Silicon carbide is a wide bandgap semiconductor with the unique electrical and physical properties, including high breakdown voltage and excellent thermal conductivity. However, for the fabrication of a new generation of electronic devices operating at high power densities, high frequencies, and high temperatures, semi-insulating silicon carbide (SI SiC) wafers with a resistivity higher than 105 Ωcm are necessary. These devices can be produced either by homoepitaxy (SiC MESFETs) or heteroepitaxy (GaN/GaAlN/GaInN structures) on SI SiC substrates. The semi-insulating SiC crystals are obtained by compensating the residual shallow donors and acceptors with deep-level centers introduced either by vanadium doping or by controlled generation of native point defects. In such crystals, the Fermi level is pinned near the midgap to a partially ionized deep level. The former method is widely used for SiC bulk crystals grown by conventional physical vapor transport (PVT) technique. The latter one is more effective for the crystals of higher purity grown by high temperature chemical vapor deposition (HTCVD).
In this work, we show a very useful tool for simulating the electron and hole concentrations in 4H and 6H-SiC as a function of temperature, assuming the concentrations of the defects participating in the charge compensation, namely: the shallow donors, shallow acceptors and deep defect centers. The charge carrier concentrations, as well as the material resistivity are calculated in the temperature range between 50 and 800 K. The model is based on the numeric solution of the charge neutrality equation and determination of the Fermi level position for a given temperature. In the calculations, the temperature dependence of the bandgap energy is taken into account. On the grounds of the simulation results, the electrical properties of 4H and 6H-SiC correlated with the temperature changes in the Fermi level position for the nitrogen, boron and vanadium (or the native defect) concentrations in the range of 1015-1018 cm-3 are discussed. The modeling results are shown to be in a good agreement with the experimental data. The importance of the simulations for the optimization of the crystal growth processes aimed at obtaining the semi-insulating material is presented.