The development of quantum materials is a source for new phenomena and innovative functionalities, and is becoming increasingly important as a starting point and foundation for next-generation industries. Predicting the functionalities of quantum materials based on existing theories is currently feasible. However, the successful synthesis of materials according to theoretical designs is not guaranteed. In reality, many theoretically predicted materials do not form, leading to explorations based on experience, intuition, and exhaustive searches, which have become bottlenecks in material development. Although a vast amount of experimental data on phase formation has accumulated through trial-and-error process in the material development, failures are barely disclosed. Furthermore, the chemical compositions of materials published in material databases are limited to those calculated or successfully synthesized by experiments. Consequently the experimental data required to accurately predict the phase formation of new materials is insufficient, making the prediction of phase formation in new materials a challenging task.

This study focuses on the layered perovskite oxyarsenides discovered by Ogino et al. These materials have a structure where an antifluorite layer and a perovskite layer are stacked. Due to the large number of possible combinations of constituent elements and the variable number of stacked layers in the perovskite structure, they are multicomponent systems with numerous variations. In ternary systems, the Goldschmidt tolerance factor, which consists of ionic radii, is known as an indicator for the phase formation of materials with an ABX₃ composition while retaining the pervskite structure. However, such phase formation determinants are not known for multicomponent systems. Therefore, this study aims to identify the phase formation determinants for multicomponent systems through a materials informatics approach.