Indonesia Conference Directory

Eu2+ ions are utilized as luminescent ions in solid-state lasers and phosphors. Therefore, the prediction of the 4f7→4f65d1 transition energy of Eu2+ in crystals is important to develop novel luminescent materials. In this work, we tried to predict the 4f7→4f65d1 transition energy of Eu2+ in oxides using first-principles calculations and machine learning. The first-principles calculations were performed based on the discrete variational multi-electron (DVME) method using small clusters composed of Eu2+ and all anions closer than the closest cation. Although the calculated 4f7→4f65d1 transition energies and the experimental ones showed some correlation, the theoretical values tend to be larger than the experimental ones by ca. 1.5 eV. Since machine learning enables one to create a predictive model of an output based on attributes, we tried to create a predictive model of the 4f7→4f65d1 transition energy of Eu2+ in oxides by machine learning using the calculated 4f7→4f65d1 transition energies and other electronic and structural parameters as the attributes. The obtained predictive model significantly improved the correlation between the predicted 4f7→4f65d1 transition energies and the experimental ones.