Designing materials for targeted performance is a dream in materials science. Although we have had some few successful cases of materials by design, we are still far from having robust models and computational tools which enable us to design materials for targeted performance. One approach to design materials is “Integrated Computational Materials Engineering (ICME)” which essentially combines different computational techniques at different length and time scales (from nanoscale to macroscale) to design more efficient materials. The scale which bridges the nanoscale to macroscale is called mesoscale (~μm). Materials behavior differently at mesoscale compared with nanoscale due to the heterogeneity nature of the mesoscale (presence of grain boundaries, defects, dislocations, etc.). This heterogeneity (and mastering that) makes the mesoscale a great opportunity for new materials design.

In this talk, I will discuss three case studies. First, I will describe a phase field model for shape memory materials and show how this model can help us to address the current challenge of shape memory materials (low durability) and design new fatigue resistant shape memory materials. The second and the third parts of the talk are about materials degradation in nuclear reactors. In these two sections I will illustrate how developing physics based models at mesoscale enable us to bridge the accelerated tests data, which are abundant, to actual light water reactor conditions, which our knowledge is limited about. This bridging empowers us to predict the life time of key components in light water reactors.