This thesis describes an experimental study of the electronic properties of the dilute nitride GaAs1¡xNx alloy. This is a semiconductor belonging to a class of highly mismatched III-N-V alloys. The incorporation of isoelectronic N on the pnictide (e.g. As) site of GaAs gives rise to a highly localised electronic state, whose energy level is resonant with the continuum of conduction band (CB) states of the host GaAs lattice. The interaction between these two sets of states causes the formation of a fully developed energy gap in the CB of the host crystal and makes possible the observation of a novel type of negative differential conductance (NDC) effect. The NDC in GaAs1¡xNx is qualitatively different from the NDC occurring in transferred electron devices (Gunn diodes) and semiconductor superlattices (SLs) and has potential for novel terahertz (THz) device applications. The emphasis of the thesis is on the experimental study of the non-linear electron dc dynamics in GaAs1¡xNx that arises when electrons are accelerated in the non-parabolic CB of GaAs1¡xNx. It also includes an investigation of the coupling of electrons to THz radiation by the measurement of harmonic generation of ac current and of changes in the dc conductivity in the presence of an applied THz radiation. The rectification effects revealed in our experiments indicate that the mechanism giving rise to NDC is a fast ( 10¡12 s) process. The fast response in time of the current is in agreement with previous calculations of the ac electron dynamics in GaAs1¡xNx predicting that the maximum response frequency associated with the NDC is governed by the time of ballistic acceleration of electrons to the N-level and that this lies in the THz frequency range. The experimental results are discussed in terms of different theoretical models and mechanisms, including the band anticrossing model, space-charge-limited current instabilities, magnetophonon resonance and classical rectification theory.

The characterisation, emulation and by-emitter degradation analysis of two types of high power semiconductor laser diodes are presented in this thesis as part of an European Union (EU) project. An attempt is made using an accurate laser simulator called Speclase to learn more about the degradation of high power semiconductor laser diodes. Speclase being a single emitter simulation tool was transformed to model a bar i.e. multiple emitters, which we have named Barlase, through an external control interface written in Labview. The concept of Barlase was based on the fact that a bar is a monolithic block of multiple emitters connected in parallel with each other with a common voltage connected across them. This tool is capable of performing simulation in different modes of operation (i.e. constant current or constant power). The tool is designed to examine and emulate the degradation processes at both the laser bar and individual emitter levels of operation. It is known that, emitter degradation is faster for emitters within a bar than for identical single emitters due to a combination of packaging-induced strain and current competition between emitters amongst others. This tool shows clear evidence of the benefits of using by-emitter degradation analysis for gaining detailed understanding of individual emitters operating in a bar and for determining bar degradation mechanisms. The tool complement to the by-emitter analysis, allowing the effects of certain factors that affect the degradation of laser bars to be investigated. Various intervention measures were taken to improve upon the results of the emulation such as modifying the trap density through local heating and the use of the global thermal solver. The modification of the trap density allowed the acceptance of a spatially variable local trap density distribution that gave a more realistic and accurate simulation of the degradation behaviour. The introduction of the global thermal solver allowed the modelling of thermal cross-talk communication between the emitters, which brings about the frown shaped current/power profiles for the unaged bars (though not as pronounced as in the experiment). An attempt was made to employ this tool in the emulation of experimentally observed degradation behaviour in a 975 nm, 16 emitter infrared tapered laser bar with each group of 4 mini-array emitters. The laser bar was first calibrated to achieve a reasonable agreement between the experimental P-I curves of unaged emitters assuming identical emitters with the simulated P-I curves. The simulated P-I curve was then used to perform simulations to emulate the degradation of the laser. The simulated output power profile did not correspond well with the experimental power profile, but a good agreement was realised between the combined output powers of the bar. Better correlation was observed between the experimental and the simulated temperature profiles. This was expected since the experimental temperature was set as input for the heatsink temperature profile. This agreement therefore must not be over-emphasised. The bar emulation model was enhanced by including a global thermal solver to model the thermal crosstalk between emitters. Emulations using this model showed a clearly defined frown shaped profile in the output current and power profiles but the change was minimal. As the emulation of laser bar degradation has not been attempted before, this work is still at a very early stage. Therefore, further work is needed to achieve better agreement in the output current/power profiles and to better the model.

In this thesis first-principles calculations of point and extended defects in diamond and silicon are reported. In single crystal diamond grown by chemical vapour deposition (CVD) dislocations are observed as mixed-type 45° and edge-type dislocations lying along <100> with 1/2<110> Burgers vectors. Results are presented on the core structures, core energies and electrical properties of both types of dislocations and their interaction with nitrogen is investigated. Then the focus turns to the brown diamond problem. Despite concerted research efforts, the origin of the brown colouration of diamond is still under discussion. Recently, the attention was drawn to vacancy-related defects. Experiments on type IIa diamonds indicate that the brown colour is caused by vacancy-type extended defects, however the shape and size of these defects remained unclear. In this work, the structural, electrical and optical properties of large spherical vacancy clusters and thin vacancy disks are investigated by means of density functional theory and the calculations are compared with recent experimental measurements on brown diamond. High pressure high temperature treatment (HPHT) of brown type Ia diamonds above 2000°C results in the loss of the brown colour and the formation of nitrogen-vacancy defects. The generation of such defects requires a source of mobile vacancies during the annealing process. It is suggested that the vacancy cluster model described in this thesis can explain the observed annealing behaviour since the break-up of the clusters leads to a supersaturation of mobile vacancies which readily complex with substitutional nitrogen atoms present in the material. Therefore, the effect of HPHT treatment of brown type Ia diamond is investigated by studying the formation energies of common and rare defects and estimates of their equilibrium concentrations at different annealing stages are given. Finally, an open problem also involving nitrogen, but in a different group IV semiconductor is considered. In Czochralski-silicon, nitrogen-related shallow thermal donors are formed between 500 and 750°C. Until now the exact chemical composition and atomic structure of these defects are not well established. Here, it is shown that NO and NO_2 belong to the family of nitrogen-oxygen related shallow thermal donors. Based on the law of mass action the equilibrium defect concentrations are predicted. Finally, the theoretical results are compared to recent Fourier transform infrared (FTIR) spectroscopy measurements.

Since the publication of research in the mid-1980s describing the formation of freeform graphene there has been an enormous growth in interest in the material. Graphene is of interest to the semiconductor industry because of the high electron mobility exhibited by the material and, as it is planar, it is compatible with silicon technology. When patterned into nanoribbons graphene can be made into regions that are semiconducting or conducting and even into entire circuits. Graphene nanoribbons can also be used to form the channel of a MOSFET. This thesis describes numerical simulations undertaken on devices formed from graphene. The energy band structure of graphene and graphene nanoribbons is obtained using nearest-neighbour and third nearest-neighbour interactions within a tight binding model. A comparison of the current-voltage characteristics of MOS structures formed on graphene nanoribbons and carbon nanotubes suggests that the nanoribbon devices may be better for switching applications. Conductivities of graphene nanoribbons and junctions formed from them were obtained using a nonequilibrium Green’s function formulation. The effects of defects and strain on these systems were also studied using this technique. Advancements were made when the self-energies used within the nonequilibrium Green’s function were obtained from an iterative scheme including third nearestneighbour interactions. An important result of this work is that accurate simulations of graphene based devices should include third nearest-neighbour interactions within the tight binding model of the energy band structure.

To enable the advancement of Si based technology, necessary to increase computing power and the manufacture of more compact circuits, significant changes to the current planar transistor are a necessity. Novel transistor architectures and materials are currently being researched vigorously. This thesis, on the electrical characterisation of non-standard orientated MOSFETs and multi-gate transistors displays detailed insight into the carrier transport and resulting performance limiting mechanisms. The results are composed of three parts. Firstly, the standard method of extracting carrier effective mobility from electrical measurements on MOSFETs is reviewed and the assumptions implicit in this method are discussed. A novel technique is suggested that corrects the difference in drain bias during current-voltage and capacitance-voltage measurements. It is further shown that the lateral field and diffusion corrections, which are commonly neglected, in fact cancel each other. The efficacy of the proposed technique is demonstrated by application to data measured on a quasi-planar SOI finFET at 300 K and 4 K. The second part is based on the electrical characterisation of n+poly-Si/SiO2/Si nand p- MOSFETs fabricated on (100) and (110) substrate orientations with the full range of channel directions. In depth analysis of the electron and hole mobility was performed at 300 K and 4 K. The 4 K mobilities were modelled in terms of ionised dopant impurity, local SiO2/Si interface charge and roughness scattering mechanisms. RMS (root mean squared) roughness values in the range 0.34 − 0.38nm and correlation lengths of 2.0 − 2.3 nm were extracted revealing comparable interface quality between the (100) and (110) surfaces. The third part examines the electrical characterisation of TiN/HfSiO2/Si n- and pfinFETs. Fin top surface and sidewalls are in the (100) and (110) planes respectively. Fins have a height of 65 nm with widths in the range of 1872 nm (quasi-planar) to 12 nm. Detailed analysis revealed vertical compressive strain induced by the gate into the fin sidewalls, which enhanced the electron mobility by 60% above the (110) reference, whilst leaving the hole mobility unaffected. Qualitative analysis of the 4 K mobilities suggests that roughness is higher on the sidewalls than on the top surface. This was attributed to the damage caused by the dry etch, used to pattern the fins. A model for remote charge scattering at the HfSiO2/SiO2 interface was developed. 4 K mobilities from the quasi-planar n- and pfinFETs were then modelled in terms of remote charge, ionised dopant impurity, local SiO2/Si interface charge and roughness scattering mechanisms. Remote charge densities of 8x1012 cm-2 were subsequently extracted. Scattering from these charges was shown to be the dominant scattering mechanism in the quasi-planar n-finFET mobility at 300 K.

- / -

- / -

Orientador : Dante Homero Mosca Jr / Dissertaçao (mestrado) - Universidade Federal do Paraná

Research in silicon photonics has recently seen a significant push to develop complete silicon-based optical components for optical communications. Silicon has shown its potential to overcome the bandwidth limitations of microprocessor interconnect, whereas, the silicon platform has already displayed the benefits of low manufacturing costs and CMOS compatibility. The work on “gated lateral silicon p-i-n junction photodiodes” has demonstrated the silicon potential, to detect optical radiations, compatibility to standard CMOS process flow and tuneable spectral response. The lateral structure of gated p-i-n junction photodiodes contributes to high responsivity to short wavelength radiations in these single and dual gate devices. The final objective of this work was to develop high responsivity, CMOS-compatible silicon photodiodes, where the spectral response can be modulated. The lateral p-i-n junction architecture led to high responsivity values, whereas, the MOS gate structure became the basis for tuneable spectral response. The MOS gate structure, made the devices appear as a transistor to the surrounding circuitry and the gate structure in dual gate devices can be used to modulate the spectral response of the device. Single gate devices showed higher responsivity values and comparatively high blue and ultraviolet (UV) response as compared to conventional photodiodes. Surface depletion region in these devices is utilized by placing a MOS gate structure and by patterning an integrated metal grating to detect polarized light. Single and dual gate devices with two variations were fabricated to characterise the device response. Novel lateral architecture of p-i-n junction photodiodes provides a surface depletion region. It is generally anticipated that photodetectors with surface depletion region might produce higher noise. In these devices the surface depletion region has a lateral continuation of gate dielectric which acts as a passivation layer and thus considerably reduced the noise. Physical device modelling studies were performed to verify the experimentally obtained results, which are provided in the relevant measurement chapters. In these devices the speed of operation is a compromise over the high responsivity, CMOS compatibility and tuneable spectral response.

Self-heating can have a detrimental effect on the performance and reliability of high power microwave devices. In this work, the thermal performance of the gallium arsenide (GaAs) Gunn diode was studied. Infrared (IR) thermal microscopy was used to measure the peak operating temperature of the graded-gap structured device. Temperature measurements were experimentally validated using micro-thermocouple probing and compared to values obtained from a standard 1D thermal resistance model. Thermal analysis of the conventionally structured Gunn diode was also undertaken using high resolution micro-Raman temperature profiling, IR thermal microscopy and electro/thermal finite element modeling. The accuracy of conventional IR temperature measurements, made on semiconductor devices, was investigated in detail. Significant temperature errors were shown to occur in IR temperature measurements made on IR transparent semiconductors layers and low emissivity/highly reflective metals. A new technique, employing spherical carbon microparticles, was developed to improve the measurement accuracy on such surfaces. The new ‘IR microparticle’ technique can be used with existing IR microscopes and potentially removes the need to coat a device with a high emissivity layer, which causes damage and heat spreading.