Most practical materials for structural applications consist of multiple components, forming alloys.
Even the simplest steel contains both iron and carbon, for example, as well as many impurities, whether they are included on purpose or not. These dopants are in most cases distributed over the crystal lattice in a disordered fashion and modify the characteristics of the host material.
Atomic-scale simulations based on quantum mechanics often neglect that disorder, however,
although it has already been shown at many occasions that ordered and disordered alloys behave very differently. The growing power of high-level calculations has moreover allowed for their use in the design of new compounds. It is therefore necessary to extend these calculations to disordered materials, allowing their properties to be taken into account as well. Since simulations of disorder are not fully mature yet, this proposal intends to develop a fast yet accurate procedure to generate temperature-dependent information about the energetic and direction-dependent elastic behaviour. Such information is, for example, essential in the prediction of ductility, as multiscale materials scientists rely on accurate elastic tensors to describe the dynamics of dislocations. This study is also expected to provide new insights as to how randomly distributed alloying elements change a material's properties, enabling a high-throughput search for more suitable materials for nuclear fusion reactors.