Metal oxide candidates for thermochemical water splitting obtained with a generative diffusion model

Generative diffusion models for inorganic crystalline materials are being actively investigated for their potential to expand the chemical and structural design spaces for known functional materials. Generative candidates are particularly useful for applications where few functional, let alone commercially viable, materials currently exist, such as metal oxides for thermochemical water-splitting, which have strict requirements for defect thermodynamics and host stability. Here, we critically examine generated metal oxides from the MatterGen diffusion model conditioned on select chemical systems for thermochemical water splitting applications. Perhaps most notably, we find that MatterGen predicts a novel, thermodynamically stable, quinary metal oxide Ba2SrInFeO6, although this compound represents an ordered and layered substitution within the same A3B2O6 structural prototype as its two ternary end members. Detailed DFT investigations and spin configuration sampling for this material and its possible decomposition products---beyond what existed in MatterGen training data---are required to validate hull energy predictions and can change the conclusion as to whether Ba2SrInFeO6 is below or slightly above the previously known energetic hull (-2 vs. 9 meV/atom above the existing hull, respectively). Nonetheless, the material exhibits oxygen defect formation energies appropriate for thermochemical water splitting, warranting targeted investigation in an experimental validation campaign, along with other future MatterGen candidates in this application space.