High-throughput identification of point defects in SiC

Point defects in wide-band-gap semiconductors are used in many applications such as single photon emitters. Before a point defect can be utilized in these applications, an important step is to identify and understand both the defect type and different configurations. This is difficult due to the vast number of possible defects present in one material. A promising way to identify a defect is to combine experimental data with ab initio calculations which include zero-phonon lines and hyperfine coupling parameters. In earlier work, we made a convergence study for divacancies in 4H-SiC. Based on our understanding of the convergence of these calculations, we made a collection of automatic workflows called ADAQ [arXiv:2008.12539]. Here, each defect is calculated for a range of different configurations, charges, spins, and possible excitations. Currently, we are running these calculations and producing a database for an array of different defects. So far, we have screened about 8000 native defects in 4H-SiC. Our preliminary results suggest that with this choice of methodology, useful data are obtained at a feasible computational cost for a large number of defect types and configurations available in SiC.