We have performed a detailed investigation of the photoluminescence features taken at 2 K on a series of GaxIn1-xN alloys grown by metal-organic vapour-phase epitaxy through the whole composition range. The evolution of the photoluminescence lineshape of GaInN alloys in the indium-rich region is dominated by doping effects rather than by band-gap tailing effects correlated to existence of random chemical crystal inhomogeneities. The lineshape of the photoluminescence indicates a residual electron concentration of about 1018–1019cm-3 in the bulk part of the epilayers. The value we get for the bowing parameter is b=2.8 eV.

We report on the MOCVD growth of InN buffer layers on sapphire substrate for InN growth. The approach used assumes that an optimized InN buffer layer has to exhibit at least the same crystalline quality and sapphire surface coverage than the GaN buffer layers allowing to grow high crystalline quality GaN on sapphire. The buffer layers were characterized by AFM and GID measurements. Sapphire nitridation was investigated: it has a strong influence on in-plane crystalline quality. Two kinds of buffer layers were optimized according to the GaN buffer layer specifications: one of them only presented In droplets at its surface. It was shown that the small amount of In droplets increases the adatoms mobility of the main layer overgrown, leading to a 25% decrease of its in-plane mosaicity, compared to InN films directly grown on sapphire. To achieve a same improvement on InN buffer layer free of In droplets, the InN main layer growth temperature had to be increased from 550 °C. to 600 °C.

We report on the use of dimethylhydrazine (DMHy) and tertiarybuthylhydrazine (TBHy), as alternative nitrogen precursor for GaN low-temperature growth, as well as to improve the InN growth rate. Lowering the GaN growth temperature, would allow growing InN/GaN heterostructures by MOVPE, without damaging the InN layers. Increasing the low InN MOVPE growth rate is of major importance to grow reasonably thick InN layers. In this respect, triethylindium (TEIn) was also used as an alternative to trimethylindium (TMIn). We demonstrate that tertiarybutylhydrazine is a suitable nitrogen source for GaN growth at 600 °C, leading to crystalline GaN layers. These layers exhibited a peak of photoluminescence emission at 3.475 eV, at 2 K, with a narrow linewidth of 18 meV; while no detectable photoluminescence was observed for layers grown with ammonia. This result opens the way to further MOVPE growth of InN and GaN based heterostructures. The unsuccessful GaN growth in a nitrogen or argon ambient using DMHy, in contrast with hydrogen, demonstrates the active role of hydrogen in the growth process. Both DMHy and TBHy were unsuitable to grow InN, but the use of TEIn instead of TMIn together with ammonia resulted at least in doubling the InN growth rate in same growth conditions

The natural aging process of InN nanostructures by the formation of indium oxides is examined by transmission electron microscopy related techniques. Uncapped and GaN-capped InN quantum dots (QDs) on GaN/sapphire substrates were grown under the same conditions and kept at room temperature/pressure conditions. The GaN capping layer is found to preserve the InN QDs in the wurtzite phase, avoiding the formation of group-III oxides, while in the uncapped sample, a thin layer of cubic phases are formed that envelops the nucleus of wurtzite InN. These cubic phases are shown to be mainly bcc-In2O3 for long aged samples where the nitrogen atoms in the InN surface layers have been substituted by atmospheric oxygen. This process implies the gradual transformation of the In sublattice from hcp to a quasi-fcc structure. Metastable zinc-blende InN phases rich in oxygen atoms are proposed to act as intermediate phases and they are evinced in samples less aged. The large concurrence of interplanar spaces, the twin formation, and the existence of a free surface that facilitates the transformation support this mechanism and would explain the high instability of the InN nanostructures at ambient conditions

We extend to any temperature, the sophisticated calculation of the evolution of the 2 K photoluminescence energy of InN proposed by Arnaudov et al. [Phys. Rev. B 69, 115216 (2004)], in view of determining the residual doping of thin films. From the detailed line shape modeling, we extract the full width at half maximum of the photoluminescence line which, in the first order, varies like n0.51 at low temperature. This allows us to propose a handy tool for rapid residual doping evaluation. Last, temperature and inhomogeneous broadening effects are analyzed. Ignoring the latter is shown to lead to an overestimation of the residual doping.

Simple calculations performed in the context of linear elasticity theory, indicate that growth of InN on (hkl)-oriented substrates minimizes the density of elastic energy stored in the strained layers and boosts the critical thickness for coherent growth of strained layers to unexpected values. This renders possible to grow two-dimensional InN–GaN structures with InN layers thicknesses larger than one monolayer and, in addition, continuous in the growth plane. How impacting this finding can be for optoelectronic devices using indium nitride-based heterostructures is finally discussed.

During molecular beam epitaxy of GaN:Tm films, substrate temperature strongly influences the rare earth incorporation, surface morphology and luminescence spectrum. The Tm incorporation into films grown between 730 and 830 degrees C was estimated by wavelength-dispersive X-ray (WDX) spectroscopy. Comparative WDX, atomic force microscopy (AFM) and cathodoluminescence (CL) mappings reveal that at an optimal growth temperature between 775 and 780 degrees C, a high Tm content (similar to 2.2 at%) and a smooth surface morphology can be obtained, leading to an intense sharp TM3+ emission. For lower substrate temperatures, Ga droplets and large (similar to 8-15 mu m) circular pits mar the sample surface; for higher temperatures, the sharp CL lines disappear due to low Tm content (<= 0.8 at%). (C) 2008 Elsevier B.V. All rights reserved.