No abstract

Phone: þ46 13 286 745, Fax: þ46 13 137 568 4H-SiC epilayers were grown on 28 off-cut substrates using standard silane/propane chemistry, with the aim of characterizing in-grown stacking faults. The stacking faults were analyzed with low temperature photoluminescence spectroscopy, room temperature photoluminescence mappings, room temperature cathodoluminescence and synchrotron white beam X-ray topography. At least three different types of ingrown stacking faults were observed, including double Shockley stacking faults, triple Shockley stacking faults and bar-shaped stacking faults. Those stacking faults are all previously found in 48 and 88 off-cut epilayers; however, the geometrical size is larger in epilayers grown on 28 off-cut substrates due to lower off-cut angle. The stacking faults were formed close to the epilayer/substrate interface during the epitaxial growth.

On-axis homoepitaxy of 4H-SiC has the advantage of producing epilayers that are free of basal plane dislocations. Such layers can be highly beneficial for SiC-based high-power bipolar electronic devices which otherwise suffer from bipolar degradation phenomena related to basal plane dislocations in epilayers. In this study, we have developed on-axis homoepitaxy on the Si-face of 100 mm diameter 4H-SiC wafers with only 4H polytype in the epilayer excluding the edges of the wafer. We have also compared standard and chloride-based growth, the influence of different ambient conditions on surface preparation of the substrate, the influence of the histories of different growth cells, and the geometry of the susceptors regarding 4H-polytype stability in the epilayer. Substrate surface preparation, growth temperature, C/Si ratio, and Si/H ratio are found to be the most influential parameters to achieve homoepitaxy. On-axis homoepitaxial growth rate is limited to a very low value of <10 μm/h. We have performed a systematic study to understand the influence of different growth parameters and gas phase chemistries to determine whether on-axis growth rate can be enhanced and, if not, what the limiting factors are. Our experimental evidence suggests that the on-axis growth rate is not limited by the gas phase chemistry or diffusion, but it is limited by the surface kinetics. A significantly low step density on on-axis substrates lowers the surface reaction rates and limits the growth rate to lower values. It may not be possible to further improve the growth rate even with chloride-based growth using epitaxial growth conditions.

No abstract