Anti-phase domains in cubic GaN

Abstract

The existence of anti-phase domains in cubic GaN grown on 3C-SiC/Si (001) substrates by plasma-assisted molecular beam epitaxy is reported. The influence of the 3C-SiC/Si (001) substrate morphology is studied with emphasis on the anti-phase domains (APDs). The GaN nucleation is governed by the APDs of the substrate, resulting in equal plane orientation and the same anti-phase boundaries. The presence of the APDs is independent of the GaN layer thickness. Atomic force microscopy surface analysis indicates lateral growth anisotropy of GaN facets in dependence of the APD orientation. This anisotropy can be linked to Ga and N face types of the {111} planes, similar to observations of anisotropic growth in 3C-SiC. In contrast to 3C-SiC, however, a difference in GaN phase composition for the two types of APDs can be measured by electron backscatter diffraction, μ-Raman and cathodoluminescence spectroscopy.

Germanium doping of cubic GaN grown by molecular beam epitaxy

“…c-GaN layers. 23 The excess material along these boundaries is only present on the two highest Ge-doped layers. Not intentionally doped layers, Si-doped layers, and layers with lower Ge-doping do not show this feature.…”

Point-group sensitive orientation mapping of non-centrosymmetric crystals

“…The implications of the broken inversion symmetry in these materials are seen, for example, in microstructural effects like antiphase domains and polarity inversion which need to be controlled in growth processes. [14][15][16][17] In this respect, EBSD is an attractive option for nondestructive analysis of semiconductor thin film systems, because of its typical depth sensitivity in the order of a few tens of nanometers. 18 In this letter, we demonstrate absolute orientation mapping for polycrystalline GaP, overcoming a fundamental limitation of the standard kinematic EBSD orientation determination which is used up to now.…”

Strain dependent electron spin dynamics in bulk cubic GaN

“…Additional sources of spin splittings would also cause a speed-up of DP relaxation. Such additional spin splittings could originate from the omnipresent inclusions of polar hexagonal GaN in the c-GaN matrix, 33 where the polar faces of the hexagonal GaN inclusions might act like a random Rashba field comparable to the random Rashba fields of dopant ions. 73,74 Furthermore, thermally activated carrier scattering between the cubic and the hexagonal GaN phases could lead to strongly enhanced spin relaxation due to the very fast spin relaxation in the polar hexagonal GaN.…”

“…[28][29][30] The experimental spin relaxation times in c-GaN are, however, still substantially shorter than theoretically predicted. 31 Unintentional strain fields caused by microstrain variations [32][33][34] or by a small h-GaN content 35 were discussed as a possible reason for the observed discrepancy. 27 The impact of strain fields on spin relaxation in c-GaN has, however, not been studied so far.…”