constrained due to their computational cost and cannot simulate systems with more than 1000 atoms. On the other hand, interatomic potentials allow the simulation of larger-scale systems but cannot examine the effects of all the common non-metallic impurities.

For the identification of the vacancy clustering process and the mobility properties of vacancy clusters in Cu, a combination of DFT, Embedded Atom Method (EAM) Potentials along with a site occupancy algorithm have been used. To determine the mobility properties of the vacancy clusters, we propose an approach that combines ab initio and interatomic potential optimisation methods with an activation relaxation technique that utilises a saddle point search algorithm. Since regular EAM potentials do not allow the simulation of non-metallic impurities, our method has been extended by using bond order potentials along with modified EAM potentials that allow us to investigate the effects of H along with Al and Ti dopants in Cu grain boundaries. The potentials were found to accurately the relaxation effects of the examined impurities and dopants. H, Al and Cu vacancies were found to be more favourable to segregate into triple junctions compared to the grain boundaries or the bulk. As a final step to bridge the gap between the examined forcefields and ab initio approaches, linear scaling DFT methods were implemented and the modelling of grain boundary systems of more than 1000 atoms with impurities was achieved.