Influence of slight misorientations of r ‐plane sapphire substrates on the growth of nonpolar a ‐plane GaN layers via HVPE

Abstract

We have investigated the growth of nonpolar GaN templates by hydride vapor phase epitaxy (HVPE). This includes a systematic study of misoriented r‐plane sapphire wafers with a miscut angle up to ±1° towards the c‐axis of the crystal as starting substrates. Starting with an AlN nucleation layer approximately 3.3 μm of nonpolar a‐plane GaN are grown by metalorganic vapor phase epitaxy (MOVPE). The subsequent growth by HVPE was optimized to obtain flat and homogeneous layers with very good crystal quality. This was illustrated by very small full widths at half maximum (FWHM) of x‐ray rocking curve (XRC) measurements of less than 500 arcsec. The surface quality was evaluated by scanning electron microscopy (SEM) and optical phase contrast microscopy showing a high number of surface defects (pits) for the samples with a high miscut towards the positive direction whereas a slight miscut of +0.5° reduces the surface roughness. Additionally, smallest XRC FWHM values have been obtained for this particular miscut. A miscut towards negative directions results in lower crystalline quality and the surfaces shows hillock‐like features. All GaN layers on misoriented substrates are tilted with respect to the substrate. Their tilt angle increases with the miscut angle. Its direction is opposed to the miscut of the sapphire. (© 2010 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Measuring strain caused by ion implantation in GaN

“…We used the standard precursors ammonia (NH 3 ), trimethylgallium (TMGa) and trimethylaluminum (TMAl) and Pd-diffused H 2 as carrier gas. Non-polar a-plane and polar c-plane GaN samples have been deposited on r-oriented [28] and coriented [29] sapphire (Al 2 O 3 ) substrates, respectively, using an AlN nucleation layer. SiN nanomask layers were also incorporated to reduce the dislocation density.…”

Semipolar GaInN quantum well structures on large area substrates

“…The MOVPE growth of approximately 1 mm GaN was initiated with our standard oxygen doped low temperature AlN nucleation layer [12,13]. A few MOVPE grown samples have been overgrown by hydride vapor phase epitaxy (HVPE) with a GaN layer of about 10 mm thickness under similar conditions as applied for the growth on non-polar aplane templates [14].…”

Semipolar GaN grown on foreign substrates: a review

“…Therefore, a lot of effort was put into applying such strategies into the a-plane GaN growth. Both low-temperature GaN (see, e.g., [27][28][29]) and hightemperature AlN nucleation layers [30][31][32][33] yield acceptable GaN layer quality [34], while a low temperature AlN layer seems to result in lower quality layers (see results reported in [35]) and hence are much less appropriate [36,37], as opposed to the experience from c-plane growth. Yan et al reported a two-step nucleation combining low temperature and high temperature AlN layers achieving fairly good GaN layer quality [38].…”