The materials properties that enable a range of energy related technologies, such as PV absorbers, transparent conductors, and photo‐catalysis, to mention just a few, tend to “live” in materials with specific compositions and crystal structures—and no others. The search for optimal energy materials should not be limited to the databases of currently known materials, but it should be extended to the large number of element combinations that define chemically possible compounds that are missing from such compilations. In order to predict the properties of such materials missing from the current databases, one first needs to predict whether they are thermodynamically stable, i.e., can be synthesized, and in what crystal structure. The problem of predicting the crystal structure of a material starting from the elemental constituents and without any bias is one of the most challenging problems in condensed matter physics. An important recent advance in this field has been the development of evolutionary algorithms coupled with density functional theory to predict the crystal structure of a solid without assumptions.

This project supports the development, testing, and application of improved crystal structure prediction methods based on state‐of‐the‐art ab initio electronic structure techniques and evolutionary algorithms to simultaneously search for materials with stable structures and target properties. The objective is to make these optimization methods scale to larger crystal structures than it is now possible and to apply them to search for materials that meet the design principles of desired functionalities without constraints and going beyond the databases of known compounds. In addition to these computational advances, in this work we will predict the stable structures and energy‐related properties of many new solids that might include numerous materials with thus far unsuspected and exciting functionalities for energy applications.