Oral: Computationally Driven Design of Ca-Ce-Ti-Mn Oxide Perovskites for Solar Thermochemical Hydrogen Generation

The development of high-performing materials for processes like two-step solar thermochemical hydrogen production (STCH) cycles is crucial to reducing the carbon footprint of hydrogen production. The computationally designed Ca-Ce-Ti-Mn oxide perovskite (CCTM), specifically Ca_2/3Ce_1/3Ti_1/3Mn_2/3O₃ (CCTM2112), has been identified as a highly promising material for STCH, outperforming the current state-of-the-art material, CeO₂ [1]. However, CCTM2112 is not yet the fully optimized composition of CCTM. Here, we optimize the composition of CCTM by employing Hubbard-U corrected density functional theory to improve stability and redox activity in STCH cycles. We compare the redox activity of various CCTM compositions, showing that the redox activity depends more strongly on the local environment of oxygen vacancies than on the global composition. Unlike CCTM2112, Ce⁴⁺ is absent in most compositions considered, leading to greater involvement of Mn cations in the redox process. Our findings show that the reduction of Ce³⁺ is more delocalized than that of Mn⁴⁺ and Mn³⁺. Additionally, we developed a crystal feature model with chemically intuitive descriptors to predict the neutral oxygen vacancy formation energy, achieving a mean absolute error of approximately 0.23 eV by utilizing available reference thermochemical data. This model enables us to explore a wide range of CCTM-based compositions.

[1] R. B. Wexler et al. Energy Environ. Sci. 16, 2550 (2023)