No description available.

No description available.

No description available.

In this thesis, calculations of Auger recombination rates in semiconductor quantum wells are presented. Chapter One introduces Auger recombination, and the reasons for studying the Auger process are explained. Basically, Auger recombination is a non- radiative recombination mechanism that becomes more important as the carrier density increases and the bandgap decreases. In direct gap semiconductors, the Auger process has an activation energy, and the resulting highly temperature dependent Auger process is thought to be a possible cause of the high temperature sensitivity of long wavelength semiconductor lasers that are being considered for use as sources in optical fibre communications systems. In Chapter Two, an expression is derived for the CHSH Auger recombination rate in a quantum well (QW) heterostructure. The possible Auger processes in a QW are discussed as are the differences between Auger recombination in a QW and in bulk semiconductors, and the magnitudes of QW and bulk Auger rates are compared. In Chapter Three, the theory of Auger recombination is extended to the case of a quantum well wire (QWW), a semiconductor structure in which carriers are free to move in one direction only. It is found that there are no significant physical differences between Auger recombination in a QW and in a QWW. The ratio of QW and QWW Auger rates is evaluated. Numerical results for Auger transition rates in 1.3µm and 1.55µm In- GaAsP/InP QWs and QWWs are presented in Chapter Four, and comparison with experimental values is made. In particular, the result found in Chapter Two, that, under certain conditions, the Auger rates in the QW and the bulk are approximately the same is found to agree with experimental results from the literature. The derivation of the CHSH Auger transition rates in QWs and QWWs that was presented in Chapters Two and Three required a number of approximations concerning the carrier statistics and the semiconductor bandstructure. In Chapter Five, these approximations are examined, and, although it is found that the use of non-degenerate carrier statistics is reasonably accurate, the assumption of parabolic energy bands can lead to overestimates of .the Auger transition rates. The first five chapters constitute the first part of the thesis, concerning Auger recombination in low-dimensional semiconductor structures. In the second part of the thesis, the realistic bandstructure of low-dimensional semi conductor structures, such as superlattices, is examined. The method used is described in Chapter Six, and is based on an empirical pseudopotential method. Results for the GaAs/AlAs superlattice are presented in Chapter Seven.

Semiconductor devices can be fabricated by growing III-V heteroepitaxial layers which are coherently strained to a III-V substrate. The relaxation of layer lattice strain through the nucleation of misfit dislocations near the interface causes a drop in performance for these devices. This thesis uses two non destructive x-ray techniques to examine relaxation in III-V epitaxial layers; double crystal diflractometry and x- ray topography. The dynamical and kinematical theories of x-ray diffraction are discussed in chapter 2. The apparatus used for double crystal diffractometry and x-ray topography and the theory of operation of these techniques is discussed in chapter 3. The properties of misfit dislocations in III-V epitaxial layers and the critical layer thickness at which relaxation occurs are discussed in chapter 4.Double crystal diffractometry and x-ray topography have been used to examine relaxation in epitaxial layers of AlAs on GaAs, InGaAs on GaAs, GaAsSb on GaAs, InGaAs on InP and an InGaAs superlattice on InP. All layers were deposited on 001 orientated substrates. Asymmetric double crystal rocking curves have been analysed using a novel technique which allows deduction of the position of an hhl layer reflection in reciprocal space. The layer unit cell parameters in the [110] and iTO] directions are determined from this. Individual misfit dislocation lines can be resolved by topography for dislocation line densities less than 0.2 μm(^-1)In each of these samples the layer relaxation was found to be asymmetric about the (110) directions. The sensitivity of diffractometry and topography to the detection of layer relaxation has been compared for samples with different thicknesses and dislocation line densities. The resolution of these techniques to the determination of layer relaxation has been shown to meet for a 1 μm layer of AlAs on GaAs. Tilt between the epitaxial layer lattice and the substrate has been measured for coherently strained and partially relaxed epitaxial layers grown on 001 orientated substrates. The lattice tilt in (110) directions was found to increase with misfit dislocation line density in these directions. Two theoretical models have been developed describing the relationship between lattice tilt and misfit dislocation line density and the tilts predicted by these compared with experiment. At high dislocation densities measurements of layer relaxation by diffractometry indicate that the images recorded by topography represent bundles of misfit dislocations and not individual dislocation fines. The number of dislocation lines per bundle was found to decrease with decreasing layer relaxation. Bunching of misfit dislocations into dislocation bundles is also observed on topographs from a low dislocation density sample where the individual dislocation hues are resolved. Screw dislocations in a strained layer and an interaction between two 60 dislocations to form a mixed dislocation have been characterised using Burgers vector analysis. Interference fringes have been observed on 004 double crystal rocking curves recorded from an ultra thin In GaAs layer sandwiched between a GaAs substrate and a GaAs cap. The position and intensity of these fringes was found to be sensitive to the composition and thickness of the In GaAs layer. Comparison between simulated and experimental rocking curve data allowed determination of the layer thickness to within a single monolayer and layer composition to within 0.5%. Topography of this sample showed that the dislocation line density varied from zero to 0.12 μm(^-1)across the wafer. The critical layer thickness and Indium concentration at which the first few misfit dislocation fines were observed was measured as 162 ± 2 A and 17 ± 0.5 %.

No description available.

No description available.

No description available.

No description available.

No description available.