Published in

IOP Publishing, Nanotechnology, 43(24), p. 435702

DOI: 10.1088/0957-4484/24/43/435702

Links

Tools

Export citation

Search in Google Scholar

Nanostructure and strain in InGaN/GaN superlattices grown in GaN nanowires

This paper is available in a repository.
This paper is available in a repository.

Full text: Download

Red circle
Preprint: archiving forbidden
Red circle
Postprint: archiving forbidden
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

The structural properties and the strain state of InGaN/GaN superlattices embedded in GaN nanowires were analyzed as a function of superlattice growth temperature, using complementary transmission electron microscopy techniques supplemented by optical analysis using photoluminescence and spatially resolved microphotoluminescence spectroscopy. A truncated pyramidal shape was observed for the 4 nm thick InGaN inclusions, where their ([Formula: see text]) central facet was delimited by six-fold {[Formula: see text]} facets towards the m-plane sidewalls of the nanowires. The defect content of the nanowires comprised multiple basal stacking faults localized at the GaN base/superlattice interface, causing the formation of zinc-blende cubic regions, and often single stacking faults at the GaN/InGaN bilayer interfaces. No misfit dislocations or cracks were detected in the heterostructure, implying a fully strained configuration. Geometrical phase analysis showed a rather uniform radial distribution of elastic strain in the ([Formula: see text]) facet of the InGaN inclusions. Depending on the superlattice growth temperature, the elastic strain energy is partitioned among the successive InGaN/GaN layers in the case of low-temperature growth, while at higher superlattice growth temperature the in-plane tensile misfit strain of the GaN barriers is accommodated through restrained diffusion of indium from the preceding InGaN layers. The corresponding In contents of the central facet were estimated at 0.42 and 0.25, respectively. However, in the latter case, successful reproduction of the experimental electron microscopy images by image simulations was only feasible, allowing for a much higher occupancy of indium adatoms at lattice sites of the semipolar facets, compared to the invariable 25% assigned to the polar facet. Thus, a high complexity in indium incorporation and strain allocation between the different crystallographic facets of the InGaN inclusions is anticipated and supported by the results of photoluminescence and spatially resolved microphotoluminescence spectroscopy.