In this contribution, we demonstrate that for non‐ and semipolar AlInN one‐dimensionally lattice‐matched to GaN, controlled relaxation in different in‐plane directions of the growth surface can be induced. All of our samples were grown on false(11true22‾false) GaN templates on patterned r‐plane sapphire as well as on m‐oriented 6H‐SiC substrates and free‐standing pseudo‐bulk GaN substrates. The InN mole fraction of the Al1−xInxN is varied from 18% up to 28% corresponding to one‐dimensional lattice matching in various in‐plane directions of the growth surface. By controlling the composition as well as the thickness, we realized one‐dimensional relaxation of the AlInN layers along the nonlattice‐matched direction. This opens new perspectives for strain and polarization engineering in optically active layers grown subsequently and gives new possibilities to design high indium‐containing GaInN quantum well structures for efficient long wavelength light emitters.
We demonstrate a strong dependence of the indium incorporation efficiency on the strain state in m-oriented GaInN/GaN multi quantum well (MQW) structures. Insertion of a partially relaxed AlInN buffer layer opens up the opportunity to manipulate the strain situation in the MQW grown on top. By lattice-matching this AlInN layer to the c- or a-axis of the underlying GaN, relaxation towards larger a- or smaller c-lattice constants can be induced, respectively. This results in a modified template for the subsequent MQW growth. From X-ray diffraction and photoluminescence measurements, we derive significant effects on the In incorporation efficiency and In concentrations in the quantum well (QW) up to x = 38% without additional accumulation of strain energy in the QW region. This makes strain manipulation a very promising method for growth of high In-containing MQW structures for efficient, long wavelength light-emitting devices.
We report on (112¯2) oriented Al1−xInxN grown by low pressure metal organic vapor phase epitaxy on (112¯2) GaN templates on patterned r-plane sapphire. The indium incorporation efficiency as well as the growth rate of (112¯2) oriented layers are similar to c-plane oriented Al1−xInxN layers. Deposition of thick Al1−xInxN layers does not lead to additional roughening like in case of c-plane oriented Al1−xInxN. Independent of the thickness, the degree of relaxation of layers lattice matched in m-direction is in the range of 33%–45% in [112¯3¯]-direction. Associated with the relaxation in [112¯3¯]-direction, there is a tilt of the Al1−xInxN layers around the [11¯00] axis due to slip of threading dislocations on the basal (0001)-plane. Relaxation in m-direction is not observable for layers lattice matched in [112¯3¯] direction. The possibility to adjust the lattice parameter of AlInN in [112¯3¯] direction without changing the lattice parameter in m-direction by anisotropic strain relaxation opens up opportunities for subsequent growth of optically active structures. One possibility is to form relaxed buffer layers for GaInN quantum well structures.
We investigated the dependence of the indium content of tenfold Al1−xInxN/GaN superlattice structures grown by metal organic vapor phase epitaxy on layer thickness and strain state. Growth conditions taken from a thick lattice-matched reference sample with an indium content of about 18% lead to reduced indium contents from 3% for 0.5 nm of Al1−xInxN to 16.5% for 5.0 nm, respectively. There is no evidence for dependences of the indium incorporation on the lattice mismatch between the Al1−xInxN and the subjacent layer. Additional supply of trimethylindium only shows a very slight, almost negligible influence on the indium content of these superlattice structures. Finally, we present a model explaining the behavior of the indium content of the Al1−xInxN layer assuming the growth of an indium depleted phase in the initial stage of growth.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.