No abstract

Magnesium (Mg) is the p-type doping of choice for GaN, and selective area doping by ion implantation is a routine technique employed during device processing. While electrically active defects have been thoroughly studied in as-grown GaN, not much is known about defects generated by ion implantation. This is especially true for the case of Mg. In this study, we carried out an electrical characterization investigation of point defects generated by Mg implantation in GaN. We have found at least nine electrically active levels in the 0.2-1.2 eV energy range, below the conduction band. The isochronal annealing behavior of these levels showed that most of them are thermally stable up to 1000 C. The nature of the detected defects is then discussed in the light of the results found in the literature.

No abstract

The static performance of different active and termination area designs for SiC-based Schottky diodes, suitable for 3.3kV applications, were investigated by means of extensive numerical simulations. We found quantitatively that the high electric field of SiC close to avalanche-breakdown is shielded most effectively from the Schottky interface by a trench-based design. Moreover, we conclude that the edge termination design with junction termination extension and four implantedp+guard rings is most robust against oxide interfacial charge.

While carbon doping is known to increase the resistivity of GaN, highly resistive layers for device isolation can also be obtained by ion implantation. In this study, we report on the electrical characterization of C-implanted n-type homoepitaxial GaN. Our investigation, carried out by capacitance-voltage measurements and deep level/minority carrier transient spectroscopy, revealed the presence of nine majority carrier traps in the 0.2–1.3 eV energy range, below the conduction band edge, and of four minority carrier traps, in the 0.1–1.4 eV energy range, above the valence band edge. The net-donor compensation mechanism and the behavior of defect centers are studied as a function of the annealing temperature in the 100–1000 °C range. While the former is explained in terms of dynamic annealing, the latter is discussed in the light of the present experimental results and those found in the literature.