Defect Formation Energies in Metals and Semiconductors made comprehensible

The study of defects in materials is of utmost importance for technological applications and the improvement and synthesis of new materials. In this work, we analyze the performance of density functional approximations (DFAs) on two prototypical sets of defective systems: monovacancies in eight fcc metals, and interstitials in the semiconductor Si-diamond. Specifically, we studied defect formation energies using LDA, PBE (GGA), the meta-GGAs SCAN, r²SCAN, and LAK, and the hybrid HSE06. For metals, LDA shows a better performance with respect to the other DFAs, whereas for Si-diamond, LAK provides an outstanding performance, overcoming HSE06 and comparable to more computationally expensive methods such as QMC. Furthermore, to explain the DFAs performance, we delved into the behavior of the DFAs’ ingredients (such as r_s, s and α) in the core, semi-core, and valence bond regions. Our preliminary results indicate the critical regions that influence the observed trends of the defect formation energies and pave the way for improving DFAs.