The need for microstructure informatics in process-structure-property relations

Abstract

Spatial hierarchy of microstructure is a defining characteristic of many practical materials systems. Elements of this hierarchy are often realized through nonequilibrium synthesis and process routes, leading to metastable structures that confer specific functionality and enhanced performance. The key to accelerating understanding and developing new and improved materials lies in quantifying microstructure in an unambiguous digital format, employing both physical models and data science methods to explore cause-and-effect relations between structure and properties and relations between composition-dependent process path history and hierarchical microstructure. Given the current state of predictive multiscale modeling, the uncertainties are simply too high to provide necessary decision support in isolation from experiments. Hence, combining experiments and computational modeling with materials data science and informatics provides the only practical path forward in replacing the historical paradigm of empirical materials development. The articles in this issue focus on microstructure informatics, which is relatively less well explored than the use of first-principles combinatorial methods applied to search the space of stable compounds, small molecules, and interface structures.