All theses since 2009 can be downloaded in full from the University library by following the given link. This list can also be downloaded as a single BibTeX file.

2023

  • D. A. Hunter, “Investigation of new semiconductor materials for wide band-gap devices,” PhD Thesis, 2023. doi:10.48730/ztp9-av57
    [BibTeX] [Abstract]

    This thesis investigates the compositional and optical properties of wide bandgap III-nitride and gallium oxide semiconductor materials using scanning electron microscopy techniques. The primarily used techniques used here were wavelength dispersive X-ray spectroscopy (WDX) and cathodoluminescence (CL) with other electron microscopy and spectroscopic techniques used to assist in the characterisation of the semiconductor materials. Compositional measurements were performed to determine bulk alloying levels within the semiconductor and trace element analyses e.g doping concentrations. Investigations benefited from the correlative nature of the WDX and CL measurements, allowing simultaneous compositional and optical mapping of the semiconductor, attributing the sample luminescence to compositional spatial regions or features of the material. Semi- and non-polar AlₓGa₁₋ₓN alloys showed similar Al incorporation as polar material, consistent across the entire AlN range. Si incorporation within semi-polar AlₓGa₁₋ₓN was found to be independent of bulk composition however the dopant concentration resulting in onset of Si self-compensation increased with AlN composition unlike polar AlₓGa₁₋ₓN. CL measurements show good agreement with the near-band edge (NBE) emission energy and the material bandgap expected from the alloy concentration for all AlₓGa₁₋ₓN crystal orientations. Compared to polar AlₓGa₁₋ₓN NBE emission showed a high degree of broadening with increasing AlN content due to compositional variation in the samples. NBE broadening was also accompanied with intense defect luminescence attributed to oxygenated VIII complexes. Indium-gallium oxide (IGO) alloys were grown with low to high In contents with notable crystallographic phase changes from monoclinic to mixed to cubic as the In content was increased. CL measurements show UV, blue and green luminescence with the spectral intensities shifting from dominant UV in near pure Ga₂O₃ to enhanced blue and green in high In containing materials. Spectral band energies also decreased as the In composition increased due to the reduction of the material bandgap. The incorporation of Sn into Ga₂O₃ alloys was investigated for thin films grown on various material substrates. Sn alloying levels was found to be dependent on multiple growth factors: Sn availability, substrate choice and growth temperature. Optical properties exhibited similar to the IGO samples for both intensity and energy shifts of the three spectral bands. Electronic investigations into the photodetector properties of the TGO films showed superior responsivity and high gain compared to Ga₂O₃ devices while operating in lower energy UV due to the bandgap reduction. An investigation into the impact of X-ray secondary fluorescence on WDX measurements on semiconductor thin films was completed as a separate computation study. Results show there was a negligible effect of secondary fluorescence on WDX measurements on multiple semiconductor families with various material substrates, unlike for geological samples. The reduced beam energy required for semiconductor thin film analyses minimises the impact of secondary fluorescence on bulk quantitative measurements. However trace element analyses may be impacted by secondary fluorescence when using the typical beam conditions for thin film specimens, particularly when the substrate also contains the trace element, the impact of secondary fluorescence may be reduced however by operating with the lowest suitable beam energy possible, minimising the size of the secondary excitation volume.

    @PhdThesis{Hunter-thesis,
    author = {Daniel A. Hunter},
    school = {University of Strathclyde},
    title = {Investigation of new semiconductor materials for wide band-gap devices},
    year = {2023},
    abstract = {This thesis investigates the compositional and optical properties of wide bandgap III-nitride and gallium oxide semiconductor materials using scanning electron microscopy techniques. The primarily used techniques used here were wavelength dispersive X-ray spectroscopy (WDX) and cathodoluminescence (CL) with other electron microscopy and spectroscopic techniques used to assist in the characterisation of the semiconductor materials. Compositional measurements were performed to determine bulk alloying levels within the semiconductor and trace element analyses e.g doping concentrations. Investigations benefited from the correlative nature of the WDX and CL measurements, allowing simultaneous compositional and optical mapping of the semiconductor, attributing the sample luminescence to compositional spatial regions or features of the material. Semi- and non-polar AlₓGa₁₋ₓN alloys showed similar Al incorporation as polar material, consistent across the entire AlN range. Si incorporation within semi-polar AlₓGa₁₋ₓN was found to be independent of bulk composition however the dopant concentration resulting in onset of Si self-compensation increased with AlN composition unlike polar AlₓGa₁₋ₓN. CL measurements show good agreement with the near-band edge (NBE) emission energy and the material bandgap expected from the alloy concentration for all AlₓGa₁₋ₓN crystal orientations. Compared to polar AlₓGa₁₋ₓN NBE emission showed a high degree of broadening with increasing AlN content due to compositional variation in the samples. NBE broadening was also accompanied with intense defect luminescence attributed to oxygenated VIII complexes. Indium-gallium oxide (IGO) alloys were grown with low to high In contents with notable crystallographic phase changes from monoclinic to mixed to cubic as the In content was increased. CL measurements show UV, blue and green luminescence with the spectral intensities shifting from dominant UV in near pure Ga₂O₃ to enhanced blue and green in high In containing materials. Spectral band energies also decreased as the In composition increased due to the reduction of the material bandgap. The incorporation of Sn into Ga₂O₃ alloys was investigated for thin films grown on various material substrates. Sn alloying levels was found to be dependent on multiple growth factors: Sn availability, substrate choice and growth temperature. Optical properties exhibited similar to the IGO samples for both intensity and energy shifts of the three spectral bands. Electronic investigations into the photodetector properties of the TGO films showed superior responsivity and high gain compared to Ga₂O₃ devices while operating in lower energy UV due to the bandgap reduction. An investigation into the impact of X-ray secondary fluorescence on WDX measurements on semiconductor thin films was completed as a separate computation study. Results show there was a negligible effect of secondary fluorescence on WDX measurements on multiple semiconductor families with various material substrates, unlike for geological samples. The reduced beam energy required for semiconductor thin film analyses minimises the impact of secondary fluorescence on bulk quantitative measurements. However trace element analyses may be impacted by secondary fluorescence when using the typical beam conditions for thin film specimens, particularly when the substrate also contains the trace element, the impact of secondary fluorescence may be reduced however by operating with the lowest suitable beam energy possible, minimising the size of the secondary excitation volume.},
    doi = {10.48730/ztp9-av57},
    }

2022

  • D. Cameron, “Nanocharacterisation of III-nitride semiconductors,” PhD Thesis, 2022. doi:10.48730/yk6s-8t20
    [BibTeX] [Abstract]

    Scanning electron microscopy allows for the investigation of materials down to the nanometre scale. In this thesis I combine cathodoluminescence spectroscopy with complementary techniques to investigate a range of III-nitride semiconductor heterostructures. I first describe AlGaN-coreshell nanorod UV-LED structures. Here, I conducted a wideranging study investigating the optical, electrical and structural properties of these structures. The electrical measurements were enabled by a nanoprobing system for in-situ contacting within the scanning electron microscope combined with bespoke focused ion beam deposition and milling to create contacts on individual nanorods. I-V measurements and electron beam induced currents confirmed the formation of a p-n junction and successful doping within the structure. In the same microscope, I measured CL to find the emission properties and found optically active quantum wells emitting near 300nm, with some variation in energy and intensity as well as clustering. In a transmission electron microscope, energy dispersive X-ray analysis confirmed the existence of clusters occurring at a-plane nanofacets. Low temperature and time resolved measurements measure the speed of the recombination in the quantum wells, showing a reduction in the quantum confined Stark effect within these structures. Using this same combination of techniques, I also examined planar UV-LED structures. I found that screw dislocations pinned the position of hexagonal hillocks during growth. In each of the penetrated layers, higher point defect densities and three-dimensional growth enclose the threading dislocations. These point defects can compensate electrical dopants and increase parasitic recombination reducing quantum well emission intensity. I found that during growth, the emergence of multiple facets with distinct incorporation rates leads to alloy composition and doping fluctuations, resulting in hexagonal structures interconnecting and forming a network. I discovered that by changing the alloy compositions present within the LED structure, screwtype threading dislocations can either enhance recombination (non-radiative) or reduced it. Finally I discuss lateral polarity heterostructures, which contain alternating stripes of Gaand N-polar material. Electron backscattered diffraction measurements confirmed the stripes were of the intended polarity but found in addition small inclusions of undesirable material in the Ga-polar regions. With cathodoluminescence spectroscopy I investigated the properties of the two distinct domains as well as their interface. I found that the Ga-polar region displayed far greater defect luminescence, consistent with some previous studies.

    @PhdThesis{Cameron-thesis,
    author = {Douglas Cameron},
    school = {University of Strathclyde},
    title = {Nanocharacterisation of {III}-nitride semiconductors},
    year = {2022},
    abstract = {Scanning electron microscopy allows for the investigation of materials down to the nanometre scale. In this thesis I combine cathodoluminescence spectroscopy with complementary techniques to investigate a range of III-nitride semiconductor heterostructures. I first describe AlGaN-coreshell nanorod UV-LED structures. Here, I conducted a wideranging study investigating the optical, electrical and structural properties of these structures. The electrical measurements were enabled by a nanoprobing system for in-situ contacting within the scanning electron microscope combined with bespoke focused ion beam deposition and milling to create contacts on individual nanorods. I-V measurements and electron beam induced currents confirmed the formation of a p-n junction and successful doping within the structure. In the same microscope, I measured CL to find the emission properties and found optically active quantum wells emitting near 300nm, with some variation in energy and intensity as well as clustering. In a transmission electron microscope, energy dispersive X-ray analysis confirmed the existence of clusters occurring at a-plane nanofacets. Low temperature and time resolved measurements measure the speed of the recombination in the quantum wells, showing a reduction in the quantum confined Stark effect within these structures. Using this same combination of techniques, I also examined planar UV-LED structures. I found that screw dislocations pinned the position of hexagonal hillocks during growth. In each of the penetrated layers, higher point defect densities and three-dimensional growth enclose the threading dislocations. These point defects can compensate electrical dopants and increase parasitic recombination reducing quantum well emission intensity. I found that during growth, the emergence of multiple facets with distinct incorporation rates leads to alloy composition and doping fluctuations, resulting in hexagonal structures interconnecting and forming a network. I discovered that by changing the alloy compositions present within the LED structure, screwtype threading dislocations can either enhance recombination (non-radiative) or reduced it. Finally I discuss lateral polarity heterostructures, which contain alternating stripes of Gaand N-polar material. Electron backscattered diffraction measurements confirmed the stripes were of the intended polarity but found in addition small inclusions of undesirable material in the Ga-polar regions. With cathodoluminescence spectroscopy I investigated the properties of the two distinct domains as well as their interface. I found that the Ga-polar region displayed far greater defect luminescence, consistent with some previous studies.},
    doi = {10.48730/yk6s-8t20},
    }

2021

  • A. Alasmari, “Characterisation of III-nitrides in the scanning electron microscope,” PhD Thesis, 2021.
    [BibTeX] [Abstract] [Library link]

    This thesis presents research on the characterisation of group III-nitrides using scanning electron microscope (SEM) techniques. In particular structural and morphological properties were investigated by electron channelling contrast imaging (ECCI) and electron backscatter diffraction (EBSD). ECCI reveals threading dislocations (TDs), sub-grains, atomic steps and step bunches in the material under study, while EBSD provides quantitative data on sub-grain misorientation. ECCI was also correlated with atomic force microscopy (AFM) to reveal and identify TDs in an epitaxially laterally overgrown GaN film, where the pattern for overgrowth comprised of stripes parallel to the GaN [1100] direction. ECCI revealed both vertically threading and inclined dislocations. Each dislocation was identified, using complementary information provided by AFM and ECCI, as either a-type or c/(a+c)-type dislocations. To extend ECCI to the study of insulating samples, ECCI micrographs were acquired in a variable pressure scanning electron microscope (VP-SEM) at pressures ranging from 0.3 mbar to 0.8 mbar using gaseous secondary electron detectors (GSEDs). ECCI micrographs are produced by detecting a gas-amplified secondary electron signal, indicating that the intensity of the detected secondary electrons is modulated as a result of diffraction of the incident electron beam. The influence of pressure, detector bias, sample tilt and working distance on the quality of ECCI micrographs were investigated. Aluminum nitride (AlN) samples grown on nano-patterned (nano-pillars) sapphire substrate (nPSS) were investigated. The miscut of the substrate and the size of the nano-pillars were shown to influence the surface morphology and TD density. The analysis of these AlN/nPSS samples showed a dependence of the density of step bunches with the density of TDs, the higher the density of step bunches, the lower the TD density. EBSD was used to measure the misorientations of sub-grains in the AlN/nPSS and the misorientation was observed to increase as the density of TDs increased.

    @PhdThesis{Aeshah-Alasmari_thesis,
    author = {Aeshah Alasmari},
    school = {University of Strathclyde},
    title = {Characterisation of {III}-nitrides in the scanning electron microscope},
    year = {2021},
    abstract = {This thesis presents research on the characterisation of group III-nitrides using scanning electron microscope (SEM) techniques. In particular structural and morphological properties were investigated by electron channelling contrast imaging (ECCI) and electron backscatter diffraction (EBSD). ECCI reveals threading dislocations (TDs), sub-grains, atomic steps and step bunches in the material under study, while EBSD provides quantitative data on sub-grain misorientation. ECCI was also correlated with atomic force microscopy (AFM) to reveal and identify TDs in an epitaxially laterally overgrown GaN film, where the pattern for overgrowth comprised of stripes parallel to the GaN [1100] direction. ECCI revealed both vertically threading and inclined dislocations. Each dislocation was identified, using complementary information provided by AFM and ECCI, as either a-type or c/(a+c)-type dislocations. To extend ECCI to the study of insulating samples, ECCI micrographs were acquired in a variable pressure scanning electron microscope (VP-SEM) at pressures ranging from 0.3 mbar to 0.8 mbar using gaseous secondary electron detectors (GSEDs). ECCI micrographs are produced by detecting a gas-amplified secondary electron signal, indicating that the intensity of the detected secondary electrons is modulated as a result of diffraction of the incident electron beam. The influence of pressure, detector bias, sample tilt and working distance on the quality of ECCI micrographs were investigated. Aluminum nitride (AlN) samples grown on nano-patterned (nano-pillars) sapphire substrate (nPSS) were investigated. The miscut of the substrate and the size of the nano-pillars were shown to influence the surface morphology and TD density. The analysis of these AlN/nPSS samples showed a dependence of the density of step bunches with the density of TDs, the higher the density of step bunches, the lower the TD density. EBSD was used to measure the misorientations of sub-grains in the AlN/nPSS and the misorientation was observed to increase as the density of TDs increased.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/2esacs/SUSTAX_T9p290969x},
    }

  • L. Spasevski, “Correlating X-ray microanalysis and cathodoluminescence data from III-nitride semiconductors,” PhD Thesis, 2021.
    [BibTeX] [Abstract] [Library link]

    Research in group III- nitride semiconductors has seen major developments during the last couple of decades. One of the materials that satisfy requirements for optoelectronic devices in the ultra-violet (UV) spectral range and high power, high frequency electronic devices is the AlGaN. Performance and reliability of these devices will strongly depend on the electronic properties of epitaxial layers which are critically affected by structural defects and unintentional and intentional doped impurities. This thesis presents research on III- nitride semiconductors, in particular AlGaN and GaN materials. It is focused on characterization of AlGaN materials and the effects of n- and p-type doping, AlN content, occurrence of defects and crystal orientation on its quality. Different electron microscopy techniques are used to investigate luminescence, composition and doping properties of semiconductor structures and their correlation with surface features. The main techniques used for the characterization consisted of cathodoluminescence spectroscopy (CL) for the probing of luminescence properties, secondary electron (SE) and backscattered electron (BSE) imaging for investigation of the sample morphology and wavelength dispersive X-ray (WDX) spectroscopy for compositional analysis. The type of growth method and choice of substrate have a great influence on the surface morphology and luminescence homogeneity of the AlGaN layer, with compositional inhomogeneity of the MBE samples confirmed only on sub μm level but having lower emission intensity compared to MOCVD samples. The thesis presents detailed steps of a procedure to quantify trace elements and investigates the associated challenges. The whole process of measurement optimization for Mg and Si dopants is described and final recipe on how to measure the concentration of major (alloy) and minor Si/Mg (dopant) elements is presented. A systematic study of polar and semipolar n-type doped AlGaN/AlN layers grown on sapphire by (MOCVD) with varied Si/group-III ratios in the gas phase was accomplished. The AlN incorporation was higher in the polar samples and the highest values of Si incorporations were observed for the polar samples with the highest Si/III ratios, while saturation of Si incorporation was seen for the semipolar samples at higher Si/III ratios. CL point spectra showed how changes in the relative intensity of the NBE peaks and impurity transitions depend strongly on the growth conditions and surface orientations. The semipolar samples showed better compositional homogeneity. A study was also performed on AlGaN:Mg samples to study the impurity transitions and luminescence properties of non LED epilayer samples grown on MOCVD AlN/sapphire templates and more complicated LED structures with different numbers of MBE-grown layers. MBE samples showed superior quality to other combinations of MBE and MOCVD structures, mainly due to problems associated with the transfer of sample between different reactors and the introduction of impurities that will form different defects within the material. Finally, some proposals for future work are presented.

    @PhdThesis{Lucia-Spasevski-thesis,
    author = {Spasevski, Lucia},
    school = {University of Strathclyde},
    title = {Correlating {X}-ray microanalysis and cathodoluminescence data from {III}-nitride semiconductors},
    year = {2021},
    abstract = {Research in group III- nitride semiconductors has seen major developments during the last couple of decades. One of the materials that satisfy requirements for optoelectronic devices in the ultra-violet (UV) spectral range and high power, high frequency electronic devices is the AlGaN. Performance and reliability of these devices will strongly depend on the electronic properties of epitaxial layers which are critically affected by structural defects and unintentional and intentional doped impurities. This thesis presents research on III- nitride semiconductors, in particular AlGaN and GaN materials. It is focused on characterization of AlGaN materials and the effects of n- and p-type doping, AlN content, occurrence of defects and crystal orientation on its quality. Different electron microscopy techniques are used to investigate luminescence, composition and doping properties of semiconductor structures and their correlation with surface features. The main techniques used for the characterization consisted of cathodoluminescence spectroscopy (CL) for the probing of luminescence properties, secondary electron (SE) and backscattered electron (BSE) imaging for investigation of the sample morphology and wavelength dispersive X-ray (WDX) spectroscopy for compositional analysis. The type of growth method and choice of substrate have a great influence on the surface morphology and luminescence homogeneity of the AlGaN layer, with compositional inhomogeneity of the MBE samples confirmed only on sub μm level but having lower emission intensity compared to MOCVD samples. The thesis presents detailed steps of a procedure to quantify trace elements and investigates the associated challenges. The whole process of measurement optimization for Mg and Si dopants is described and final recipe on how to measure the concentration of major (alloy) and minor Si/Mg (dopant) elements is presented. A systematic study of polar and semipolar n-type doped AlGaN/AlN layers grown on sapphire by (MOCVD) with varied Si/group-III ratios in the gas phase was accomplished. The AlN incorporation was higher in the polar samples and the highest values of Si incorporations were observed for the polar samples with the highest Si/III ratios, while saturation of Si incorporation was seen for the semipolar samples at higher Si/III ratios. CL point spectra showed how changes in the relative intensity of the NBE peaks and impurity transitions depend strongly on the growth conditions and surface orientations. The semipolar samples showed better compositional homogeneity. A study was also performed on AlGaN:Mg samples to study the impurity transitions and luminescence properties of non LED epilayer samples grown on MOCVD AlN/sapphire templates and more complicated LED structures with different numbers of MBE-grown layers. MBE samples showed superior quality to other combinations of MBE and MOCVD structures, mainly due to problems associated with the transfer of sample between different reactors and the introduction of impurities that will form different defects within the material. Finally, some proposals for future work are presented.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/2esacs/SUSTAX_Tc534fp459},
    }

2020

  • E. Pascal, “Dynamical models for novel diffraction techniques in SEM,” PhD Thesis, 2020.
    [BibTeX] [Abstract] [Library link]

    The scanning electron microscope is a powerful nanocharacterisation tool for a variety of materials including semiconductors and metals. Less known for its diffraction abilities than its transmission counterpart, scanning electron microscopy (SEM) can be used in a number of diffraction modalities to provide information on crystal imperfections at the nanoscale level. This comes with the added benefit of SEM requiring minimal sample preparation. Models for diffraction in the SEM are still being developed and improved, hence in this work I explore the physics and implementations of such models. I focus on the two main branches of SEM diffraction techniques:incident beam channelling, or diffraction in, powerful when it comes to resolving individual dislocations close to the surface; and back(/forward)scattering diffraction,or diffraction out, which provides a variety of information about grain distribution,orientation and strain. Both of these diffraction modalities involve the same physical processes, so it makes sense to use the same models, namely dynamical scattering in the column approximation. I use the two beam Bloch waves approach for electron channelling contrast imaging (ECCI) of threading dislocations (TDs) normal to the surface in wurtzite group-III nitride materials. I also introduce and use the notion of ECC-strain to study crystal features and to predict the behaviour of TDs contrast. For the electron back(/forward)scatter modality, I show the first application of the new energy-weighted dynamical scattering capabilities of EMsoft to study the novel transmission mode (TKD) of the SEM.

    @PhdThesis{Elena-Pascal-thesis,
    author = {Elena Pascal},
    school = {University of Strathclyde},
    title = {Dynamical models for novel diffraction techniques in {SEM}},
    year = {2020},
    abstract = {The scanning electron microscope is a powerful nanocharacterisation tool for a variety of materials including semiconductors and metals. Less known for its diffraction abilities than its transmission counterpart, scanning electron microscopy (SEM) can be used in a number of diffraction modalities to provide information on crystal imperfections at the nanoscale level. This comes with the added benefit of SEM requiring minimal sample preparation. Models for diffraction in the SEM are still being developed and improved, hence in this work I explore the physics and implementations of such models. I focus on the two main branches of SEM diffraction techniques:incident beam channelling, or diffraction in, powerful when it comes to resolving individual dislocations close to the surface; and back(/forward)scattering diffraction,or diffraction out, which provides a variety of information about grain distribution,orientation and strain. Both of these diffraction modalities involve the same physical processes, so it makes sense to use the same models, namely dynamical scattering in the column approximation. I use the two beam Bloch waves approach for electron channelling contrast imaging (ECCI) of threading dislocations (TDs) normal to the surface in wurtzite group-III nitride materials. I also introduce and use the notion of ECC-strain to study crystal features and to predict the behaviour of TDs contrast. For the electron back(/forward)scatter modality, I show the first application of the new energy-weighted dynamical scattering capabilities of EMsoft to study the novel transmission mode (TKD) of the SEM.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/k7ss9a/SUSTAX_Tj3860696g},
    }

  • J. Denholm, “Freezing in kinetic ferromagnets,” PhD Thesis, 2020.
    [BibTeX] [Abstract] [Library link]

    In this thesis I present studies of the non-equilibrium dynamics of Ising and Potts ferromagnets in two spatial dimensions. While the conception of these models was originally motivated by a desire to understand the phenomena surrounding equilibrium phase transitions, they have found considerable use in non-equilibrium Statistical Physics. Despite being well studied, even from a non-equilibrium viewpoint, a number of surprisingly basic and fundamental gaps in our understanding of Ising and Potts models have come to light in recent decades. For example, the tacit assumption that Ising ferromagnets should always reach their ground state at zero-temperature has been found incorrect, and unexpected features of the associated relaxation process have come to light. With the Potts model the situation is stranger still: the late time final states that persist in two dimensions are considerably richer than those of the Ising model, and the relaxation times are complex and not yet well understood. In this thesis I examine basic aspects of zero-temperature coarsening in two dimensional Ising and Potts ferromagnets. I explore the timescales associated with zero-temperature freezing in the Ising model, and uncover the existence of an overlooked relaxation timescale. I then investigate the final states of the zero-temperature Potts model on the triangular lattice, which prior to this work had not been examined. I continue my studies of the Potts model to situations of increased ground state degeneracy and extended local interaction rules.

    @PhdThesis{James-Denholm_thesis,
    author = {James Denholm},
    school = {Univesrity of Strathclyde},
    title = {Freezing in kinetic ferromagnets},
    year = {2020},
    abstract = {In this thesis I present studies of the non-equilibrium dynamics of Ising and Potts ferromagnets in two spatial dimensions. While the conception of these models was originally motivated by a desire to understand the phenomena surrounding equilibrium phase transitions, they have found considerable use in non-equilibrium Statistical Physics. Despite being well studied, even from a non-equilibrium viewpoint, a number of surprisingly basic and fundamental gaps in our understanding of Ising and Potts models have come to light in recent decades. For example, the tacit assumption that Ising ferromagnets should always reach their ground state at zero-temperature has been found incorrect, and unexpected features of the associated relaxation process have come to light. With the Potts model the situation is stranger still: the late time final states that persist in two dimensions are considerably richer than those of the Ising model, and the relaxation times are complex and not yet well understood. In this thesis I examine basic aspects of zero-temperature coarsening in two dimensional Ising and Potts ferromagnets. I explore the timescales associated with zero-temperature freezing in the Ising model, and uncover the existence of an overlooked relaxation timescale. I then investigate the final states of the zero-temperature Potts model on the triangular lattice, which prior to this work had not been examined. I continue my studies of the Potts model to situations of increased ground state degeneracy and extended local interaction rules.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/k7ss9a/SUSTAX_T7d278t113},
    }

  • M. B. Jablon, “Understanding crystallographic defects in hard materials: a study in ECCI and EBSD,” PhD Thesis, 2020.
    [BibTeX] [Abstract] [Library link]

    This thesis investigates the crystallographic defects inherent in two hard materials, tungsten carbide embedded in a cobalt matrix (WC-Co) and diamonds. The scanning electron microscopy (SEM) techniques of electron channelling contrast imaging (ECCI), cathodoluminescence (CL) imaging and electron backscatter diffraction (EBSD) allowed for the investigation of the crystallographic defects. Initial work on WC identified the dislocations and subgrain structures in grains of known crystallographic orientation. By combining ECCI and EBSD, new insights into the microstructure of WC grains were revealed. The advantage of using these techniques over the more common TEM based techniques is that ECCI and EBSD allow for analysis over a large field of view, in this case 75 μm ×75 μm. These large scale analyses reveal the microstructure and misorientation and dislocations throughout an entire WC grain. ECCI exposed dislocations while EBSD data revealed small changes in crystallographic orientation resulting from dislocations and subgrains, thereby corroborating the ECCI data. Further investigation of dislocations in WC categorized the type of dislocations found in grains of known crystallographic orientation using ECCI and EBSD. Dislocations and their Burgers vectors are described. The type of dislocations found in a particular grain are discussed in relation to their orientation with respect to neighbouring grains. Using ECCI and EBSD to categorize dislocations in the basal and prismatic planes helps explain how WC grains deform plastically. Plastic deformation is studied by indenting WC grains and studying the damage with ECCI and EBSD analysis. Misorientation surrounding the indent is investigated by EBSD for two different indenter tips. Using ECCI to reveal slip bands around indents in grains exposes different defects and deformation depending on the crystallographic orientation of grains. This novel ECCI work agrees with previous TEM data on slip systems in indented WC grains. Using ECCI presents an advantage over TEM in this case in that damage in grains can be imaged over a wider field of view with simpler sample preparation. The characterization of twinned gem-quality diamonds, also known as macles, using ECCI, EBSD and CL reveals that twinned diamonds are all composed of two sibling twins with around 180° between the siblings. CL is helpful in understanding the initial twin nucleation and distinct periods of crystallization in the diamond’s history while EBSD is used to corroborate CL. This study shows that the twinning plane is not linear, but jagged, corroborating previous research indicating that these are not true contact twins, rather they contain some intergrowths. EBSD confirms that these are undeniable twins as the plane separating the two siblings is a twin Coincident Site Lattices (CSL) grain boundary. CL and EBSD together are powerful tools for characterizing the twinning plane and growth mechanism in diamonds and this work describes the novel use of these techniques to understand diamond crystallization.

    @PhdThesis{Matat-Jablon_thesis,
    author = {B. Matat Jablon},
    school = {University of Strathclyde},
    title = {Understanding crystallographic defects in hard materials: a study in {ECCI} and {EBSD}},
    year = {2020},
    abstract = {This thesis investigates the crystallographic defects inherent in two hard materials, tungsten carbide embedded in a cobalt matrix (WC-Co) and diamonds. The scanning electron microscopy (SEM) techniques of electron channelling contrast imaging (ECCI), cathodoluminescence (CL) imaging and electron backscatter diffraction (EBSD) allowed for the investigation of the crystallographic defects. Initial work on WC identified the dislocations and subgrain structures in grains of known crystallographic orientation. By combining ECCI and EBSD, new insights into the microstructure of WC grains were revealed. The advantage of using these techniques over the more common TEM based techniques is that ECCI and EBSD allow for analysis over a large field of view, in this case 75 μm ×75 μm. These large scale analyses reveal the microstructure and misorientation and dislocations throughout an entire WC grain. ECCI exposed dislocations while EBSD data revealed small changes in crystallographic orientation resulting from dislocations and subgrains, thereby corroborating the ECCI data. Further investigation of dislocations in WC categorized the type of dislocations found in grains of known crystallographic orientation using ECCI and EBSD. Dislocations and their Burgers vectors are described. The type of dislocations found in a particular grain are discussed in relation to their orientation with respect to neighbouring grains. Using ECCI and EBSD to categorize dislocations in the basal and prismatic planes helps explain how WC grains deform plastically. Plastic deformation is studied by indenting WC grains and studying the damage with ECCI and EBSD analysis. Misorientation surrounding the indent is investigated by EBSD for two different indenter tips. Using ECCI to reveal slip bands around indents in grains exposes different defects and deformation depending on the crystallographic orientation of grains. This novel ECCI work agrees with previous TEM data on slip systems in indented WC grains. Using ECCI presents an advantage over TEM in this case in that damage in grains can be imaged over a wider field of view with simpler sample preparation. The characterization of twinned gem-quality diamonds, also known as macles, using ECCI, EBSD and CL reveals that twinned diamonds are all composed of two sibling twins with around 180° between the siblings. CL is helpful in understanding the initial twin nucleation and distinct periods of crystallization in the diamond’s history while EBSD is used to corroborate CL. This study shows that the twinning plane is not linear, but jagged, corroborating previous research indicating that these are not true contact twins, rather they contain some intergrowths. EBSD confirms that these are undeniable twins as the plane separating the two siblings is a twin Coincident Site Lattices (CSL) grain boundary. CL and EBSD together are powerful tools for characterizing the twinning plane and growth mechanism in diamonds and this work describes the novel use of these techniques to understand diamond crystallization.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/2esacs/SUALMA21149718300002996},
    }

2018

  • S. Vespucci, “Development of an energy filtering direct electron detector for diffraction studies in the SEM,” PhD Thesis, 2018.
    [BibTeX] [Abstract] [Library link]

    This thesis describes the application of an energy filtering digital direct electron detector for diffraction studies of materials in the field emission scanning electron microscope (SEM). The main aim was the development of the digital complementary metal-oxide-semiconductor hybrid pixel detector, “Timepix” for electron backscatter diffraction (EBSD), a technique which allows the acquisition of precise crystallographic information from the surface of a sample, such as crystallographic orientation, phase and strain. EBSD results from nitride semiconductor, silicon and diamond thin films and tungsten-carbide cobalt samples are presented and used to illustrate the advantages of acquiring EBSD patterns with the Timepix detector, in particular to demonstrate the improvement in the contrast and increase in the detail contained in the EBSD patterns as consequence of the energy filtering. Alongside EBSD, new applications were developed such as re ection high energy electron diffraction (RHEED) in the SEM. RHEED is a very surface sensitive technique which in principle could allow the study of ultrathin samples where conventional SEM based methods are limited. The combination of RHEED and Kikuchi diffraction, led furthermore to the development of surface wave resonance electron channelling contrast imaging (SWRECCI), which allows crystalline defects such as surface steps, grain boundaries, dislocations and stacking faults to be imaged with a high level of surface sensitivity, extending furthermore the application of ECCI to non-continuous surfaces. This is obtained by selecting experimental geometries which stimulate the surface wave resonance at the specimen surface. Transmission diffraction in the SEM was also explored, resulting in the acquisition of transmission diffraction patterns and in the generation of images of the sample obtained under experimental conditions analogous to scanning transmission electron microscopy. This allowed for example, bright and dark field images of the specimen to be obtained. The resulting images exhibited crystalline contrast not often observed in the SEM. The Timepix sensor is constructed from a piece of single crystal silicon. Diffraction effects within this single crystal were found to result in the Timepix detector response exhibiting an underlying diffraction pattern; that is a detector diffraction pattern (DDP). The DDP provides a watermark from which the location of the camera relative to the position of the electron beam on the sample may be precisely and accurately determined. This opens up new opportunities for improved mapping of the strain distribution in materials for example. The development of all the novel techniques summarized above opens up new horizons which need to be explored.

    @PhdThesis{Stefano-Vespucci_thesis,
    author = {Stefafano Vespucci},
    school = {University of Strathclyde},
    title = {Development of an energy filtering direct electron detector for diffraction studies in the {SEM}},
    year = {2018},
    abstract = {This thesis describes the application of an energy filtering digital direct electron detector for diffraction studies of materials in the field emission scanning electron microscope (SEM). The main aim was the development of the digital complementary metal-oxide-semiconductor hybrid pixel detector, ``Timepix'' for electron backscatter diffraction (EBSD), a technique which allows the acquisition of precise crystallographic information from the surface of a sample, such as crystallographic orientation, phase and strain. EBSD results from nitride semiconductor, silicon and diamond thin films and tungsten-carbide cobalt samples are presented and used to illustrate the advantages of acquiring EBSD patterns with the Timepix detector, in particular to demonstrate the improvement in the contrast and increase in the detail contained in the EBSD patterns as consequence of the energy filtering.
    Alongside EBSD, new applications were developed such as re
    ection high energy electron diffraction (RHEED) in the SEM. RHEED is a very surface sensitive technique which in principle could allow the study of ultrathin samples where conventional SEM based methods are limited.
    The combination of RHEED and Kikuchi diffraction, led furthermore to the development of surface wave resonance electron channelling contrast imaging (SWRECCI), which allows crystalline defects such as surface steps, grain boundaries, dislocations and stacking faults to be imaged with a high level of surface sensitivity, extending furthermore the application of ECCI to non-continuous surfaces. This is obtained by selecting experimental geometries which stimulate the surface wave resonance at the specimen surface.
    Transmission diffraction in the SEM was also explored, resulting in the acquisition of transmission diffraction patterns and in the generation of images of the sample obtained under experimental conditions analogous to scanning transmission electron microscopy. This allowed for example, bright and dark field images of the specimen to be obtained. The resulting images exhibited crystalline contrast not often observed in the SEM.
    The Timepix sensor is constructed from a piece of single crystal silicon. Diffraction effects within this single crystal were found to result in the Timepix detector response exhibiting an underlying diffraction pattern; that is a detector diffraction pattern (DDP). The DDP provides a watermark from which the location of the camera relative to the position of the electron beam on the sample may be precisely and accurately determined. This opens up new opportunities for improved mapping of the strain distribution in materials for example.
    The development of all the novel techniques summarized above opens up new horizons which need to be explored.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/k7ss9a/SUALMA21106168740002996},
    }

2017

  • E. Skidchenko, “Photoluminescence study of Cu(In,Ga)Se₂ and Cu₂ZnSnSe₄ used for thin film solar cells,” Master Thesis, 2017.
    [BibTeX] [Abstract] [Library link]

    In this study we were trying to answer the question why the best laboratory size CuIn₁₋ₓGaₓSe₂-based (CIGSe) solar cells have been achieved at about 30% Ga, whereas the optimal band gap requires much higher Ga content (around 50%). Also for the first time ever we examined the effects of 4 keV Ar⁺ irradiation on optical properties of Cu₂ZnSnSe₄ (CZTSe) thin films. Photoluminescence (PL) measurements were performed on samples of CIGSe single crystals with x=0, 0.05, 0.10, 0.25, 0.50, 0.75, 1 and on samples of CZTSe thin films before and after Ar⁺ irradiation. The impact of variations in excitation intensity and temperature on the nature of the PL bands was investigated. For most CIGSe samples band-tail (BT) recombination coupled with band-band (BB) transition were identified as being responsible for the obtained PL peaks. The calculated band tails are deeper in CuGaSe₂ than in CuInSe₂, explaining why an increase in Ga content x > 0.3 does not improve the efficiency of solar cell. Also band-impurity (BI) transition and donor acceptor pair (DAP) recombination were discovered. For the first time ever excitons in the CIGSe samples with 5% and 10% Ga content were clearly observed and identified. In the CZTSe samples BI recombination was established as the main source of PL. Ar⁺ irradiation of CZTSe produced significant changes in the material: reduced PL intensity and creation of tail defect complexes supported by an increase in the average band tail depth γ from 25 to 30 meV as well as increase in the activation energy from 75 to 88 meV. The composition of the CIGSe and CZTSe samples was investigated using energy dispersive x-ray (EDX) analysis and wavelength dispersive x-ray (WDX) technique, respectively. According to the microscopy results, calculated cation ratios proved the nature of established recombinations.

    @MastersThesis{Ekaterina-Skidchenko-thesis,
    author = {Ekaterina Skidchenko},
    school = {University of Strathclyde},
    title = {Photoluminescence study of {Cu(In,Ga)Se₂} and {Cu₂ZnSnSe₄} used for thin film solar cells},
    year = {2017},
    abstract = {In this study we were trying to answer the question why the best laboratory size CuIn₁₋ₓGaₓSe₂-based (CIGSe) solar cells have been achieved at about 30% Ga, whereas the optimal band gap requires much higher Ga content (around 50%). Also for the first time ever we examined the effects of 4 keV Ar⁺ irradiation on optical properties of Cu₂ZnSnSe₄ (CZTSe) thin films. Photoluminescence (PL) measurements were performed on samples of CIGSe single crystals with x=0, 0.05, 0.10, 0.25, 0.50, 0.75, 1 and on samples of CZTSe thin films before and after Ar⁺ irradiation.
    The impact of variations in excitation intensity and temperature on the nature of the PL bands was investigated. For most CIGSe samples band-tail (BT) recombination coupled with band-band (BB) transition were identified as being responsible for the obtained PL peaks. The calculated band tails are deeper in CuGaSe₂ than in CuInSe₂, explaining why an increase in Ga content x > 0.3 does not improve the efficiency of solar cell.
    Also band-impurity (BI) transition and donor acceptor pair (DAP) recombination were discovered. For the first time ever excitons in the CIGSe samples with 5% and 10% Ga content were clearly observed and identified. In the CZTSe samples BI recombination was established as the main source of PL. Ar⁺ irradiation of CZTSe produced significant changes in the material: reduced PL intensity and creation of tail defect complexes supported by an increase in the average band tail depth γ from 25 to 30 meV as well as increase in the activation energy from 75 to 88 meV.
    The composition of the CIGSe and CZTSe samples was investigated using energy dispersive x-ray (EDX) analysis and wavelength dispersive x-ray (WDX) technique, respectively. According to the microscopy results, calculated cation ratios proved the nature of established recombinations.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tdv13zt305},
    }

2016

  • M. J. Wallace, “Optoelectronic study of InGaN/GaN LEDs,” PhD Thesis, 2016.
    [BibTeX] [Abstract] [Library link]

    The quality of light emitting diodes (LEDs) has increased to a point where solid state lighting is becoming fairly common. Despite this, greater understanding of the effect of the device structure and the electric fields within them is helpful to continue improving device efficiency and uniformity and in reducing costs. In this thesis the optical and electronic properties of InGaN/GaN LEDs have been studied with a combination of luminescence spectroscopy, microscopy, conductivity mapping and efficiency measurements. A study was made of the effects of the various electric fields, and the interplay between them, on LED luminescence and conductivity. Cathodoluminescence (CL) mapping shows die to die variation across large wafers revealing the powerful effects of a induced electric field on spectral intensity/position/width, in uncontacted devices. Micron scale spots in the LED material, lower in luminescence intensity and which trap charge, were revealed by CL/EBIC mapping with the origin attributed to cluster point defects in the active region. Depth resolved CL and CL under bias reveal the extent of asymmetry in carrier transport in the p/n type GaN around the active region. LEDs grown with different active region temperature profiles were studied. Devices exposed to high temperature after quantum well growth (2T) were found to have a uniform spatial luminescence and a peak efficiency that is higher and occurs at a lower current density (0.1 W/A @ 1 Acm⁻²). By contrast those with a low temperature cap (Q2T) exhibit dark spots in the luminescence, and a lower peak efficiency at a higher current density (0.04 W/A @ 10 Acm⁻²). The effect of improvement in LED design and material quality on the device efficiency, uniformity and spectral characteristics was studied. The addition of an Al₀.₂₃Ga₀.₇₇N electron blocking layer (EBL) was found to reduce the size and strength of the dark spots by about a factor of 2, while an additional In₀.₀₅Ga₀.₉₅N underlayer (UL) removed the dark spots entirely and shifted the luminescence peak by around 100 meV. The effect on the electroluminescence efficiency of the reduction in template dislocation density was found to depend strongly on the drive current density, with defect non-radiative recombination more important at low currents. Overall device efficiency was shown to be improved with an EBL and UL. The most efficient devices were those with the 2T type growth but the relative improvements are larger in LEDs grown with the Q2T method. Together, the results present a number of factors limiting the performance of current LEDs and suggest potential routes for improvement and optimisation.

    @PhdThesis{Michael-Wallace-thesis,
    author = {Wallace, Michael J.},
    school = {University of Strathlcyde},
    title = {Optoelectronic study of {InGaN/GaN LED}s},
    year = {2016},
    abstract = {The quality of light emitting diodes (LEDs) has increased to a point where solid state lighting is becoming fairly common. Despite this, greater understanding of the effect of the device structure and the electric fields within them is helpful to continue improving device efficiency and uniformity and in reducing costs. In this thesis the optical and electronic properties of InGaN/GaN LEDs have been studied with a combination of luminescence spectroscopy, microscopy, conductivity mapping and efficiency measurements. A study was made of the effects of the various electric fields, and the interplay between them, on LED luminescence and conductivity. Cathodoluminescence (CL) mapping shows die to die variation across large wafers revealing the powerful effects of a induced electric field on spectral intensity/position/width, in uncontacted devices. Micron scale spots in the LED material, lower in luminescence intensity and which trap charge, were revealed by CL/EBIC mapping with the origin attributed to cluster point defects in the active region. Depth resolved CL and CL under bias reveal the extent of asymmetry in carrier transport in the p/n type GaN around the active region. LEDs grown with different active region temperature profiles were studied. Devices exposed to high temperature after quantum well growth (2T) were found to have a uniform spatial luminescence and a peak efficiency that is higher and occurs at a lower current density (0.1 W/A @ 1 Acm⁻²). By contrast those with a low temperature cap (Q2T) exhibit dark spots in the luminescence, and a lower peak efficiency at a higher current density (0.04 W/A @ 10 Acm⁻²). The effect of improvement in LED design and material quality on the device efficiency, uniformity and spectral characteristics was studied. The addition of an Al₀.₂₃Ga₀.₇₇N electron blocking layer (EBL) was found to reduce the size and strength of the dark spots by about a factor of 2, while an additional In₀.₀₅Ga₀.₉₅N underlayer (UL) removed the dark spots entirely and shifted the luminescence peak by around 100 meV. The effect on the electroluminescence efficiency of the reduction in template dislocation density was found to depend strongly on the drive current density, with defect non-radiative recombination more important at low currents. Overall device efficiency was shown to be improved with an EBL and UL. The most efficient devices were those with the 2T type growth but the relative improvements are larger in LEDs grown with the Q2T method. Together, the results present a number of factors limiting the performance of current LEDs and suggest potential routes for improvement and optimisation.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tnv935295j},
    }

  • G. Kusch, “Characterization of low conductivity wide band gap semiconductors,” PhD Thesis, 2016.
    [BibTeX] [Abstract] [Library link]

    This thesis covers research on low electric conductivity wide band gap semiconductors of the group-III nitride material system. The work presented focussed on using multi-mode scanning electron microscope (SEM) techniques to investigate the luminescence properties and their correlation with surface effects, doping concentration and structure of semiconductor structures. The measurement techniques combined cathodoluminescence (CL) for the characterization of luminescence properties, secondary electron (SE) imaging for imaging of the morphology and wavelength dispersive X-ray (WDX) spectroscopy for compositional analysis. The high spatial resolution of CL and SE-imaging allowed for the investigation of nanometer sized features, whilst environmental SEM allowed the characterisation of low conductivity samples. The investigated AₓGa₁₋ₓN samples showed a strong dependence on the miscut of the substrate, which was proven to influence the surface morphology and the compositional homogeneity. Studying the influence of the AlₓGa₁₋ₓN sample thickness displayed a reduced strain in the samples with increasing thickness as well as an increasing crystalline quality. The analysis of AlxGa₁₋xN:Si samples showed the incorporation properties of Si in AlₓGa₁₋ₓN, the correlation between defect luminescence, Si concentration and resistivity as well as the influence of threading dislocations on the luminescence properties and incorporation of point defects. The characterization of UV-LED structures demonstrated that a change in the band structure is one of the main reasons for a decreasing output power in AlₓGa₁₋ₓN based UV-LEDs. In addition the dependence of the luminescence properties and crystalline quality of InₓAl₁₋ₓN based UV-LEDs on various growth parameters (e.g. growth temperature, quantum well thickness) was investigated. The study of nanorods revealed the influence of the template on the compositional homogeneity and luminescence of InₓAl₁₋ₓN nanorod LEDs. Furthermore,the influence of optical modes in these structures was studied and found to provide an additional engineering parameter for the design of nanorod LEDs.

    @PhdThesis{Gunnar-Kusch-thesis,
    author = {Gunnar Kusch},
    school = {University of Strathclyde},
    title = {Characterization of low conductivity wide band gap semiconductors},
    year = {2016},
    abstract = {This thesis covers research on low electric conductivity wide band gap semiconductors of the group-III nitride material system. The work presented focussed on using multi-mode scanning electron microscope (SEM) techniques to investigate the luminescence properties and their correlation with surface effects, doping concentration and structure of semiconductor structures. The measurement techniques combined cathodoluminescence (CL) for the characterization of luminescence properties, secondary electron (SE) imaging for imaging of the morphology and wavelength dispersive X-ray (WDX) spectroscopy for compositional analysis. The high spatial resolution of CL and SE-imaging allowed for the investigation of nanometer sized features, whilst environmental SEM allowed the characterisation of low conductivity samples. The investigated AₓGa₁₋ₓN samples showed a strong dependence on the miscut of the substrate, which was proven to influence the surface morphology and the compositional homogeneity. Studying the influence of the AlₓGa₁₋ₓN sample thickness displayed a reduced strain in the samples with increasing thickness as well as an increasing crystalline quality. The analysis of AlxGa₁₋xN:Si samples showed the incorporation properties of Si in AlₓGa₁₋ₓN, the correlation between defect luminescence, Si concentration and resistivity as well as the influence of threading dislocations on the luminescence properties and incorporation of point defects. The characterization of UV-LED structures demonstrated that a change in the band structure is one of the main reasons for a decreasing output power in AlₓGa₁₋ₓN based UV-LEDs. In addition the dependence of the luminescence properties and crystalline quality of InₓAl₁₋ₓN based UV-LEDs on various growth parameters (e.g. growth temperature, quantum well thickness) was investigated. The study of nanorods revealed the influence of the template on the compositional homogeneity and luminescence of InₓAl₁₋ₓN nanorod LEDs. Furthermore,the influence of optical modes in these structures was studied and found to provide an additional engineering parameter for the design of nanorod LEDs.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tbn999684n},
    }

2015

  • E. Taylor-Shaw, “Composition and luminescence studies of InGaN and InAlN alloys,” PhD Thesis, 2015.
    [BibTeX] [Abstract] [Library link]

    III-nitride semiconductors are the leading material for use in solid state lighting (SSL), with highly efficient blue and white nitride-based light emitting diodes (LEDs) commercially available. However challenges still remain to improve their efficiency. The work in this thesis focuses on the optical and compositional characterisation of InGaN and InAlN alloys, which are widely used as active regions in such light emitters. Composition and luminescence properties of InGaN epilayers with varying growth temperature and hydrogen flow rates are investigated. The measurements revealed that the samples grown with small amounts of hydrogen improved in surface quality, compositional and luminescence homogeneity when compared with samples grown at equivalent temperature. The additional hydrogen did reduce the InN fraction slightly. Investigations of the optical, compositional and structural properties of Ga auto-incorporated InAl(Ga)N epilayers are made. Composition measurements revealed 1 2-28 % of Ga incorporated. The growth parameters and resultant Ga indicated the likely cause is residual Ga coming from the reactor walls and delivery pipes, as by increasing the total flow rate from 8000 sccm to 24000 sccm was seen to suppress the GaN from 28 to 12 %. A broad spectral emission peak was seen, whose energy varied with InN content and not GaN. A large set of InAlN epilayers grown on AlN buffers are studied. Composition measurements revealed a wide range of InN contents from 0.1 % to 25.6 %. The analysis revealed no presence of Ga within the samples. Optical measurements produced broad InAlN luminescence spectra which varied with InN content. The peak energy was found to be 3.46-3.93 eV for InN compositions of 0.7-6.6 %. Analysis suggests this is not bandedge emission due to the low peak energy and very wide FWHMs. Finally, a home built PL mapping system is demonstrated, along with the design and operation challenges. Utilising this mapping system, investigations of InGa N/GaN MQW LED samples grown under different barrier growth methods are made.

    @PhdThesis{Elaine-Taylor-Shaw-thesis,
    author = {Elaine Taylor-Shaw},
    school = {University of Strathclyde},
    title = {Composition and luminescence studies of {InGaN} and {InAlN} alloys},
    year = {2015},
    abstract = {III-nitride semiconductors are the leading material for use in solid state lighting (SSL), with highly efficient blue and white nitride-based light emitting diodes (LEDs) commercially available. However challenges still remain to improve their efficiency. The work in this thesis focuses on the optical and compositional characterisation of InGaN and InAlN alloys, which are widely used as active regions in such light emitters. Composition and luminescence properties of InGaN epilayers with varying growth temperature and hydrogen flow rates are investigated. The measurements revealed that the samples grown with small amounts of hydrogen improved in surface quality, compositional and luminescence homogeneity when compared with samples grown at equivalent temperature. The additional hydrogen did reduce the InN fraction slightly. Investigations of the optical, compositional and structural properties of Ga auto-incorporated InAl(Ga)N epilayers are made. Composition measurements revealed 1 2-28 % of Ga incorporated. The growth parameters and resultant Ga indicated the likely cause is residual Ga coming from the reactor walls and delivery pipes, as by increasing the total flow rate from 8000 sccm to 24000 sccm was seen to suppress the GaN from 28 to 12 %. A broad spectral emission peak was seen, whose energy varied with InN content and not GaN. A large set of InAlN epilayers grown on AlN buffers are studied. Composition measurements revealed a wide range of InN contents from 0.1 % to 25.6 %. The analysis revealed no presence of Ga within the samples. Optical measurements produced broad InAlN luminescence spectra which varied with InN content. The peak energy was found to be 3.46-3.93 eV for InN compositions of 0.7-6.6 %. Analysis suggests this is not bandedge emission due to the low peak energy and very wide FWHMs. Finally, a home built PL mapping system is demonstrated, along with the design and operation challenges. Utilising this mapping system, investigations of InGa N/GaN MQW LED samples grown under different barrier growth methods are made.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_T9019s256w},
    }

2013

  • S. Kraeusel, “Native defects in the group III nitrides,” PhD Thesis, 2013.
    [BibTeX] [Abstract] [Library link]

    The promise of the broad range of direct band gaps of the (Al,Ga,In)N system is limited by the crystal quality of current material. As grown defect densities of InN, when compared with the more mature GaN, are extremely high and InN is strongly influenced by these defects. This is particularly important due to the unusual position of the charge neutrality level of InN, leading to both the well known surface charge accumulation and difficulties in p-type doping. While impurities and native defects clearly impact on the bulk carrier density in InN, the effects of threading dislocations on the electrical properties are still in dispute. Issues such as whether the dislocation line is charged or contains dangling bonds remain open. In this work an empirical Stillinger-Weber inter-atomic potential method is employed in a systematic global search for possible dislocation core reconstructions for screw and edge dislocations in GaN. The global optimisation of the dislocation cores is performed for a wide variety of core stoichiometries ranging from Ga rich to N rich. The most promising optimised core configurations are subsequently investigated using density functional theory for GaN and InN, in order to discuss relative stability under a wide range of growth conditions and their influence on the electronic properties of the bulk material.

    @PhdThesis{Simon-Kraeusel-thesis,
    author = {Simon Kraeusel},
    school = {University of Strathclyde},
    title = {Native defects in the group {III} nitrides},
    year = {2013},
    abstract = {The promise of the broad range of direct band gaps of the (Al,Ga,In)N system is limited by the crystal quality of current material. As grown defect densities of InN, when compared with the more mature GaN, are extremely high and InN is strongly influenced by these defects. This is particularly important due to the unusual position of the charge neutrality level of InN, leading to both the well known surface charge accumulation and difficulties in p-type doping. While impurities and native defects clearly impact on the bulk carrier density in InN, the effects of threading dislocations on the electrical properties are still in dispute. Issues such as whether the dislocation line is charged or contains dangling bonds remain open. In this work an empirical Stillinger-Weber inter-atomic potential method is employed in a systematic global search for possible dislocation core reconstructions for screw and edge dislocations in GaN. The global optimisation of the dislocation cores is performed for a wide variety of core stoichiometries ranging from Ga rich to N rich. The most promising optimised core configurations are subsequently investigated using density functional theory for GaN and InN, in order to discuss relative stability under a wide range of growth conditions and their influence on the electronic properties of the bulk material.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tzw12z534t},
    }

  • J. Bruckbauer, “Luminescence study of III-nitride semiconductor nanostructures and LEDs,” PhD Thesis, 2013.
    [BibTeX] [Abstract] [Library link]

    In this work, cathodoluminescence (CL) hyperspectral imaging, photoluminescence (PL) and electroluminescence are used to study the optical properties of III-nitride semiconductor materials. III-nitride semiconductors have successfully opened up the solid-state lighting market. Light-emitting diodes (LEDs) fabricated using III-nitrides, however, still suffer from numerous deficiencies such as high defect densities, efficiency droop and the ‘green gap’. In order to investigate the type and properties of the defects, CL and electron channelling contrast imaging (ECCI) were performed on the same micron-scale area of a GaN thin film. A one-to-one correlation between isolated dark spots in CL and threading dislocations (TDs) in ECCI showed that TDs of pure edge character and TDs with a screw component act as non-radiative recombination centres. Secondary electron imaging of planar InGaN/GaN multiple quantum well (MQW) structures identified trench defects of varying width. CL imaging re vealed a strong redshift (90 meV) and intensity increase for trench defects with wide trenches compared with the defect-free surrounding area. Narrower trench defects showed a small redshift (10 meV) and a slight reduction in intensity. The optical properties of nanorods fabricated from planar InGaN/GaN MQW structures were investigated using PL and CL. PL spectroscopy identified reduced strain within the MQW stack in the nanorods compared with the planar structure. CL imaging of single nanorods revealed a redshift of 18 meV of the MQW emission along the nanorod axis and provided an estimate of 55 nm for the carrier diffusion length. Colour conversion using novel organic compounds as energy down-converters was studied. The first molecules absorbed in the ultra-violet and emitted in the yellow spectral region. Further modification of the organic compound shifted the absorption into the blue and white light generation was investigated by coating blue-emitting nanorods and blue LED determination of the colour rendering index and colour temperature showed “warm white” light emission with values of 70 and 3220 K, respectively.

    @PhdThesis{Jochen-Bruckbauer-thesis,
    author = {Jochen Bruckbauer},
    school = {University of Strathclyde},
    title = {Luminescence study of {III}-nitride semiconductor nanostructures and {LED}s},
    year = {2013},
    abstract = {In this work, cathodoluminescence (CL) hyperspectral imaging, photoluminescence (PL) and electroluminescence are used to study the optical properties of III-nitride semiconductor materials. III-nitride semiconductors have successfully opened up the solid-state lighting market. Light-emitting diodes (LEDs) fabricated using III-nitrides, however, still suffer from numerous deficiencies such as high defect densities, efficiency droop and the 'green gap'. In order to investigate the type and properties of the defects, CL and electron channelling contrast imaging (ECCI) were performed on the same micron-scale area of a GaN thin film. A one-to-one correlation between isolated dark spots in CL and threading dislocations (TDs) in ECCI showed that TDs of pure edge character and TDs with a screw component act as non-radiative recombination centres. Secondary electron imaging of planar InGaN/GaN multiple quantum well (MQW) structures identified trench defects of varying width. CL imaging re vealed a strong redshift (90 meV) and intensity increase for trench defects with wide trenches compared with the defect-free surrounding area. Narrower trench defects showed a small redshift (10 meV) and a slight reduction in intensity. The optical properties of nanorods fabricated from planar InGaN/GaN MQW structures were investigated using PL and CL. PL spectroscopy identified reduced strain within the MQW stack in the nanorods compared with the planar structure. CL imaging of single nanorods revealed a redshift of 18 meV of the MQW emission along the nanorod axis and provided an estimate of 55 nm for the carrier diffusion length. Colour conversion using novel organic compounds as energy down-converters was studied. The first molecules absorbed in the ultra-violet and emitted in the yellow spectral region. Further modification of the organic compound shifted the absorption into the blue and white light generation was investigated by coating blue-emitting nanorods and blue LED determination of the colour rendering index and colour temperature showed "warm white" light emission with values of 70 and 3220 K, respectively.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tq237hs021},
    }

  • F. Elfituri, “Computational investigation of InGaN alloys over the full composition range,” PhD Thesis, 2013.
    [BibTeX] [Abstract] [Library link]

    The properties of InGaN alloys are important for many applications in optoelectronics, since the fundamental band gap of this material system spans the visible range. Calculating properties, particularly for InN, is theoretically challenging, especially obtaining accurate values for the band gap. We have developed a semiempirical parameterization for the simulation of (In,Ga)N using the density functional based tight binding method (DFTB), where the band gaps of InN and GaN have been empirically adjusted to match experiment. This is the first application of this method to In containing materials. We demonstrate the performance of this method by calculating a range of properties for both compounds and also their alloy for a range of crystal structures (wurtzite, zincblende and, for the pure compounds, rocksalt). There are several methods to model alloys of these materials, here the virtual crystal approximation and the cluster expansion method been used to study the alloy system of InGaN. While 8, 16 atom supercells are commonly used for cluster expansions, in this work these results are critically compared against the larger 32 atom cell, the effect of the ensemble used to simulate the alloy is also investigated by using both the Strictly Regular Solution and Generalised Quasi-Chemical approximations to provide limiting cases around the experimental conditions of Molecular Beam Epitaxy (MBE) and Metal-Organic Chemical Vapor Deposition (MOCVD) alloys.

    @PhdThesis{Fathi-Elfituri-thesis,
    author = {Fathi Elfituri},
    school = {University of Strathclyde},
    title = {Computational investigation of InGaN alloys over the full composition range},
    year = {2013},
    abstract = {The properties of InGaN alloys are important for many applications in optoelectronics, since the fundamental band gap of this material system spans the visible range. Calculating properties, particularly for InN, is theoretically challenging, especially obtaining accurate values for the band gap. We have developed a semiempirical parameterization for the simulation of (In,Ga)N using the density functional based tight binding method (DFTB), where the band gaps of InN and GaN have been empirically adjusted to match experiment. This is the first application of this method to In containing materials. We demonstrate the performance of this method by calculating a range of properties for both compounds and also their alloy for a range of crystal structures (wurtzite, zincblende and, for the pure compounds, rocksalt). There are several methods to model alloys of these materials, here the virtual crystal approximation and the cluster expansion method been used to study the alloy system of InGaN. While 8, 16 atom supercells are commonly used for cluster expansions, in this work these results are critically compared against the larger 32 atom cell, the effect of the ensemble used to simulate the alloy is also investigated by using both the Strictly Regular Solution and Generalised Quasi-Chemical approximations to provide limiting cases around the experimental conditions of Molecular Beam Epitaxy (MBE) and Metal-Organic Chemical Vapor Deposition (MOCVD) alloys.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_T3t945q80c},
    }

2012

  • L. Krishnan Jagadamma, “Characterisation of nanostructured light emitters,” PhD Thesis, 2012.
    [BibTeX] [Abstract] [Library link]

    Group III-nitride semiconductors are the dominant inorganic solid state light emitting materials, spanning the UV to infra-red spectral range. InGaN/GaN based LEDs and lasers are commercially available and intense research is being pursued to improve their efficiency. One practical approach is the development of functionalised and/or improved materials patterned on a nanometre length scale. This thesis presents the optical, morphological and compositional characterisation of III-nitride based nanostructured light emitters. The III-nitride nanostructures studied are GaN coalesced above arrays of either nanopyramids or nanocolumns, semipolar and nonpolar InGaN QWs on the facets of GaN nanopyramids, and thin epilayers of AlInN and AlInGaN. Spatially resolved optical characterisation of nano-ELOG GaN layers revealed a shift in the exciton-related band edge emission across the coalesced layer. This is related to Si doping and to strain effects. Study of the semipolar {10-11} InGaN QWs grown on the facets of GaN nanopyramids identified a blue shift in QW emission energy as the sampled region is moved up the pyramid facets. This shift is found to follow the release of the tensile strain towards the top of nanopyramid. Luminescence properties of nearly lattice matched AlInN epilayers investigated using CL, PL and PLE spectroscopic techniques revealed that the emission and bandgap energy of the AlInN layers are at higher energy than that of GaN. Results obtained from polarisation resolved PL measurements of AlInN epilayers point to two possible implications: the observed higher energy AlInN emission is either related to defects or this emission is due to carrier recombination occurring in InN clusters similar to those of InGaN epilayers. Optical properties of thin AlInGaN epilayers investigated using PL and PLE spectroscopy revealed a redshift in bandgap energy with increase in InN fraction. The observed spatial intensity fluctuations are discussed in terms of the InN compositional fluctuations and inhomogeneous strain effects.

    @PhdThesis{Lethy-Krishnan-Jagadamma-thesis,
    author = {Krishnan Jagadamma, Lethy},
    school = {University of Strathclyde},
    title = {Characterisation of nanostructured light emitters},
    year = {2012},
    abstract = {Group III-nitride semiconductors are the dominant inorganic solid state light emitting materials, spanning the UV to infra-red spectral range. InGaN/GaN based LEDs and lasers are commercially available and intense research is being pursued to improve their efficiency. One practical approach is the development of functionalised and/or improved materials patterned on a nanometre length scale. This thesis presents the optical, morphological and compositional characterisation of III-nitride based nanostructured light emitters. The III-nitride nanostructures studied are GaN coalesced above arrays of either nanopyramids or nanocolumns, semipolar and nonpolar InGaN QWs on the facets of GaN nanopyramids, and thin epilayers of AlInN and AlInGaN. Spatially resolved optical characterisation of nano-ELOG GaN layers revealed a shift in the exciton-related band edge emission across the coalesced layer. This is related to Si doping and to strain effects. Study of the semipolar {10-11} InGaN QWs grown on the facets of GaN nanopyramids identified a blue shift in QW emission energy as the sampled region is moved up the pyramid facets. This shift is found to follow the release of the tensile strain towards the top of nanopyramid. Luminescence properties of nearly lattice matched AlInN epilayers investigated using CL, PL and PLE spectroscopic techniques revealed that the emission and bandgap energy of the AlInN layers are at higher energy than that of GaN. Results obtained from polarisation resolved PL measurements of AlInN epilayers point to two possible implications: the observed higher energy AlInN emission is either related to defects or this emission is due to carrier recombination occurring in InN clusters similar to those of InGaN epilayers. Optical properties of thin AlInGaN epilayers investigated using PL and PLE spectroscopy revealed a redshift in bandgap energy with increase in InN fraction. The observed spatial intensity fluctuations are discussed in terms of the InN compositional fluctuations and inhomogeneous strain effects.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tft848q64r},
    }

  • G. Naresh-Kumar, “Electron channelling contrast imaging of nitride semiconductor thin films,” PhD Thesis, 2012.
    [BibTeX] [Abstract] [Library link]

    This work reports on the development of a scanning electron microscopy based technique known as electron channelling contrast imaging for characterising extended defects in nitride semiconductors. Extended defects in nitride semiconductors act as scattering centres for light and charge carriers and thus limit the performance of optoelectronic devices. Developing the capability to rapidly analyse extended defects, namely threading dislocation and stacking fault densities without any sample preparation represents a real step forward in the development of more efficient nitride-based semiconductor devices. In electron channelling contrast imaging, changes in crystallographic orientation, or changes in lattice constant due to local strain, are revealed by changes in grey scale in an image constructed by monitoring the intensity of backscattered electrons as an electron beam is scanned over a suitably oriented sample. Extremely small orientation changes are detectable, enabling small an gle tilt and rotation boundaries and extended defects to be imaged. Images with a resolution of tens of nanometres are obtainable with electron channelling contrast imaging. Vertical threading dislocations are revealed as spots with black-white contrast in electron channelling contrast imaging. A simple geometric procedure was developed which exploits the differences observed in the direction of this black-white contrast for screw, edge, and mixed dislocations for two electron channelling contrast images acquired from two symmetrically equivalent crystal planes. By using this procedure, an order of magnitude reduction in the time required to obtain quantitative information on dislocations is envisaged compared to the presently available techniques. The use of electron channelling contrast imaging to reveal and characterise basal plane stacking faults and partial dislocations in m-InGaN thin films and quantifying threading dislocations in InAlN thin films is also demonstrated. Preliminary work on combining electron channelling contrast imaging and cathodoluminescence imaging has been demonstrated for the first time to understand the effect of threading dislocations on light emission characteristics in nitride semiconductors.

    @PhdThesis{Gunasekar-Naresh-Kumar-thesis,
    author = {Gunasekar Naresh-Kumar},
    school = {University of Strathclyde},
    title = {Electron channelling contrast imaging of nitride semiconductor thin films},
    year = {2012},
    abstract = {This work reports on the development of a scanning electron microscopy based technique known as electron channelling contrast imaging for characterising extended defects in nitride semiconductors. Extended defects in nitride semiconductors act as scattering centres for light and charge carriers and thus limit the performance of optoelectronic devices. Developing the capability to rapidly analyse extended defects, namely threading dislocation and stacking fault densities without any sample preparation represents a real step forward in the development of more efficient nitride-based semiconductor devices. In electron channelling contrast imaging, changes in crystallographic orientation, or changes in lattice constant due to local strain, are revealed by changes in grey scale in an image constructed by monitoring the intensity of backscattered electrons as an electron beam is scanned over a suitably oriented sample. Extremely small orientation changes are detectable, enabling small an gle tilt and rotation boundaries and extended defects to be imaged. Images with a resolution of tens of nanometres are obtainable with electron channelling contrast imaging. Vertical threading dislocations are revealed as spots with black-white contrast in electron channelling contrast imaging. A simple geometric procedure was developed which exploits the differences observed in the direction of this black-white contrast for screw, edge, and mixed dislocations for two electron channelling contrast images acquired from two symmetrically equivalent crystal planes. By using this procedure, an order of magnitude reduction in the time required to obtain quantitative information on dislocations is envisaged compared to the presently available techniques. The use of electron channelling contrast imaging to reveal and characterise basal plane stacking faults and partial dislocations in m-InGaN thin films and quantifying threading dislocations in InAlN thin films is also demonstrated. Preliminary work on combining electron channelling contrast imaging and cathodoluminescence imaging has been demonstrated for the first time to understand the effect of threading dislocations on light emission characteristics in nitride semiconductors.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tft848q64r},
    }

  • F. Luckert, “Optical properties of the chalcopyrite semiconductors CuInSe₂, CuInS₂ and CuGaSe₂,” PhD Thesis, 2012.
    [BibTeX] [Abstract] [Library link]

    CuInSe₂, CuInS₂ and CuGaSe₂ are I-III-VI₂ compound semiconductors with a chalcopyrite structure. These ternary compounds exhibit favourable properties, such as direct band gaps and high absorption coefficients, for application as absorber layers in thin-film solar cells. Recently Cu(In,Ga)Se₂ based photovoltaic devices have demonstrated conversion efficiencies of 20.3 % which is the highest amongst polycrystalline thin-film solar cell technologies. This thesis describes a study of excitonic recombination processes in high quality CuInSe₂, CuInS₂ and CuGaSe₂ single crystals using photoluminescence (PL) spectroscopy as a function of excitation power, temperature and applied magnetic field. Excitation power dependent measurements confirm the identification of the free excitons in the PL spectra of the three chalcopyrite semiconductor compounds. Additional sharp lines in the PL spectra appear to be due to the radiative recombination of excitons bound to shallow hydrogenic defects. PL lines due to excitons bound to more complex defects with a low concentration of defects are also found in CuInSe₂ and CuInS₂. Analysis of the temperature dependent PL spectra lead to activation energies of the free and bound excitons in CuInSe₂, CuInS₂ and CuGaSe₂. In addition, phonon energies have been obtained from the temperature dependence of the free exciton spectral positions and of the full width at half maximum. PL spectra measured in applied magnetic fields allow estimation of the diamagnetic shift rates for CuInSe₂, CuInS₂ and CuGaSe₂. A first-order perturbation model leads to values for the excitonic reduced masses and the effective hole masses can be estimated. For CuInSe₂ a theoretically predicted anisotropy of the effective hole masses is demonstrated. The study of the excitonic states in CuInSe₂, CuInS₂ and CuGaSe₂ provides a deeper understanding of the electronic material properties which can facilitate further improvements in solar cell efficiencies.

    @PhdThesis{Franziska-Luckert-thesis,
    author = {Franziska Luckert},
    school = {University of Strathclyde},
    title = {Optical properties of the chalcopyrite semiconductors {CuInSe₂}, {CuInS₂} and {CuGaSe₂}},
    year = {2012},
    abstract = {CuInSe₂, CuInS₂ and CuGaSe₂ are I-III-VI₂ compound semiconductors with a chalcopyrite structure. These ternary compounds exhibit favourable properties, such as direct band gaps and high absorption coefficients, for application as absorber layers in thin-film solar cells. Recently Cu(In,Ga)Se₂ based photovoltaic devices have demonstrated conversion efficiencies of 20.3 % which is the highest amongst polycrystalline thin-film solar cell technologies. This thesis describes a study of excitonic recombination processes in high quality CuInSe₂, CuInS₂ and CuGaSe₂ single crystals using photoluminescence (PL) spectroscopy as a function of excitation power, temperature and applied magnetic field. Excitation power dependent measurements confirm the identification of the free excitons in the PL spectra of the three chalcopyrite semiconductor compounds. Additional sharp lines in the PL spectra appear to be due to the radiative recombination of excitons bound to shallow hydrogenic defects. PL lines due to excitons bound to more complex defects with a low concentration of defects are also found in CuInSe₂ and CuInS₂. Analysis of the temperature dependent PL spectra lead to activation energies of the free and bound excitons in CuInSe₂, CuInS₂ and CuGaSe₂. In addition, phonon energies have been obtained from the temperature dependence of the free exciton spectral positions and of the full width at half maximum. PL spectra measured in applied magnetic fields allow estimation of the diamagnetic shift rates for CuInSe₂, CuInS₂ and CuGaSe₂. A first-order perturbation model leads to values for the excitonic reduced masses and the effective hole masses can be estimated. For CuInSe₂ a theoretically predicted anisotropy of the effective hole masses is demonstrated. The study of the excitonic states in CuInSe₂, CuInS₂ and CuGaSe₂ provides a deeper understanding of the electronic material properties which can facilitate further improvements in solar cell efficiencies.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_T7m01bk82b},
    }

2009

  • L. Tan, “Nanometric characterisation of III-nitride semiconductors,” PhD Thesis, 2009.
    [BibTeX] [Abstract] [Library link]

    An investigation on the optical, composition and surface properties of the III-nitride ternary alloys InₓGa₁₋ₓN and Al₁₋ₓInₓN is presented. The structures studied are single quantum wells (SQWs) and epilayers. The effects of various GaN cap thicknesses, well widths, Si doping concentrations in the barriers, InN compositions, epilayer thicknesses and substrates are examined. The techniques used are photoluminescence (PL) and PL excitation (PLE) spectroscopy for investigating the optical properties, wavelength dispersive x-ray (WDX) and Rutherford backscattering spectrometry (RBS) for measuring InN composition. RBS is also used to study the crystalline quality of the samples; and secondary electron microscopy (SEM) and atomic force microscopy (AFM) for examining the surface quality. InₓGa₁₋ₓN/GaN SQWs with a variety of GaN cap thicknesses and different barrier doping levels were studied using PL spectroscopy. It was demonstrated that the InₓGa₁₋ₓN/GaN SQWs with thin caps give reduced PL intensity and larger PL linewidths, indicating degradation due to the increasing effects of the surface. As the doping level in the GaN barrier layers increases the luminescence intensity decreases, the linewidth increases and the SQW transition energy red-shifts. The latter is attributed to band-gap renormalisation. The properties of GaN/A₁₋ₓInₓN SQWs were investigated for varying quantum well width and for different substrates. The GaN/A₁₋ₓInₓN SQW luminescence peak energy decreases as the well-width increases, mainly due to the intense spontaneous polarisation fields, and its dependence on excitation power reveals the effects of carrier screening. Al₁₋ₓInₓN epilayers were also characterised and here the PL peak emission energies decrease with increasing InN composition. The energy bandgap bowing parameter is found to be linearly decreasing with increasing InN composition in the limited composition range studied. Cracks are observed in highly tensile strained epilayers with x < 0.10. For variations in layer thickness the surface quality and crystal quality worsens with increasing thickness. Finally, the samples on FS-GaN substrates have 3 to 4 atomic % more InN than those grown on sapphire. Surface quality degradation is observed in samples grown on non-polished sapphire substrates and AlyGa₁₋yN templates with high Al content.

    @PhdThesis{Lay-Theng-Tan-thesis,
    author = {Tan, Lay-Theng},
    school = {University of Strathclyde},
    title = {Nanometric characterisation of {III}-nitride semiconductors},
    year = {2009},
    abstract = {An investigation on the optical, composition and surface properties of the III-nitride ternary alloys InₓGa₁₋ₓN and Al₁₋ₓInₓN is presented. The structures studied are single quantum wells (SQWs) and epilayers. The effects of various GaN cap thicknesses, well widths, Si doping concentrations in the barriers, InN compositions, epilayer thicknesses and substrates are examined. The techniques used are photoluminescence (PL) and PL excitation (PLE) spectroscopy for investigating the optical properties, wavelength dispersive x-ray (WDX) and Rutherford backscattering spectrometry (RBS) for measuring InN composition. RBS is also used to study the crystalline quality of the samples; and secondary electron microscopy (SEM) and atomic force microscopy (AFM) for examining the surface quality. InₓGa₁₋ₓN/GaN SQWs with a variety of GaN cap thicknesses and different barrier doping levels were studied using PL spectroscopy. It was demonstrated that the InₓGa₁₋ₓN/GaN SQWs with thin caps give reduced PL intensity and larger PL linewidths, indicating degradation due to the increasing effects of the surface. As the doping level in the GaN barrier layers increases the luminescence intensity decreases, the linewidth increases and the SQW transition energy red-shifts. The latter is attributed to band-gap renormalisation. The properties of GaN/A₁₋ₓInₓN SQWs were investigated for varying quantum well width and for different substrates. The GaN/A₁₋ₓInₓN SQW luminescence peak energy decreases as the well-width increases, mainly due to the intense spontaneous polarisation fields, and its dependence on excitation power reveals the effects of carrier screening. Al₁₋ₓInₓN epilayers were also characterised and here the PL peak emission energies decrease with increasing InN composition. The energy bandgap bowing parameter is found to be linearly decreasing with increasing InN composition in the limited composition range studied. Cracks are observed in highly tensile strained epilayers with x < 0.10. For variations in layer thickness the surface quality and crystal quality worsens with increasing thickness. Finally, the samples on FS-GaN substrates have 3 to 4 atomic % more InN than those grown on sapphire. Surface quality degradation is observed in samples grown on non-polished sapphire substrates and AlyGa₁₋yN templates with high Al content.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/1vf34ij/SUSTAX_Tzc77sq19f},
    }

2008

  • K. Bejtka, "GaN-based microcavities : fabrication, characterisation and development," PhD Thesis, 2008.
    [BibTeX] [Abstract] [Library link]

    This thesis describes the fabrication and characterisation of double dielectric mirror GaN-based microcavities (MCs) along with investigation of the properties of various materials required for them, including GaN on AlInN, AlGaN, AlInGaN, and the use of FS-GaN substrate. Several processing routes for MC fabrication are detailed, with characterisation measurements after each step and for completed structures. Strong coupling between an exciton and a photon was observed for some approaches. The structures were grown by MOVPE and MBE on FS-GaN, sapphire and silicon substrates. Microcavities were fabricated using various techniques for substrate removal in order to access the back-side of active region for deposition of the bottom mirror. The finalised structures were characterised by optical spectroscopy. The structures grown on silicon resulted in the first observation of SC in transmission measurements for III-nitrides. High quality factors were observed from MCs grown on FS-GaN and on GaN-on-sapphire templates. These approaches open the way to improved structural quality of the active region, resulting from the use of substrates with low TDD.

    @PhdThesis{Katarzyna-Bejtka-thesis,
    author = {Katarzyna Bejtka},
    school = {University of Strathclyde},
    title = {{GaN}-based microcavities : fabrication, characterisation and development},
    year = {2008},
    abstract = {This thesis describes the fabrication and characterisation of double dielectric mirror GaN-based microcavities (MCs) along with investigation of the properties of various materials required for them, including GaN on AlInN, AlGaN, AlInGaN, and the use of FS-GaN substrate. Several processing routes for MC fabrication are detailed, with characterisation measurements after each step and for completed structures. Strong coupling between an exciton and a photon was observed for some approaches. The structures were grown by MOVPE and MBE on FS-GaN, sapphire and silicon substrates. Microcavities were fabricated using various techniques for substrate removal in order to access the back-side of active region for deposition of the bottom mirror. The finalised structures were characterised by optical spectroscopy. The structures grown on silicon resulted in the first observation of SC in transmission measurements for III-nitrides. High quality factors were observed from MCs grown on FS-GaN and on GaN-on-sapphire templates. These approaches open the way to improved structural quality of the active region, resulting from the use of substrates with low TDD.},
    url = {https://suprimo.lib.strath.ac.uk/permalink/f/17m2831/SUALMA2161967200002996},
    }

  • I. Roqan, "Experimental and theoretical studies of rare earth doped III-nitride semiconductor properties," PhD Thesis, 2008.
    [BibTeX] [Abstract] [Library link]

    In this Thesis present optical and structural investigations of RE³⁺ ions doped in-situ or implanted into III-nitride semiconductors. The optical studies used cathodoluminescence (CL), photoluminescence (PL) and photoluminescence excitation (PLE) spectroscopies. The structural properties employed atomic force microscopy (AFM), scanning electron microscopy (SEM), Rutherford backscattering spectrometry (RBS) and x-ray Diffraction (XRD) techniques. The elemental compositions were measured by wavelength dispersive x-ray analysis (WDX). The local structure of selected dopants was investigated using extended x-ray absorption fine structure (EXAFS). In addition, a theoretical investigation of RE clustering in GaN was carried out using local density functional theory (AIMPRO supercell code). The GaN growth temperature during in-situ doping with Tm shows a significant effect on the luminescence spectra, structural and compositional properties. The ¹D₂→³F₄, transition of Tm³⁺ in GaN:Tm films (~ 465 nm) has been observed for the first time to our knowledge. RE-implanted AlGaN and AlInN show significant improvements in the emission intensity and temperature quenching, compared to GaN, which dissociates at annealing temperatures greater than 1200 °C. The linewidth of the Dy³⁺ line shows a dependence on the alloy similar to that of the bound exciton in AlₓGa₁₋ₓN. AlInN:Eu shows no site multiplicity, in contrast to GaN:Eu. From CL and RBS measurements, a significant enhancement in the implantation damage recovery is found for an implantation temperature of 700 °C. The implantation fluence plays an important role in the implantation damage. CL, PL and PLE of Eu-implanted zincblende-GaN (ZB-GaN:Eu) and wurtzite-GaN (W-GaN:Eu) are compared to investigate the optical activation of GaN by Eu. A new sharp emission line at 627 nm is ascribed to an Eu³⁺ optical centre in ZB-GaN:Eu. The energy transfer process can modify the spectral signature of Eu³⁺ emission in an AlGaN sample co-implanted with Gd depending on the magnitude of the band gap. The theoretical studies of Tm clustering in GaN confirm the experimental EXAFS results. Electrically active states above the valence band are found for Tm clusters in GaN but not for isolated Tm atoms.

    @PhdThesis{Iman-Roqan-thesis,
    author = {Iman Roqan},
    title = {Experimental and theoretical studies of rare earth doped III-nitride semiconductor properties},
    school = {University of Strathclyde},
    year = {2008},
    abstract = {In this Thesis present optical and structural investigations of RE³⁺ ions doped in-situ or implanted into III-nitride semiconductors. The optical studies used cathodoluminescence (CL), photoluminescence (PL) and photoluminescence excitation (PLE) spectroscopies. The structural properties employed atomic force microscopy (AFM), scanning electron microscopy (SEM), Rutherford backscattering spectrometry (RBS) and x-ray Diffraction (XRD) techniques. The elemental compositions were measured by wavelength dispersive x-ray analysis (WDX). The local structure of selected dopants was investigated using extended x-ray absorption fine structure (EXAFS). In addition, a theoretical investigation of RE clustering in GaN was carried out using local density functional theory (AIMPRO supercell code).
    The GaN growth temperature during in-situ doping with Tm shows a significant effect on the luminescence spectra, structural and compositional properties. The ¹D₂→³F₄, transition of Tm³⁺ in GaN:Tm films (~ 465 nm) has been observed for the first time to our knowledge. RE-implanted AlGaN and AlInN show significant improvements in the emission intensity and temperature quenching, compared to GaN, which dissociates at annealing temperatures greater than 1200 °C. The linewidth of the Dy³⁺ line shows a dependence on the alloy similar to that of the bound exciton in AlₓGa₁₋ₓN. AlInN:Eu shows no site multiplicity, in contrast to GaN:Eu.
    From CL and RBS measurements, a significant enhancement in the implantation damage recovery is found for an implantation temperature of 700 °C. The implantation fluence plays an important role in the implantation damage. CL, PL and PLE of Eu-implanted zincblende-GaN (ZB-GaN:Eu) and wurtzite-GaN (W-GaN:Eu) are compared to investigate the optical activation of GaN by Eu. A new sharp emission line at 627 nm is ascribed to an Eu³⁺ optical centre in ZB-GaN:Eu.
    The energy transfer process can modify the spectral signature of Eu³⁺ emission in an AlGaN sample co-implanted with Gd depending on the magnitude of the band gap.
    The theoretical studies of Tm clustering in GaN confirm the experimental EXAFS results. Electrically active states above the valence band are found for Tm clusters in GaN but not for isolated Tm atoms.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2159936650002996}
    }

2007

  • K. Wang, "Investigation of the optical properties of rare earth doped and novel nitride semiconducters," PhD Thesis, 2007.
    [BibTeX] [Abstract] [Library link]

    In the past decade, the Group III- nitride semiconductors have emerged as the leading materials for green, blue and ultraviolet light emitting devices. Recently, rare earth (RE) doped III-nitrides have attracted considerable interest because of their potential application in light emitting devices covering the wide spectral range from infrared to visible. This thesis presents a study of a range of III-nitride hosts, with ion implantation used to introduce RE ions. The unimplanted AlInN epilayers are also studied in depth as the properties of this material are relatively less well known. Optical properties of RE implanted and annealed GaN, AlN, AlGaN and AlInN alloys have been investigated through photoluminescence (PL) and PL excitation (PLE) spectroscopy. Structural properties have been studied by Rutherford Backscattering Spectroscopy (RBS), x-ray diffraction (XRD), electron probe micro-analysis (EPMA), and secondary ion mass spectrometry (SIMS). PL and PLE study have demonstrated multiple Eu luminescent centres in the GaN host with distinct spectral patterns, excitation pathways, thermal quenching, and annealing dependences. In contrast, only one principle Eu centre has been observed in AlGaN alloys across the entire composition range. The linewidth of the Eu luminescence shows a similar broadening to the exciton linewidth as a function of AlN fraction. PLE spectra have demonstrated two excitation bands below the bandgaps with small energy shift (X₁ is from 3.26 to 3.54 eV and X₂ from 4.47 to 4.61 eV), which have been identified as core-exciton like complexes of Eu centres, formally equivalent to charge transfer states. The structural and optical properties of the AlInN epilayers have been characterized using RBS, XRD, XRD reciprocal space mapping (RSM), PL and PLE spectroscopy. Strong UV luminescence bands have been observed in the energy range from ~4.0 to ~3.5 eV. The effective bandgaps have been measured through PLE spectra and found to be from ~4.0 to 4.5 eV. AlInN epilayers implanted with Eu and Er ions have been systematically studied and compared with GaN. The intensity of RE luminescence is in the same order of magnitude for both hosts in the annealing temperature range from 800 to 1300 °C. The linewidth however is much broadened for RE ions in AlInN due to inhomogeneous broadening in the alloy.

    @PhdThesis{Ke-Wang-thesis,
    author = {Wang, Ke},
    title = {Investigation of the optical properties of rare earth doped and novel nitride semiconducters},
    school = {University of Strathclyde},
    year = {2007},
    abstract = {In the past decade, the Group III- nitride semiconductors have emerged as the leading materials for green, blue and ultraviolet light emitting devices. Recently, rare earth (RE) doped III-nitrides have attracted considerable interest because of their potential application in light emitting devices covering the wide spectral range from infrared to visible.
    This thesis presents a study of a range of III-nitride hosts, with ion implantation used to introduce RE ions. The unimplanted AlInN epilayers are also studied in depth as the properties of this material are relatively less well known. Optical properties of RE implanted and annealed GaN, AlN, AlGaN and AlInN alloys have been investigated through photoluminescence (PL) and PL excitation (PLE) spectroscopy. Structural properties have been studied by Rutherford Backscattering Spectroscopy (RBS), x-ray diffraction (XRD), electron probe micro-analysis (EPMA), and secondary ion mass spectrometry (SIMS).
    PL and PLE study have demonstrated multiple Eu luminescent centres in the GaN host with distinct spectral patterns, excitation pathways, thermal quenching, and annealing dependences. In contrast, only one principle Eu centre has been observed in AlGaN alloys across the entire composition range. The linewidth of the Eu luminescence shows a similar broadening to the exciton linewidth as a function of AlN fraction. PLE spectra have demonstrated two excitation bands below the bandgaps with small energy shift (X₁ is from 3.26 to 3.54 eV and X₂ from 4.47 to 4.61 eV), which have been identified as core-exciton like complexes of Eu centres, formally equivalent to charge transfer states.
    The structural and optical properties of the AlInN epilayers have been characterized using RBS, XRD, XRD reciprocal space mapping (RSM), PL and PLE spectroscopy. Strong UV luminescence bands have been observed in the energy range from ~4.0 to ~3.5 eV. The effective bandgaps have been measured through PLE spectra and found to be from ~4.0 to 4.5 eV. AlInN epilayers implanted with Eu and Er ions have been systematically studied and compared with GaN. The intensity of RE luminescence is in the same order of magnitude for both hosts in the annealing temperature range from 800 to 1300 °C. The linewidth however is much broadened for RE ions in AlInN due to inhomogeneous broadening in the alloy.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2160605500002996}
    }

2006

  • V. Kachkanov, "Structural properties of luminescent nitride semiconductors," PhD Thesis, 2006.
    [BibTeX] [Abstract] [Library link]

    This thesis presents the results of structural studies of InGaN, AlGaN, InAIN and rare-earth (RE) doped (Tm, Er and Eu) GaN by Extended X-ray Absorption Fine Structure (EXAFS), optical characterisation of RE-doped GaN by cathodoluminescence (CL), photoluminescence (PL) and photoluminescence excitation (PLE) spectroscopies. First attempts to identify the lattice location of emitting centres in nitride semiconductors using X-ray Excited Optical Luminescence (XEOL) for EXAFS detection are also presented. The local structure of InGaN showed no deviation from that of a random alloy in terms of cation coordination numbers. However, a comparative analysis of the In and Ga local structures showed evidence of a partial phase separation in the form of In-rich and Ga-rich InGaN regions. The degree of the phase separation was found to increase with decreasing InN fraction. This fact, considered together with a concurrent increase of luminescence efficiency, implies that In-rich InGaN clusters are responsible for efficient exciton localisation. For AlGaN samples, the observed segregation of GaN was attributed to spontaneous formation of a GaN/AIGaN superlattice. For InAIN, a partial phase separation in the form of In-rich and Al-rich InAIN regions was observed. The majority of RE³⁺ ions were found to occupy Ga substitutional sites in GaN. A smooth transition from isolated RE³⁺ centres, to the formation of RE³⁺ clusters and finally to the growth of a rare-earth nitride phase was observed with increasing impurity concentration. Analysis of CL, PL and PLE spectra suggests the prevalence of site multiplicity of RE³⁺ ions in GaN. At least two Eu³⁺ emitting species in GaN were clearly identified: one of which exhibits a broad absorption band centred at 385 nm and an other which requires higher excitation energy. In the case of Tm, the existence of Tm clusters inferred from EXAFS analysis was justified independently using PLE spectroscopy.

    @PhdThesis{Vyacheslav-Kachkanov-thesis,
    author = {Vyacheslav Kachkanov},
    title = {Structural properties of luminescent nitride semiconductors},
    school = {University of Strathclyde},
    year = {2006},
    abstract = {This thesis presents the results of structural studies of InGaN, AlGaN, InAIN and rare-earth (RE) doped (Tm, Er and Eu) GaN by Extended X-ray Absorption Fine Structure (EXAFS), optical characterisation of RE-doped GaN by cathodoluminescence (CL), photoluminescence (PL) and photoluminescence excitation (PLE) spectroscopies. First attempts to identify the lattice location of emitting centres in nitride semiconductors using X-ray Excited Optical Luminescence (XEOL) for EXAFS detection are also presented.
    The local structure of InGaN showed no deviation from that of a random alloy in terms of cation coordination numbers. However, a comparative analysis of the In and Ga local structures showed evidence of a partial phase separation in the form of In-rich and Ga-rich InGaN regions. The degree of the phase separation was found to increase with decreasing InN fraction. This fact, considered together with a concurrent increase of luminescence efficiency, implies that In-rich InGaN clusters are responsible for efficient exciton localisation. For AlGaN samples, the observed segregation of GaN was attributed to spontaneous formation of a GaN/AIGaN superlattice. For InAIN, a partial phase separation in the form of In-rich and Al-rich InAIN regions was observed.
    The majority of RE³⁺ ions were found to occupy Ga substitutional sites in GaN. A smooth transition from isolated RE³⁺ centres, to the formation of RE³⁺ clusters and finally to the growth of a rare-earth nitride phase was observed with increasing impurity concentration. Analysis of CL, PL and PLE spectra suggests the prevalence of site multiplicity of RE³⁺ ions in GaN. At least two Eu³⁺ emitting species in GaN were clearly identified: one of which exhibits a broad absorption band centred at 385 nm and an other which requires higher excitation energy. In the case of Tm, the existence of Tm clusters inferred from EXAFS analysis was justified independently using PLE spectroscopy.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2160916940002996}
    }

  • D. Amabile, "Investigation of the relationship between composition and luminescence properties of novel nitride based semiconductors," PhD Thesis, 2006.
    [BibTeX] [Abstract] [Library link]

    An investigation of the composition, structure and luminescence properties of the III-Nitride alloys AlInN, AlGaN, InGaN and AlInGaN is presented along with preliminary studies of europium (Eu) implanted AlInN. Wavelength dispersive x-ray (WDX) measurements have been used to determine sample composition and photoluminescence (PL) to probe luminescence properties. WDX has successfully determined the composition of thin layers of AlInGaN and InGaN, using multiple electron energies to probe depth effects. Challenges associated with quantifying the InN content of thin AlN-rich AlInN epilayers are addressed. These InN contents are compared with those determined by Rutherford backscattering spectroscopy (RBS) and high resolution x-ray diffraction (HR-XRD). All techniques yield values within ~3% InN of each other. Variations in AlN content across the surface of a quarter wafer of AlGaN have been investigated through WDX mapping with the AIN content at the centre being lower than at the edge due to parasitic reactions. PL peak emission energies increased (decreased) with increasing AlN (InN) content in AlInGaN and decreased with increasing InN content in InGaN and AlInN, reflecting the behaviour of the band-gap. Peak widths were shown to increase with AIN content in AlInGaN and to be smaller than those of InGaN. Both InGaN and AlInGaN samples exhibited S-shaped temperature dependence of PL peak energy and peak width. Different behaviour to that previously seen in MOVPE InGaN epilayers was observed for MBE InGaN, with the MBE layers exhibiting lower PL peak energies. Comparison of the emission properties of InGaN layers grown entirely by MBE with those grown on MOVPE-GaN templates showed the latter to be of superior quality. The increase of the AlInGaN band-gap with AIN content was followed using photoluminescence excitation (PLE). AlInN:Eu was found to exhibit broader and more intense Eu-related emission than GaN:Eu along with reduced thermal quenching.

    @PhdThesis{Debbie-Amabile-thesis,
    author = {Debbie Amabile},
    title = {Investigation of the relationship between composition and luminescence properties of novel nitride based semiconductors},
    school = {University of Strathclyde},
    year = {2006},
    abstract = {An investigation of the composition, structure and luminescence properties of the III-Nitride alloys AlInN, AlGaN, InGaN and AlInGaN is presented along with preliminary studies of europium (Eu) implanted AlInN. Wavelength dispersive x-ray (WDX) measurements have been used to determine sample composition and photoluminescence (PL) to probe luminescence properties. WDX has successfully determined the composition of thin layers of AlInGaN and InGaN, using multiple electron energies to probe depth effects. Challenges associated with quantifying the InN content of thin AlN-rich AlInN epilayers are addressed. These InN contents are compared with those determined by Rutherford backscattering spectroscopy (RBS) and high resolution x-ray diffraction (HR-XRD). All techniques yield values within ~3% InN of each other. Variations in AlN content across the surface of a quarter wafer of AlGaN have been investigated through WDX mapping with the AIN content at the centre being lower than at the edge due to parasitic reactions. PL peak emission energies increased (decreased) with increasing AlN (InN) content in AlInGaN and decreased with increasing InN content in InGaN and AlInN, reflecting the behaviour of the band-gap. Peak widths were shown to increase with AIN content in AlInGaN and to be smaller than those of InGaN. Both InGaN and AlInGaN samples exhibited S-shaped temperature dependence of PL peak energy and peak width. Different behaviour to that previously seen in MOVPE InGaN epilayers was observed for MBE InGaN, with the MBE layers exhibiting lower PL peak energies. Comparison of the emission properties of InGaN layers grown entirely by MBE with those grown on MOVPE-GaN templates showed the latter to be of superior quality. The increase of the AlInGaN band-gap with AIN content was followed using photoluminescence excitation (PLE). AlInN:Eu was found to exhibit broader and more intense Eu-related emission than GaN:Eu along with reduced thermal quenching.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2159994770002996}
    }

2005

  • S. M. de Sousa Pereira, "Structural and optical characterization of light emitting InGaN/GaN epitaxial layers," PhD Thesis, 2005.
    [BibTeX] [Abstract] [Library link]

    This thesis presents an experimental investigation of light emitting epitaxial layers based on indium gallium nitride (InₓGa₁₋ₓN). This group III-nitride ternary semiconductor alloy is used as the active layer in a novel class of optoelectronic devices, including light emitting diodes (LEDs) and laser diodes (LDs), under development to operate in the visible and ultraviolet regions of the electromagnetic spectrum. The structural and optical properties of InₓGa₁₋ₓN/GaN single layers and multiple quantum wells (MQWs) are characterized with an emphasis on their fundamental physical properties. The fundamental purpose of this work is to provide grounds for better understanding of the yet unclear physics of this important material system, and help to fill the gap between basic scientific knowledge and technological applications. Namely, this work addresses the issues of accurate measurement of the InN mole fraction (x), the influence of composition and strain in the structural and optical properties and the topic of phase segregation in InₓGa₁₋ₓN. The approach taken in this thesis is to integrate information provided by several complementary structural and optical characterization techniques through a systematic and multidisciplinary analysis. Specifically we combine: 1) sample growth by metal organic chemical vapour deposition (MOCVD) with specific features in an attempt to isolate the influence of structural parameters, such as layer thickness and composition; 2) Structural characterization by atomic force microscopy (AFM), scanning electron microscopy (SEM), X-ray diffraction (XRD) and Rutherford Backscattering spectrometry (RBS); 3) Optical characterisation at complementary length scales by: optical absorption (OA), photoluminescence (PL), and cathodoluminescence (CL) spectroscopy and confocal microscopy (CM) spectroscopy. Based on the results obtained, interpretation models to describe the structural and optical features are proposed, with a particular emphasis on the establishment of direct correlations between both. It is concluded that strainrelated phenomena must be taken into account to interpret the experimental results. Moreover it is also deduced that a detailed knowledge on the nanostructure is essential to explain InₓGa₁₋ₓN optical properties.

    @PhdThesis{Sergio-Pereira-thesis,
    author = {de Sousa Pereira, Sérgio M.},
    title = {Structural and optical characterization of light emitting {InGaN/GaN} epitaxial layers},
    school = {Universidade de Aveiro},
    year = {2005},
    abstract = {This thesis presents an experimental investigation of light emitting epitaxial layers based on indium gallium nitride (InₓGa₁₋ₓN). This group III-nitride ternary semiconductor alloy is used as the active layer in a novel class of optoelectronic devices, including light emitting diodes (LEDs) and laser diodes (LDs), under development to operate in the visible and ultraviolet regions of the electromagnetic spectrum. The structural and optical properties of InₓGa₁₋ₓN/GaN single layers and multiple quantum wells (MQWs) are characterized with an emphasis on their fundamental physical properties. The fundamental purpose of this work is to provide grounds for better understanding of the yet unclear physics of this important material system, and help to fill the gap between basic scientific knowledge and technological applications. Namely, this work addresses the issues of accurate measurement of the InN mole fraction (x), the influence of composition and strain in the structural and optical properties and the topic of phase segregation in InₓGa₁₋ₓN. The approach taken in this thesis is to integrate information provided by several complementary structural and optical characterization techniques through a systematic and multidisciplinary analysis. Specifically we combine: 1) sample growth by metal organic chemical vapour deposition (MOCVD) with specific features in an attempt to isolate the influence of structural parameters, such as layer thickness and composition; 2) Structural characterization by atomic force microscopy (AFM), scanning electron microscopy (SEM), X-ray diffraction (XRD) and Rutherford Backscattering spectrometry (RBS); 3) Optical characterisation at complementary length scales by: optical absorption (OA), photoluminescence (PL), and cathodoluminescence (CL) spectroscopy and confocal microscopy (CM) spectroscopy. Based on the results obtained, interpretation models to describe the structural and optical features are proposed, with a particular emphasis on the establishment of direct correlations between both. It is concluded that strainrelated phenomena must be taken into account to interpret the experimental results. Moreover it is also deduced that a detailed knowledge on the nanostructure is essential to explain InₓGa₁₋ₓN optical properties.},
    url = {http://ria.ua.pt/handle/10773/2666}
    }

  • F. Sweeney, "Characterisation of nitride thin films by electron backscattered diffraction correlated with cathodoluminescence spectroscopy and imaging," PhD Thesis, 2005.
    [BibTeX] [Abstract] [Library link]

    In this thesis I will illustrate how the techniques of electron backscatter diffraction (EBSD), cathodoluminescence spectroscopy and spectral mapping are very powerful for the characterisation of nitride thin films. EBSD is an exciting technique to apply to nitrides as we can use it to measure the orientation between nitride thin films and their substrates, measure tilt in nitride thin films and identify zincblende inclusions in a predominantly wurtzite film and vice-versa. EBSD may also be used to map the strain in nitride films due to the lattice mismatch between the nitride films and their substrates. The attraction of CL is its sensitivity to a solid’s crystal structure, composition, strain, doping and defect concentrations and to the ability to probe these material properties on a sub-micron level with varying degrees of penetration. EBSD, x-ray diffraction and cathodoluminescence have all been used to understand the properties of a series of GaN thin films grown on on-axis and two off-axis (4° and 10°) sapphire substrates. EBSD measurements reveal a tilt of 13±1° towards [10-10]GaN for the GaN thin films grown on the 10° off-axis substrate, but no discernable tilt is observed for the other two films. EBSD patterns also show an in-plane rotation of 90° of the GaN epilayer with respect to the sapphire substrate. EBSD and X-ray diffraction were also used to identify the misorientation of the sapphire substrate, both show a misorientation of approximately 10° towards the [12-10]sapphire direction. I also show results from an epitaxially laterally overgrown. GaN (ELOG) thin film. where EBSD reveals a lateral variation in both strain and tilt. CL mapping shows a similar trend in strain variation to that revealed by the EBSD. Finally I report on the use of EBSD to identify and map zinc blende regions in a predominantly wurtzite MBE film, with cathodoluminescence used to obtain correlated luminescence spectra and spectral maps.

    @PhdThesis{Francis-Sweeney-thesis,
    author = {Francis Sweeney},
    title = {Characterisation of nitride thin films by electron backscattered diffraction correlated with cathodoluminescence spectroscopy and imaging},
    school = {University of Strathclyde},
    year = {2005},
    abstract = {In this thesis I will illustrate how the techniques of electron backscatter diffraction (EBSD), cathodoluminescence spectroscopy and spectral mapping are very powerful for the characterisation of nitride thin films. EBSD is an exciting technique to apply to nitrides as we can use it to measure the orientation between nitride thin films and their substrates, measure tilt in nitride thin films and identify zincblende inclusions in a predominantly wurtzite film and vice-versa. EBSD may also be used to map the strain in nitride films due to the lattice mismatch between the nitride films and their substrates. The attraction of CL is its sensitivity to a solid’s crystal structure, composition, strain, doping and defect concentrations and to the ability to probe these material properties on a sub-micron level with varying degrees of penetration.
    EBSD, x-ray diffraction and cathodoluminescence have all been used to understand the properties of a series of GaN thin films grown on on-axis and two off-axis (4° and 10°) sapphire substrates. EBSD measurements reveal a tilt of 13±1° towards [10-10]GaN for the GaN thin films grown on the 10° off-axis substrate, but no discernable tilt is observed for the other two films. EBSD patterns also show an in-plane rotation of 90° of the GaN epilayer with respect to the sapphire substrate. EBSD and X-ray diffraction were also used to identify the misorientation of the sapphire substrate, both show a misorientation of approximately 10° towards the [12-10]sapphire direction.
    I also show results from an epitaxially laterally overgrown. GaN (ELOG) thin film. where EBSD reveals a lateral variation in both strain and tilt. CL mapping shows a similar trend in strain variation to that revealed by the EBSD.
    Finally I report on the use of EBSD to identify and map zinc blende regions in a predominantly wurtzite MBE film, with cathodoluminescence used to obtain correlated luminescence spectra and spectral maps.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2160729950002996}
    }

2004

  • R. Pecharromán-Gallego, "Investigations of the luminescence of GaN and InGaN/GaN quantum wells," PhD Thesis, 2004.
    [BibTeX] [Abstract] [Library link]

    This thesis presents investigations on the optical properties of the III-nitride material InGaN, as well as GaN. Studies of IngaN/GaN quantum wells grown by metalorganic vapour phase epitaxy (MOVPE) have been performed utilising mainly photoluminescence spectroscopy (PL) and also absorption spectroscopy. A set of InGaN single quantum wells (SQWs) emitting over the entire visible range has been studied and a technique to identify and minimise extra peaks in PL spectra due to Fabry-Perot interferences has been developed. These PL spectra were discussed and the long-wavelength emission was explained by intense piezoelectric fields coupled with InN-GaN segregation. The relatively strong phonon sidebands (PSBs) observed in the PL from SQWs and multi quantum well (MQWs) structures were also studied, and a longitudinal optical (LO) phonon energy of 89 meV was measured independent of the InN fraction. From an analysis of the Huang-Rhys parameters it was concluded that the fraction of excitons strongly localised was around 40 %. The dependence of the degree of localisation on the emission wavelength and number of wells was discussed. The unusual temperature dependence of the PL peak energy and linewidth of a number of samples emitting over the entire visible spectrum is also investigated. The origin is attributed to the competition of two recombination energy levels, as a result of fluctuations in the well width, in In content and/or its distribution. The temperature dependence of the PL emission peak energy was successfully fitted to a theoretical curve. Optical properties of GaN epilayers have also been investigated. The influence of the layer thickness on the PL peak energy was investigated and related to the compressive strain. The influence of the growth temperature of GaN structures has been studied, and a thermal stress model has been applied and discussed. The increase in PL intensity of lateral epitaxial overgrowth (LEO) grown GaN structures is discussed.

    @PhdThesis{Raul-Pecharroman-Gallego-thesis,
    author = {Raúl Pecharromán-Gallego},
    title = {Investigations of the luminescence of {GaN} and {InGaN/GaN} quantum wells},
    school = {University of Strathclyde},
    year = {2004},
    abstract = {This thesis presents investigations on the optical properties of the III-nitride material InGaN, as well as GaN. Studies of IngaN/GaN quantum wells grown by metalorganic vapour phase epitaxy (MOVPE) have been performed utilising mainly photoluminescence spectroscopy (PL) and also absorption spectroscopy. A set of InGaN single quantum wells (SQWs) emitting over the entire visible range has been studied and a technique to identify and minimise extra peaks in PL spectra due to Fabry-Perot interferences has been developed. These PL spectra were discussed and the long-wavelength emission was explained by intense piezoelectric fields coupled with InN-GaN segregation. The relatively strong phonon sidebands (PSBs) observed in the PL from SQWs and multi quantum well (MQWs) structures were also studied, and a longitudinal optical (LO) phonon energy of 89 meV was measured independent of the InN fraction. From an analysis of the Huang-Rhys parameters it was concluded that the fraction of excitons strongly localised was around 40 %. The dependence of the degree of localisation on the emission wavelength and number of wells was discussed. The unusual temperature dependence of the PL peak energy and linewidth of a number of samples emitting over the entire visible spectrum is also investigated. The origin is attributed to the competition of two recombination energy levels, as a result of fluctuations in the well width, in In content and/or its distribution. The temperature dependence of the PL emission peak energy was successfully fitted to a theoretical curve. Optical properties of GaN epilayers have also been investigated. The influence of the layer thickness on the PL peak energy was investigated and related to the compressive strain. The influence of the growth temperature of GaN structures has been studied, and a thermal stress model has been applied and discussed. The increase in PL intensity of lateral epitaxial overgrowth (LEO) grown GaN structures is discussed.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2157546140002996}
    }

2003

  • I. Fernández-Torrente, "MBE-grown InGaN epilayers," Master Thesis, 2003.
    [BibTeX] [Abstract] [Library link]

    Two main methods are used for InₓGa₁₋ₓN growth: high-temperature Metalorganic Chemical Vapour Deposition (MOCVD), proven to be successful for the growth of materials for commercial devices with low InN content (x<0.3), and low-temperature Molecular Beam Epitaxy (MBE), with better prospects for obtaining higher InN fractions (x>0.5). A direct comparison between MOCVD and MBE InGaN epilayers with similar InN concentration has been performed for the first time. The InN fraction in available MOCVD epilayers varies from 0 to 0.4 while in our MBE samples the range is 0.13-0.35. Wavelength Dispersive X-ray (WDX) analysis was performed to determine the composition of the samples. In-situ cathodoluminescence (CL) and ex-situ photoluminescence (PL) energy mapping were combined with large-area optical absorption spectroscopy in order to assess the optical properties of the materials. The composition dependencies of the optical energies are found to vary with the growth method. The trendline of PL/CL peak energy versus InN fraction is linear in both cases with comparable slopes but different intercepts for zero InN fraction. An “S-shape effect” of PL emission energy also appears with temperature, accompanied by an anomalous behaviour of the emission linewidth. We will try to explain these results.

    @MastersThesis{Isabel-Fernandez-Torres-thesis,
    author = {Fernández-Torrente, Isabel},
    title = {{MBE}-grown {InGaN} epilayers},
    school = {University of Strathclyde},
    year = {2003},
    abstract = {Two main methods are used for InₓGa₁₋ₓN growth: high-temperature Metalorganic Chemical Vapour Deposition (MOCVD), proven to be successful for the growth of materials for commercial devices with low InN content (x<0.3), and low-temperature Molecular Beam Epitaxy (MBE), with better prospects for obtaining higher InN fractions (x>0.5). A direct comparison between MOCVD and MBE InGaN epilayers with similar InN concentration has been performed for the first time. The InN fraction in available MOCVD epilayers varies from 0 to 0.4 while in our MBE samples the range is 0.13-0.35. Wavelength Dispersive X-ray (WDX) analysis was performed to determine the composition of the samples. In-situ cathodoluminescence (CL) and ex-situ photoluminescence (PL) energy mapping were combined with large-area optical absorption spectroscopy in order to assess the optical properties of the materials. The composition dependencies of the optical energies are found to vary with the growth method. The trendline of PL/CL peak energy versus InN fraction is linear in both cases with comparable slopes but different intercepts for zero InN fraction. An “S-shape effect” of PL emission energy also appears with temperature, accompanied by an anomalous behaviour of the emission linewidth. We will try to explain these results.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2157611590002996}
    }

2002

  • M. E. White, "Investigation of the optical properties, composition and logical structure of InGaN," PhD Thesis, 2002.
    [BibTeX] [Abstract] [Library link]

    This thesis presents work on the optical properties, composition and local structure of InGaN. Low temperature photoluminescence (PL) spectroscopy of InGaN epilayers was used to study linewidth, intensity and uniformity of the samples. Temperature dependence of the PL spectrum showed a decrease in the peak emission energy with increasing temperature for one epilayer sample and an ‘S’-shape temperature dependence of peak emission energy for a second sample. Composition analysis of the epilayers using three different experimental techniques; Rutherford backscattering spectrometry (RBS), electron probe microanalysis (EPMA) and extended X-ray absorption fine structure (EXAFS), found that the indium nitride fraction varies linearly with the PL peak emission energy. Confocal microscopy of InGaN epilayer, quantum well (QW) and quantum box (QB) samples revealed that InGaN luminescence segregates into bright and dark spots, ~1µm in diameter, and indicates that the origin of InGaN luminescence is of sub µm-scale. InGaN local structure parameters were derived from EXAFS measurements. Epilayer samples exhibited a single-phase InGaN alloy while a QB sample showed a two-phase mixture of components, one of which is nearly pure InN. Photoluminescence excitation (PLE) spectroscopy on MOCVD epilayers and MBE QWs and QBs showed that the bandgap energy varies linearly with detected emission energy for an individual sample. The PLE results from the MOCVD samples were consistent with data taken using optical absorption spectroscopy. Two different linear relationships between Stokes’ shift and detected emission energy were obtained for the MOCVD and MBE samples. Analysis of the accumulated results suggests that InGaN luminescence may be related to the combined effects of exciton localisation in InN-rich QDs, formed as a result of phase segregation in InGaN, accompanied by strain-induced piezoelectric fields.

    @PhdThesis{Madeleine-White-thesis,
    author = {Madeleine E. White},
    title = {Investigation of the optical properties, composition and logical structure of {InGaN}},
    school = {University of Strathclyde},
    year = {2002},
    abstract = {This thesis presents work on the optical properties, composition and local structure of InGaN. Low temperature photoluminescence (PL) spectroscopy of InGaN epilayers was used to study linewidth, intensity and uniformity of the samples. Temperature dependence of the PL spectrum showed a decrease in the peak emission energy with increasing temperature for one epilayer sample and an ‘S’-shape temperature dependence of peak emission energy for a second sample. Composition analysis of the epilayers using three different experimental techniques; Rutherford backscattering spectrometry (RBS), electron probe microanalysis (EPMA) and extended X-ray absorption fine structure (EXAFS), found that the indium nitride fraction varies linearly with the PL peak emission energy. Confocal microscopy of InGaN epilayer, quantum well (QW) and quantum box (QB) samples revealed that InGaN luminescence segregates into bright and dark spots, ~1µm in diameter, and indicates that the origin of InGaN luminescence is of sub µm-scale. InGaN local structure parameters were derived from EXAFS measurements. Epilayer samples exhibited a single-phase InGaN alloy while a QB sample showed a two-phase mixture of components, one of which is nearly pure InN. Photoluminescence excitation (PLE) spectroscopy on MOCVD epilayers and MBE QWs and QBs showed that the bandgap energy varies linearly with detected emission energy for an individual sample. The PLE results from the MOCVD samples were consistent with data taken using optical absorption spectroscopy. Two different linear relationships between Stokes’ shift and detected emission energy were obtained for the MOCVD and MBE samples. Analysis of the accumulated results suggests that InGaN luminescence may be related to the combined effects of exciton localisation in InN-rich QDs, formed as a result of phase segregation in InGaN, accompanied by strain-induced piezoelectric fields.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2155209590002996}
    }

  • C. Lewis, "Compositional analysis of dilute nitride semiconductor structures," Master Thesis, 2002.
    [BibTeX] [Abstract] [Library link]

    Using Wavelength Dispersive Spectrometry (WDS) on an Electron Probe Micro-Analyser (EPMA), the nitrogen compositions of a series of GaAs₁₋ₓNₓ samples grown by Molecular Beam Epitaxy (MBE) were investigated. By measuring characteristic X-rays intensities from the samples and comparing with known standards the nitrogen fractions can be calculated. Much consideration of experimental conditions utilised enabled a refined procedure that could detect and quantify the low levels of nitrogen present, down to 0.01 percentage by weight. The nitrogen compositions (x) are found to be in good agreement with measurements of X-Ray Diffraction (XRD) and Rutherford BackScattering Spectrometry (RBS) by the material growers. Nitrogen compositions (quoted as percentages) of 1.04% < x <2.42% were measured in the samples studied, with fractional errors of ~ 1 %. There was no evidence of compositional non-uniformity in the samples studied.

    @MastersThesis{Chris-Lewis-thesis,
    author = {Chris Lewis},
    title = {Compositional analysis of dilute nitride semiconductor structures},
    school = {University of Strathclyde},
    year = {2002},
    abstract = {Using Wavelength Dispersive Spectrometry (WDS) on an Electron Probe Micro-Analyser (EPMA), the nitrogen compositions of a series of GaAs₁₋ₓNₓ samples grown by Molecular Beam Epitaxy (MBE) were investigated. By measuring characteristic X-rays intensities from the samples and comparing with known standards the nitrogen fractions can be calculated. Much consideration of experimental conditions utilised enabled a refined procedure that could detect and quantify the low levels of nitrogen present, down to 0.01 percentage by weight.
    The nitrogen compositions (x) are found to be in good agreement with measurements of X-Ray Diffraction (XRD) and Rutherford BackScattering Spectrometry (RBS) by the material growers. Nitrogen compositions (quoted as percentages) of 1.04% < x <2.42% were measured in the samples studied, with fractional errors of ~ 1 %. There was no evidence of compositional non-uniformity in the samples studied.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2155450170002996}
    }

2000

  • S. K. Manson-Smith, "Investigation into scanning tunnelling luminescence microscopy," PhD Thesis, 2000.
    [BibTeX] [Abstract] [Library link]

    This work reports on the development of a scanning tunnelling luminescence (STL) microscope and its application to the study of III-nitride semiconductor materials used in the production of light emitting devices STL microscopy is a technique which uses the high resolution topographic imaging capabilities of the scanning tunnelling microscope (STM) to generate high resolution luminescence images. The STM tunnelling current acts as a highly localised source of electrons (or holes) which generates luminescence in certain materials. Light generated at the STM tunnelling junction is collected concurrently with the height variation of the tunnelling probe as it is scanned across a sample surface, producing simultaneous topographic and luminescence images. Due to the very localised excitation source high resolution luminescence images can be obtained. Spectroscopic resolution can be obtained by using filters. Additionally, the variation of luminescence intensity with tunnel current and with bias voltage can provide information on recombination processes and material properties. The design and construction of a scanning tunnelling luminescence microscope is described in detail. Operating under ambient conditions, the microscope has several novel features, including a new type of miniature inertial slider-based approach motor, large solid-angle light collection optical arrangement and a tip-height regulation system which requires the minimum of operator input Room temperature STE images of an InGaN quantum well-based light emitting diode structure are presented. The images reveal correlation between surface topography and luminescence. Topographic imaging shows hexagonal pits on the top surface. Luminescence imaging shows strong emission from the hexagonal pits. Low-temperature STL studies were performed at Nagoya University, Japan, using a modified commercial STM. Topographic and luminescence images of an InGaN epilayer of low indium concentration were obtained by STL. Luminescence segregation on a nanometre scale can be seen. The material presented in this thesis clearly demonstrates the potential of STL for the high-resolution topographic and luminescence characterisation of materials and light emitting devices.

    @PhdThesis{Sacha-Manson-Smith-thesis,
    author = {Sacha K. Manson-Smith},
    title = {Investigation into scanning tunnelling luminescence microscopy},
    school = {University of Strathclyde},
    year = {2000},
    abstract = {This work reports on the development of a scanning tunnelling luminescence (STL) microscope and its application to the study of III-nitride semiconductor materials used in the production of light emitting devices
    STL microscopy is a technique which uses the high resolution topographic imaging capabilities of the scanning tunnelling microscope (STM) to generate high resolution luminescence images. The STM tunnelling current acts as a highly localised source of electrons (or holes) which generates luminescence in certain materials. Light generated at the STM tunnelling junction is collected concurrently with the height variation of the tunnelling probe as it is scanned across a sample surface, producing simultaneous topographic and luminescence images. Due to the very localised excitation source high resolution luminescence images can be obtained. Spectroscopic resolution can be obtained by using filters. Additionally, the variation of luminescence intensity with tunnel current and with bias voltage can provide information on recombination processes and material properties.
    The design and construction of a scanning tunnelling luminescence microscope is described in detail. Operating under ambient conditions, the microscope has several novel features, including a new type of miniature inertial slider-based approach motor, large solid-angle light collection optical arrangement and a tip-height regulation system which requires the minimum of operator input
    Room temperature STE images of an InGaN quantum well-based light emitting diode structure are presented. The images reveal correlation between surface topography and luminescence. Topographic imaging shows hexagonal pits on the top surface. Luminescence imaging shows strong emission from the hexagonal pits.
    Low-temperature STL studies were performed at Nagoya University, Japan, using a modified commercial STM. Topographic and luminescence images of an InGaN epilayer of low indium concentration were obtained by STL. Luminescence segregation on a nanometre scale can be seen.
    The material presented in this thesis clearly demonstrates the potential of STL for the high-resolution topographic and luminescence characterisation of materials and light emitting devices.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2155915310002996}
    }

1999

  • K. G. Chinyama, "Morphological and photoluminescence studies of ultrathin CdSe in ZnSe well structures," PhD Thesis, 1999.
    [BibTeX] [Abstract] [Library link]

    Using a combination of transmission electron microscopy (TEM), high resolution transmission electron microscopy (HRTEM), digital analysis of lattice images (DALI), correspondence analysis and composition evaluation by lattice fringe analysis (CELFA), we present at near-atomic resolution the morphology of cadmium selenide quantum confinement layers embedded in zinc selenide matrix. We reveal the first direct measurements of the composition of these quantum wells (QWS) clearly indicating that the nominal CdSe inserted layer is completely transformed into a ZnₓCd₁₋ₓSe disordered alloy resulting in the Cadmium-content being high in the core of the self-assembled islands, commonly referred to as quantum dots (QDS). These QDs themselves are not binary CdSe but contain a high zinc content. Photoluminescence (PL) measurements show that PL surface mapping is a useful tool to distinguish between a QW structure characterised by interface roughness from that containing QDs. The mapping indicates that in the former structures, there is strong spatial correlation between the variation in the peak energy and that in the linewidth compared to the latter structures. Temperature-dependent PL measurements reveal strong PL band line-narrowing in the rough structures in comparison to those with QDs leading to some critical temperature Tc, after which thermal broadening sets in. This temperature is sensitive to well width, thus presenting a good parameter to characterise exciton localisation. Furthermore time-resolved results reveal anomalous temporal behaviour of PL intensity suggesting the exciton relaxation process in these structures takes other complicated routes. Finally, we investigated the distribution of oscillator strengths for the recombination of excitons in QW structures, trapped in local minima of the confinement potential using a statistical topographic model of the potential. The predicted distribution of oscillator strengths is very different from the Porter-Thomas distribution which characterises disordered systems, and is notable for the fact that small oscillator strengths are extremely rare.

    @PhdThesis{Kaumba-Chinyama-thesis,
    author = {Kaumba G. Chinyama},
    title = {Morphological and photoluminescence studies of ultrathin {CdSe} in {ZnSe} well structures},
    school = {University of Strathclyde},
    year = {1999},
    abstract = {Using a combination of transmission electron microscopy (TEM), high resolution transmission electron microscopy (HRTEM), digital analysis of lattice images (DALI), correspondence analysis and composition evaluation by lattice fringe analysis (CELFA), we present at near-atomic resolution the morphology of cadmium selenide quantum confinement layers embedded in zinc selenide matrix. We reveal the first direct measurements of the composition of these quantum wells (QWS) clearly indicating that the nominal CdSe inserted layer is completely transformed into a ZnₓCd₁₋ₓSe disordered alloy resulting in the Cadmium-content being high in the core of the self-assembled islands, commonly referred to as quantum dots (QDS). These QDs themselves are not binary CdSe but contain a high zinc content.
    Photoluminescence (PL) measurements show that PL surface mapping is a useful tool to distinguish between a QW structure characterised by interface roughness from that containing QDs. The mapping indicates that in the former structures, there is strong spatial correlation between the variation in the peak energy and that in the linewidth compared to the latter structures. Temperature-dependent PL measurements reveal strong PL band line-narrowing in the rough structures in comparison to those with QDs leading to some critical temperature Tc, after which thermal broadening sets in. This temperature is sensitive to well width, thus presenting a good parameter to characterise exciton localisation. Furthermore time-resolved results reveal anomalous temporal behaviour of PL intensity suggesting the exciton relaxation process in these structures takes other complicated routes.
    Finally, we investigated the distribution of oscillator strengths for the recombination of excitons in QW structures, trapped in local minima of the confinement potential using a statistical topographic model of the potential. The predicted distribution of oscillator strengths is very different from the Porter-Thomas distribution which characterises disordered systems, and is notable for the fact that small oscillator strengths are extremely rare.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2152900580002996}
    }

  • F. J. McGow, "Optical properties of strained layer semiconductor structures," PhD Thesis, 1999.
    [BibTeX] [Abstract] [Library link]

    A series of strain-balanced InGaAs/In(Ga,Al)As multiple quantum well samples were investigated using optical spectroscopy. An analysis of magneto-absorption spectra provided information on the effects of inbuilt strain on the relative positions of heavy and light hole valence subbands, and on the in-plane effective masses of the holes. Both tension and compression have been shown to reduce the value of the heavy hole in-plane effective mass from its value in the lattice matched system. For one material three sets of excitonic Landau levels could be resolved in the spectra, allowing estimates of the reduced excitonic effective mass of the first and second electron to heavy hole (0.040±0.001me and 0.034±0.02me) and the first electron to light hole (0.053±0.002me) excitons. Enhancement in the light hole in-plane effective mass, relative to the heavy hole, is shown to result from the tensile strain in the quantum wells. The theoretical fitting of transition energies observed in the 0T absorption spectra of samples with different degrees of strain in the wells is described. This allowed estimates of the conduction band offset ratio (Qc) to be made for strain balanced systems. The value of Qc was found to be Qc=0.74±0.10, 0.55±0.08 and 0.52±0.08 for samples with well strain of -0.10%, 0.14% and 0.40% respectively. This decrease in Qc as the degree of tension in the wells was increased was more greatly pronounced than in single strain systems. Photoluminescence measurements made across the surface of the materials show little shift, indicating homogenous sample compositions. Temperature dependant photoluminescence measurements of these samples show evidence of thermal excitation of carriers between the first light and heavy hole states. The results of photomodulated transmission measurements demonstrate the potential application of In(Ga,Al)As strain balanced superlattices as all-optical modulators.

    @PhdThesis{Fiona-McGow-thesis,
    author = {Fiona J. McGow},
    title = {Optical properties of strained layer semiconductor structures},
    school = {University of Strathclyde},
    year = {1999},
    abstract = {A series of strain-balanced InGaAs/In(Ga,Al)As multiple quantum well samples were investigated using optical spectroscopy. An analysis of magneto-absorption spectra provided information on the effects of inbuilt strain on the relative positions of heavy and light hole valence subbands, and on the in-plane effective masses of the holes. Both tension and compression have been shown to reduce the value of the heavy hole in-plane effective mass from its value in the lattice matched system. For one material three sets of excitonic Landau levels could be resolved in the spectra, allowing estimates of the reduced excitonic effective mass of the first and second electron to heavy hole (0.040±0.001me and 0.034±0.02me) and the first electron to light hole (0.053±0.002me) excitons. Enhancement in the light hole in-plane effective mass, relative to the heavy hole, is shown to result from the tensile strain in the quantum wells.
    The theoretical fitting of transition energies observed in the 0T absorption spectra of samples with different degrees of strain in the wells is described. This allowed estimates of the conduction band offset ratio (Qc) to be made for strain balanced systems. The value of Qc was found to be Qc=0.74±0.10, 0.55±0.08 and 0.52±0.08 for samples with well strain of -0.10%, 0.14% and 0.40% respectively. This decrease in Qc as the degree of tension in the wells was increased was more greatly pronounced than in single strain systems.
    Photoluminescence measurements made across the surface of the materials show little shift, indicating homogenous sample compositions. Temperature dependant photoluminescence measurements of these samples show evidence of thermal excitation of carriers between the first light and heavy hole states. The results of photomodulated transmission measurements demonstrate the potential application of In(Ga,Al)As strain balanced superlattices as all-optical modulators.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2153476590002996}
    }

  • A. Mohammed, "Optical and structural characterisation of low dimensional structures using electron beam excitation systems," PhD Thesis, 1999.
    [BibTeX] [Abstract] [Library link]

    This thesis presents studies on optical and structural characterisation of low dimensional structures that emit light within the visible region of the spectrum (ZnSe-based and GaN-based) using electron beam excitation systems. A guideline for theoretical design of strained and strain-free ZnCdSe/ZnSe quantum well (QW) structures with good carrier confinements has been provided. The emission energies of several QW structures have been estimated prior to the growth and are compared with the experimental results. The diffusion Monte Carlo technique that requires no trial wave function has been employed for the first time to calculate exciton binding energies in these structures. Characterisation of ZnCdSe/ZnSe single- and multi- QW structures by Cathodoluminescence (CL) and Photoluminescence (PL) techniques have revealed potential fluctuations and disorder and vertical carrier transport at low excitation energy in some structures. The temperature dependence of CL intensity of some ZnCdSe/ZnSe QW structures at low electron beam excitation energies have indicated a blue shift in emission energy with increasing temperature due to the presence of potential fluctuations. The luminescence thermal quenching process at high temperatures is found to be due to emission of carriers or excitons out of the confined QW states. The emitted carriers are subsequently captured by the wider neighbouring QW in good quality structures, nonradiative recombination centres suppress this carrier transfer in bad quality structures. The temperatures at which the QW luminescence starts to quench and the activation energies of luminescence quenching are found to depend on excitation conditions, sample quality and QW depth. The results of CL intensity dependence on the excitation intensity revealed that luminescence from good quality QW structures is dominated by radiative recombination processes even at high temperatures during thermal quenching. In contrast, in defected structures non-radiative recombination mechanisms dominate the luminescence properties at all temperatures. Secondary electron images of hexagonal growth hillocks of GaN obtained at a range of electron beam excitation energies vary because of the different signals involved in the imaging. Electron backscatter diffraction measurements have been used for phase identification and lattice constants determination in a strained GaN epilayer.

    @PhdThesis{Abdullahi-Mohammed-thesis,
    author = {Abdullahi Mohammed},
    title = {Optical and structural characterisation of low dimensional structures using electron beam excitation systems},
    school = {University of Strathclyde},
    year = {1999},
    abstract = {This thesis presents studies on optical and structural characterisation of low dimensional structures that emit light within the visible region of the spectrum (ZnSe-based and GaN-based) using electron beam excitation systems.
    A guideline for theoretical design of strained and strain-free ZnCdSe/ZnSe quantum well (QW) structures with good carrier confinements has been provided. The emission energies of several QW structures have been estimated prior to the growth and are compared with the experimental results. The diffusion Monte Carlo technique that requires no trial wave function has been employed for the first time to calculate exciton binding energies in these structures.
    Characterisation of ZnCdSe/ZnSe single- and multi- QW structures by Cathodoluminescence (CL) and Photoluminescence (PL) techniques have revealed potential fluctuations and disorder and vertical carrier transport at low excitation energy in some structures.
    The temperature dependence of CL intensity of some ZnCdSe/ZnSe QW structures at low electron beam excitation energies have indicated a blue shift in emission energy with increasing temperature due to the presence of potential fluctuations. The luminescence thermal quenching process at high temperatures is found to be due to emission of carriers or excitons out of the confined QW states. The emitted carriers are subsequently captured by the wider neighbouring QW in good quality structures, nonradiative recombination centres suppress this carrier transfer in bad quality structures. The temperatures at which the QW luminescence starts to quench and the activation energies of luminescence quenching are found to depend on excitation conditions, sample quality and QW depth.
    The results of CL intensity dependence on the excitation intensity revealed that luminescence from good quality QW structures is dominated by radiative recombination processes even at high temperatures during thermal quenching. In contrast, in defected structures non-radiative recombination mechanisms dominate the luminescence properties at all temperatures.
    Secondary electron images of hexagonal growth hillocks of GaN obtained at a range of electron beam excitation energies vary because of the different signals involved in the imaging. Electron backscatter diffraction measurements have been used for phase identification and lattice constants determination in a strained GaN epilayer.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2153170170002996}
    }

1997

  • J. P. Creasey, "Photoluminescence studies of space-charge effects in II-VI type II strained layer superlattices," PhD Thesis, 1997.
    [BibTeX] [Abstract] [Library link]

    CdS/CdSe and ZnSe/CdS strained superlattices (SLS) on (100) and (111) GaAs substrates were grown by Metal Organic Vapour Phase Epitaxy. Low temperature photoluminescence (PL) studies on these type-II superlattices are presented in this thesis. CB offset values of 300 meV and 1040 meV are determined for the cubic heterojunction and the cubic ZnSe/CdS heterojunction, on (100) GaAs, respectively from PL. A refined estimate of the wurtzite CdS/CdSe CB offset is made by comparing PL observations which show double phase growth of CdS/CdSe and ZnSe/CdS SLS on (100) GaAs, this estimate agrees with the value estimated by Halsall et al (1992) from piezoelectric CdS/CdSe SLS grown on (111)GaAs. PL studies of non-piezoelectric CdS/CdSe and CdS/ZnSe SLS grown on (100) GaAs substrates show SL peak shifts of up to 29 meV per order increase in the excitation density (shift-factor). These shifts are explained in terms of band bending in the quantum well due to large photo-induced space-charge fields. Broad linewidths in type-II SLS are attributed to a combination of interdiffusion at the heterojunction interfaces and well-width fluctuations in the superattice. A shift-factor of 57 meV observed in a ZnSe/CdS SLS, on (111), is attributed to large piezoelectric fields; allowed in cubic ZnSe due to the dependence of the Piezoelectric coefficients on layer strain (non-linear piezoelectric effect). PL band-energies are compared to a calculation of SL transition energy in piezoelectric CdS/CdSe SLS as a function of carrier density. An excellent fit is obtained. A highly piezoelectric structure with 16 nm period length shows the largest shift-factor ever measured (~200 meV). This result suggests that the critical period is greater than 16 nm for symmetric CdS/CdSe SLS. Finally, a stimulated emission band in CdS/CdSe SLS at ~1.78 eV is shown to originate from a real-space exciton transition in the CdSe layers of the SLS.

    @PhdThesis{Jonathan-Creasey-thesis,
    author = {Jonathan P. Creasey},
    title = {Photoluminescence studies of space-charge effects in {II-VI} type {II} strained layer superlattices},
    school = {University of Strathclyde},
    year = {1997},
    abstract = {CdS/CdSe and ZnSe/CdS strained superlattices (SLS) on (100) and (111) GaAs substrates were grown by Metal Organic Vapour Phase Epitaxy. Low temperature photoluminescence (PL) studies on these type-II superlattices are presented in this thesis.
    CB offset values of 300 meV and 1040 meV are determined for the cubic heterojunction and the cubic ZnSe/CdS heterojunction, on (100) GaAs, respectively from PL. A refined estimate of the wurtzite CdS/CdSe CB offset is made by comparing PL observations which show double phase growth of CdS/CdSe and ZnSe/CdS SLS on (100) GaAs, this estimate agrees with the value estimated by Halsall et al (1992) from piezoelectric CdS/CdSe SLS grown on (111)GaAs.
    PL studies of non-piezoelectric CdS/CdSe and CdS/ZnSe SLS grown on (100) GaAs substrates show SL peak shifts of up to 29 meV per order increase in the excitation density (shift-factor). These shifts are explained in terms of band bending in the quantum well due to large photo-induced space-charge fields. Broad linewidths in type-II SLS are attributed to a combination of interdiffusion at the heterojunction interfaces and well-width fluctuations in the superattice.
    A shift-factor of 57 meV observed in a ZnSe/CdS SLS, on (111), is attributed to large piezoelectric fields; allowed in cubic ZnSe due to the dependence of the Piezoelectric coefficients on layer strain (non-linear piezoelectric effect).
    PL band-energies are compared to a calculation of SL transition energy in piezoelectric CdS/CdSe SLS as a function of carrier density. An excellent fit is obtained.
    A highly piezoelectric structure with 16 nm period length shows the largest shift-factor ever measured (~200 meV). This result suggests that the critical period is greater than 16 nm for symmetric CdS/CdSe SLS.
    Finally, a stimulated emission band in CdS/CdSe SLS at ~1.78 eV is shown to originate from a real-space exciton transition in the CdSe layers of the SLS.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2154447200002996}
    }

1992

  • F. Yang, "Wide bandgap II-VI quantum wells : optical characterisation and the role of disorder," PhD Thesis, 1992.
    [BibTeX] [Abstract] [Library link]

    Photoluminescence (PL) and absorption spectroscopic studies of II-VI multiple quantum wells (MQWs) are presented in this thesis. PL spectra of MQW samples are investigated at various temperature and excitation intensities. The peak positions of PL in MQW samples agreed qualitatively with the assignment of the PL quantum confined hh-exciton transitions. A rate equation analysis of the excitation power dependence of the PL intensity shows that under super band gap excitation such dependences can not distinguish band-to-band from excitonic transitions, but do provide information on the carrier generation rate coefficient. A study of the absorption spectra of excitons in ZnSe-ZnS MQWs with well widths ranging from 0.6nm to 7.6nm is reported. The n=1 heavy hole (hh) and light hole (lh) exciton absorptions are clearly resolved for all samples even at room temperature. Theoretical estimates of energies of the 1s and 2s states of n=1 hh-exciton, and 1s state of n=1 lh-exciton, which takes account of partial strain relaxation, quantum confinement of free carriers and exciton binding energy, are in excellent agreement with the experimental peak energies. Special optical properties of ultrathin II-VI QWs are studied using a CdSe-ZnSe MQW as an example. Laser excitation close to, or lying within, the disordered broadened exciton PL spectrum reveals a series of sharp lines separated by the LO-phonon energy, which increases with the exciting photon energy. The variation of the phonon energy is explained by sampling by localised exciton of effective alloys in different wells. Time-resolved PL spectra show that the sharp lines have the same decay time as the broad band luminescence. For each exciting photon energy, the photon energy does not change with the phonon order. This suggests the sharp lines are due to resonant processes. In addition, time-resolved PL spectra reveal an almost constant decay time across the PL band. A new theory of the inhomogeneous broadening of the exciton optical transitions is presented. The line shape of exciton absorption is explained by the distribution of the band gap. The line shape of the exciton PL is explained by the distribution of heights (energies) of spatial minima of the band gap, which implies that all PL emission is due to classically localised excitons. A semi-empirical and an analytical approach predict successfully the PL lineshape and peak position for a Gaussian bandgap distribution. Both approaches reveal a quasi-Gaussian PL band, which is consistent with most experimental results. This theory reveals a linear relation between the hh-exciton linewidth Γ and the Stokes' shift Δ: Δ=0.6 Γ. The sharp line structures under resonant excitation are explained by LO-phonon assisted tunneling of exciton between localised states.

    @PhdThesis{Yang-Fang-thesis,
    author = {Yang, Fang},
    title = {Wide bandgap {II-VI} quantum wells : optical characterisation and the role of disorder},
    school = {University of Strathclyde},
    year = {1992},
    abstract = {Photoluminescence (PL) and absorption spectroscopic studies of II-VI multiple quantum wells (MQWs) are presented in this thesis. PL spectra of MQW samples are investigated at various temperature and excitation intensities. The peak positions of PL in MQW samples agreed qualitatively with the assignment of the PL quantum confined hh-exciton transitions. A rate equation analysis of the excitation power dependence of the PL intensity shows that under super band gap excitation such dependences can not distinguish band-to-band from excitonic transitions, but do provide information on the carrier generation rate coefficient.
    A study of the absorption spectra of excitons in ZnSe-ZnS MQWs with well widths ranging from 0.6nm to 7.6nm is reported. The n=1 heavy hole (hh) and light hole (lh) exciton absorptions are clearly resolved for all samples even at room temperature. Theoretical estimates of energies of the 1s and 2s states of n=1 hh-exciton, and 1s state of n=1 lh-exciton, which takes account of partial strain relaxation, quantum confinement of free carriers and exciton binding energy, are in excellent agreement with the experimental peak energies.
    Special optical properties of ultrathin II-VI QWs are studied using a CdSe-ZnSe MQW as an example. Laser excitation close to, or lying within, the disordered broadened exciton PL spectrum reveals a series of sharp lines separated by the LO-phonon energy, which increases with the exciting photon energy. The variation of the phonon energy is explained by sampling by localised exciton of effective alloys in different wells. Time-resolved PL spectra show that the sharp lines have the same decay time as the broad band luminescence. For each exciting photon energy, the photon energy does not change with the phonon order. This suggests the sharp lines are due to resonant processes. In addition, time-resolved PL spectra reveal an almost constant decay time across the PL band.
    A new theory of the inhomogeneous broadening of the exciton optical transitions is presented. The line shape of exciton absorption is explained by the distribution of the band gap. The line shape of the exciton PL is explained by the distribution of heights (energies) of spatial minima of the band gap, which implies that all PL emission is due to classically localised excitons. A semi-empirical and an analytical approach predict successfully the PL lineshape and peak position for a Gaussian bandgap distribution. Both approaches reveal a quasi-Gaussian PL band, which is consistent with most experimental results. This theory reveals a linear relation between the hh-exciton linewidth Γ and the Stokes' shift Δ: Δ=0.6 Γ. The sharp line structures under resonant excitation are explained by LO-phonon assisted tunneling of exciton between localised states.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2139666550002996}
    }

1991

  • C. Xiaoyuan, "Photoluminescence studies of II-VI compound strained layer superlattices," PhD Thesis, 1991.
    [BibTeX] [Abstract] [Library link]

    This thesis descrides experimental and theoretical studies of the optical properties of semiconductors, in particular binary superlattices of wide bandgap II-IV compounds. A simple equation to the temperature dependence of semiconductor bandgaps is justified on both theoretical and experimental background. The temperature dependence of semiconductor bandgaps follows a universal relationship characterised for each material by an average phonon energy and electron-phonon coupling strength. This coupling strength seems to be uneffected by the quantum confinement in strained layer superlattices. The optical emissions of the superlattices CdS-CdSe, ZnS-ZnSe, ZnSe-CdSe and ZnS-Cds, grown by metal organic vapour phase epitaxy, are investigated for the first time by means of steady-state and time-resolved photoluminescence spectroscopy measurements. It is shown that the excitonic emission of wurtzite CdS-CdSe SLS is consistent with a type II staggered energy band alignment modified by the presence of giant strain-induced piezoelectric fields. The energy position and line shape of the exciton emission are shown to depend largely on the excitation power intensity. Such a excitation dependence of the excitonic emission in these intrinsic Stark superlattices can be described by a model in which bandgap restructuring results from the partial cancellation of the internal fields by electrostatic fields arising from the spatial separation of photoexcited carriers. The dynamics of the bandgap renormalisation are studied in details by means of time-resolved spectroscopy. The exciton lifetimes are estimated to span the range from ns to ms depending on the layer thickness and the exciton population. Quantum fluctuations mainly due to well width fluctuations have large effects on optical properties of ZnSe-CdSe short period SLS. Various features of the optical spectroscopy show that these materials form a model disordered solid, in which the fluctuations form random potential traps to localize excitons to lead to broad exciton emissions, and result in exciton decay with a particular nonexponential pattern. The decay of exciton emission, especially for the long tails, is determined by the carrier hopping processes between the localized sites. A simple expression is found for the decay profile of exciton emission in disordered superlattices. This expression is proven to be consistent with a development of recent theoretical ideas about the motion of charge carriers on random low-dimensional networks. A new experimental parameter which characterises photoluminescence decay in such disordered media is introduced. This tentative investigation establishes a framework for future work.

    @PhdThesis{Xiaoyuan-Chen-thesis,
    author = {Xiaoyuan, Chen},
    school = {University of Strathclyde},
    title = {Photoluminescence studies of {II-VI} compound strained layer superlattices},
    year = {1991},
    abstract = {This thesis descrides experimental and theoretical studies of the optical properties of semiconductors, in particular binary superlattices of wide bandgap II-IV compounds. A simple equation to the temperature dependence of semiconductor bandgaps is justified on both theoretical and experimental background. The temperature dependence of semiconductor bandgaps follows a universal relationship characterised for each material by an average phonon energy and electron-phonon coupling strength. This coupling strength seems to be uneffected by the quantum confinement in strained layer superlattices. The optical emissions of the superlattices CdS-CdSe, ZnS-ZnSe, ZnSe-CdSe and ZnS-Cds, grown by metal organic vapour phase epitaxy, are investigated for the first time by means of steady-state and time-resolved photoluminescence spectroscopy measurements. It is shown that the excitonic emission of wurtzite CdS-CdSe SLS is consistent with a type II staggered energy band alignment modified by the presence of giant strain-induced piezoelectric fields. The energy position and line shape of the exciton emission are shown to depend largely on the excitation power intensity. Such a excitation dependence of the excitonic emission in these intrinsic Stark superlattices can be described by a model in which bandgap restructuring results from the partial cancellation of the internal fields by electrostatic fields arising from the spatial separation of photoexcited carriers. The dynamics of the bandgap renormalisation are studied in details by means of time-resolved spectroscopy. The exciton lifetimes are estimated to span the range from ns to ms depending on the layer thickness and the exciton population. Quantum fluctuations mainly due to well width fluctuations have large effects on optical properties of ZnSe-CdSe short period SLS. Various features of the optical spectroscopy show that these materials form a model disordered solid, in which the fluctuations form random potential traps to localize excitons to lead to broad exciton emissions, and result in exciton decay with a particular nonexponential pattern. The decay of exciton emission, especially for the long tails, is determined by the carrier hopping processes between the localized sites. A simple expression is found for the decay profile of exciton emission in disordered superlattices. This expression is proven to be consistent with a development of recent theoretical ideas about the motion of charge carriers on random low-dimensional networks. A new experimental parameter which characterises photoluminescence decay in such disordered media is introduced. This tentative investigation establishes a framework for future work.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2149448380002996},
    }

  • P. J. Parbrook, "The growth by metalorganic vapour phase epitaxy and charactisation of wide-bangap II-VI epilayers and superlattices," PhD Thesis, 1991.
    [BibTeX] [Abstract] [Library link]

    This thesis describes the growth of the Zn(Cd)S(Se) family of compounds and superlattices by metal organic vapour phase epitaxy (MOVPE). The MOVPE growth of wide-bandgap II-VI compounds is hampered by gas phase prereaction between the group-II precursor, a metalorganic compound, and the group-VI hydride. The novel precursors dimethylzinc-triethylamine and dimethylzinc-triazine strongly inhibit prereaction when growing zinc chalcogenides, and prereaction is also virtually eliminated by adding pyridine to the flow. Mechanisms for this action are discussed. Epilayers were characterised by a variety of techniques. Photoluminescence spectra of zinc chalcogenides feature strong free exciton emission and electrical measurements show the low carrier concentrations and high mobilities characteristic of pure material. The zinc compounds are invariably cubic on (100) GaAs substrates whilst mixed phase growth is obtained for Cd chalcogenides grown on (100) GaAs. Hexagonal CdS and CdSe are obtained on (111)A GaAs. ZnS-Znse, Zns-CdS, ZnSe-CdSe superlattices were grown. Photoluminescence allows an estimate of the conduction band offset between ZnSe and ZnS of -25±meV. The common anion structures ZnS-CdS and ZnSe-CdSe both have large lattice mismatch. A critical thickness of about 1.3nm for a CdSe (CdS) well in a ZnSe (ZnS) lattice was witnessed by rapid decay in exciton photoluminescence for wells exceeding this thickness. Both systems showed strongly blue shifted luminescence compared to the ZnS-ZnSe system suggesting that these structures have large conduction band offsets. Estimates of the band lineup are extremely difficult for these systems. Conduction band offsets of 750±300meV and 850±400meV are estimated for ZnSe-CdSe and ZnS-CdS superlattices respectively. Annealing experiments on the three superlattice systems show that ZnS-ZnSe structures are stable against interdiffusion at high temperatures (up to 550°C). Common-anion superlattices have been shown for the first time to be stable below a growth temperature of 400°C.

    @PhdThesis{Peter-Parbrook-thesis,
    author = {Peter J. Parbrook},
    title = {The growth by metalorganic vapour phase epitaxy and charactisation of wide-bangap {II-VI} epilayers and superlattices},
    school = {University of Strathclyde},
    year = {1991},
    abstract = {This thesis describes the growth of the Zn(Cd)S(Se) family of compounds and superlattices by metal organic vapour phase epitaxy (MOVPE).
    The MOVPE growth of wide-bandgap II-VI compounds is hampered by gas phase prereaction between the group-II precursor, a metalorganic compound, and the group-VI hydride. The novel precursors dimethylzinc-triethylamine and dimethylzinc-triazine strongly inhibit prereaction when growing zinc chalcogenides, and prereaction is also virtually eliminated by adding pyridine to the flow. Mechanisms for this action are discussed. Epilayers were characterised by a variety of techniques. Photoluminescence spectra of zinc chalcogenides feature strong free exciton emission and electrical measurements show the low carrier concentrations and high mobilities characteristic of pure material. The zinc compounds are invariably cubic on (100) GaAs substrates whilst mixed phase growth is obtained for Cd chalcogenides grown on (100) GaAs. Hexagonal CdS and CdSe are obtained on (111)A GaAs.
    ZnS-Znse, Zns-CdS, ZnSe-CdSe superlattices were grown. Photoluminescence allows an estimate of the conduction band offset between ZnSe and ZnS of -25±meV. The common anion structures ZnS-CdS and ZnSe-CdSe both have large lattice mismatch. A critical thickness of about 1.3nm for a CdSe (CdS) well in a ZnSe (ZnS) lattice was witnessed by rapid decay in exciton photoluminescence for wells exceeding this thickness. Both systems showed strongly blue shifted luminescence compared to the ZnS-ZnSe system suggesting that these structures have large conduction band offsets. Estimates of the band lineup are extremely difficult for these systems. Conduction band offsets of 750±300meV and 850±400meV are estimated for ZnSe-CdSe and ZnS-CdS superlattices respectively. Annealing experiments on the three superlattice systems show that ZnS-ZnSe structures are stable against interdiffusion at high temperatures (up to 550°C). Common-anion superlattices have been shown for the first time to be stable below a growth temperature of 400°C.},
    url = {http://suprimo.lib.strath.ac.uk/SUVU01:LSCOP_SU:SUALMA2145263100002996}
    }