Comparison of nitrogen-vacancy complexes in diamond and cubic SiC: dose dependencies and spin-Hamiltonian parameters

At present the nitrogen-vacancy (NV) complex in diamond is the most promising defect for application in the area of quantum computing. This provides a stimulus for an extensive search of other defects in semiconductors with similar properties. Recently it was shown that the N_CV_Si defect complex in SiC is perspectively appropriate for this goal as well. In the present work we perform comparative ab initio studies of NV complexes in diamond and 3C-SiC. We focus both on radiospectroscopic characterization of these defects and on the calculation of the equilibrium concentration of complexes in irradiated crystals. In particular a full set of spin-Hamiltonian parameters including g-tensors, hyperfine tensors and the spin-spin part of zero-field splitting constant D^ss were calculated for both negative and neutral charge states as well as for excited quartet states of neutral complexes. Comparison of calculated values with the available experimental data and results of other calculations show good agreement, especially in the case when hybrid and meta-hybrid functionals were used. This makes the unambiguous identification of negative NV complexes in both materials possible. Our calculations reveal that the ground states of neutral complexes are a difficult case for both DFT calculations and experimental observations. This is caused by multi-determinantal behavior of wave function for such complexes, which leads to a large amount of spin contamination and to the broken symmetry solution which appeared for single Slater determinant DFT calculations. Based on the calculated minimum of free energy of neutral and negative complexes in SiC and diamond we obtained the equilibrium concentrations of these complexes depending on the vacancy concentration produced by irradiation. We show that in some dose regions both negative and neutral complexes coexist, while in other regions only one charge state prevails. Comparison of the calculated and experimental dose dependencies for diamond shows good qualitative agreement.