A Multilevel Modeling Strategy Using GULP–ORCA Integration for Crystalline Materials

Abstract

Computational modeling of materials is a fundamental technological tool for scientific research. Its application enables the prediction of various material properties and provides explanations for the observed phenomena. Nevertheless, despite the range of existing computational techniques, their investigation can be challenging, especially given available computational resources. This work presents a multilevel computational methodology that integrates classical atomistic simulations performed with the General Utility Lattice Program (GULP) and electronic-structure calculations performed with ORCA software, aiming to significantly reduce the computational cost of studying crystalline materials. In this context, this study proposes a multilevel methodology for the computational investigation of crystalline materials, aiming to combine the strengths of the classical static atomistic method (CM) with DFT. The material used as the basis for the calculations was Y2O3, which is well-studied and has its properties thoroughly documented in the literature. In this multilevel methodology, optimization of the crystalline structure of Y2O3 was performed using CM, showing >99.6% agreement with experimentally measured lattice parameters and achieving a reduction of over 99% in optimization computation time compared to the DFT. Based on the structural information obtained through CM, calculations of the electronic properties of Y2O3 were performed using DFT. For this purpose, 26 different molecular structures of Y2O3 were considered with the number of atoms ranging from 5 to 350. The calculated bandgap energy values for these molecules ranged from 5.12 to 6.62 eV, and the electronic density of states (DOS) was also generated and analyzed. Calculations were additionally performed for the Y2O3 system doped with Eu3+ ions, where the formation of new electronic states in the DOS, attributed to the presence of the dopant, was observed.