An Efficient ab Initio Scheme for Discovering Organic-Inorganic Hybrid Materials by Using Genetic Algorithms

Organic-inorganic hybrid materials (OIHMs), such as methylammonium lead triiodide (MAPbI₃), have a wide composition space because of the various potential combinations of organic molecules and inorganic cages. However, for unknown OHIMs, it is difficult to predict what kind of crystal structure will be stable without any experimental data. In this work, we report an efficient scheme for predicting crystal structures and phase diagrams of MA-Pb-I systems from first-principles calculations and genetic algorithms. In our scheme, OIHMs are divided into organic molecules and inorganic clusters. A pseudobinary phase diagram of MAI-PbI₂ was obtained by predicting structures at each composition. These results indicated that only MAPbI₃ and MA₂PbI₄ are stable phases, consistent with the experiments. In addition, the electronic and optical properties of the predicted structures were calculated and the solar cell performance was evaluated. Thus, our method allowed us to search for unknown OIHMs without any experimental data.