Search

This thesis describes the designs and measurements of three nano-photonic device schemes. The aim of which is to achieve the scale-up of III-V semiconductor based nano-photonic systems. Scaleup of the number of resonant quantum dots, and nano-structures on-chip is vital for the realisation of a quantum network using the III-V semiconductor quantum dot architecture. The effect of varying the growth conditions of site-controlled quantum dots on their optical properties is investigated. This is combined with a study of a recently demonstrated in-situ local quantum dot strain tuning method. Optical waveguides are conformally coated in a dielectric material which when crystallised by a heating laser, induces a strain field in the quantum dots embedded in the waveguide. An electro-mechanically tunable photonic crystal cavity design is proposed. The cavity and cantilever designs are optimised and tested for robustness against fabrication imperfections in simulation. The cavity design exhibits world-leading figures of merit for both the weak and strong coupling regimes. The cavity fabrication was optimised through several test sample iterations. The final design was fabricated, and a voltage-controllable cavity wavelength tuning range much larger than the standard deviation of the cavity mode wavelength distribution was measured. A scheme which enables a quantum dot to exhibit both strong light-matter interaction enhancement, and strongly directional emission is presented. The design consists of a photonic crystal cavity coupled to two feeding photonic crystal waveguides. The operation of the design is verified through a theoretical model and simulations. A test device sample is measured to characterise various waveguide orientation designs and parameters. Strongly directional emission and a low Purcell enhancement are measured from a quantum dot embedded in a waveguide coupled cavity.

This thesis is trying to explore a possible routine to manufacture flexible sensor via vacuum-based processes. It starts with an exploration of a route to manufacture Organic Thin Film Transistors (OTFTs) and then a series of stabilization methods are discussed to reduce the probability of short circuit, current leakage and improve the performance of OTFTs, i.e., increasing mobility and decreasing threshold voltage. In addition, a possible mechanism of bias stress is investigated with a model and the relevant evolution of trapping state and trapping energy is plotted over time and distance from the dielectric/semiconductor interface. Next, a short-time bias scheme is demonstrated to reduce error of bias stress during strain tests. A first example of a strain sensor is based on the architecture of floating gate OTFTs, i.e., a PVDF coated floating gate that transfers the signal of charge imbalance from the ferroelectric effect to transistors, and then the OTFT will output the amplified signal. The advantage of floating gate structure is the separation of the transistor and sensor part to avoid interference between transistor and sensing reaction and giving the option for the transistor to be encapsulated from the environment. However, the bias voltage is a key parameter to control the floating gate sensor, with the sensitivity of the device dependant on it, i.e., it is very difficult to obtain both a sensitive signal at same time as one unaffected by bias stress. In order to obtain a sensitive and stable signal, a series of OTFT-based functional circuits are developed and a transimpedance converter is fabricated to convert nano-scale input current to a readable voltage output (0 - 3V). After encapsulation, the linear converting relationship is maintained and the mechanical flexibility is not affected. Finally, a series of sensors are built with the OTFT-based signal conditioning circuit to detect arterial pulse from human body, pH, hydrogen peroxide and glucose from solution. Conclusively, this thesis demonstrates a new architecture of OTFT-based sensing system, and that can work as a versatile platform to integrate with different functionalized sensing electrodes to process signals to a readable level and visualized with smartphone via Bluetooth microcontroller.

Magnetic exchange spring systems which consist of alternating thin hard and soft magnetic layers provide a great potential in applications such as (i) permanent magnets and (ii) magnetic data storage. The molecular beam epitaxy-grown samples provide an excellent model system with which to study the magnetic switchings and reversals. X-ray magnetic circular dichroism (XMCD) allows element-selective hysteresis loops with the results presented alongside bulk magnetometry and micromagnetic modelling. This gives a complete picture of REFe2 magnetic exchange springs. This work unveiled the spin-flop and transverse exchange springs in the ErFe2/YFe2 and DyFe2/Fe2 thin film multilayers respectively. This work has also focused upon studying the magnetisation dynamics of exchange-coupled bilayers. Ferromagnetic resonance spectroscopy gives multiple resonant modes of the NiFe/CoFe exchange-coupled bilayer. X-ray detected ferromagnetic resonance was used to investigate the relationship between the magnetic layers by using XMCD for the magnetic contrast. How the anisotropy of one layer can influence the second, as well as determination of the origin of the two resonant modes are discussed. A new microwave device is the final aspect of this work. The new topic of metamaterials was studied and opened up new research directions. This work demonstrates that tunability of the metamaterial resonance by applied magnetic field is possible by coupling the resonant split ring resonator structure to magnetic thin films, providing large modulations in transmitted microwave power previously unattained.

As silicon MOSFETs have reached their saturation limits due to aggressive scaling, high mobility channel materials like germanium are offering a future generation of semiconductor materials. Germanium's smaller band-gap broadens the optical absorption spectrum, and its low melting point considerably reduces the process temperature compared with silicon. However its unstable native oxide, large diffusion constant of n-type dopants, high series resistance of metal contacts on germanium, susceptibility to implantation damage, Fermi-level pinning of metal contacts on n-type source/drain regions, poor-quality interfaces between germanium and high-k dielectrics, etc. are compromising the performance of germanium MOSFETS. Implant-less nickel germanide source and drain p-channel MOSFETs were successfully fabricated at a low thermal budget. A effective hole mobility of -340 cm2/Vsec was measured on nickel germanide source/drain p-MOSFETs with Hf02/GeOxNy as gate dielectric. The Fermi level of aluminium contacts on germanium was successfully unpinned by using atomic layer deposited layers. A barrier height from ~O.7e V for Al contacts on n-Ge to a value of 0.28 e V was realized with insertion of a thin alumina interfacial layer (28.5 A)

Silicon Photonics has been a major success story in the last decade, with many photonic devices having been successfully demonstrated. The only missing component is the light source, however, as making an efficient light source in silicon is challenging due to the material's indirect bandgap. The development of a silicon light source would enable us to make an all-silicon chip, which would find many practical applications. The most notable among these applications are on-chip communications and sensing applications. In this PhD project, I have worked on enhancing silicon light emission by combining material processing and device engineering methods. Regarding materials processing, the emission level was increased by taking three routes. In all the three cases the emission was further enhanced by coupling it with a photonic crystal (PhC) cavity via Purcell effect. The three different approaches taken in this PhD project are listed below. 1. The first approach involves incorporation of optically active defects into the silicon lattice by hydrogen plasma treatment or ion implantation. This process results in broad luminescence bands centered at 1300 and 1500 nm. By coupling these emission bands with the photonic crystal cavity, I was able to demonstrate a narrowband silicon light emitting diode at room temperature. This silicon nano light emitting diode has a tunable emission line in the 1300-1600 nm range. 2. In the second approach, a narrow emission line at 1.28µm was created by carbon ion implantation, termed “G-line” emission. The possibility of enhancing the emission intensity of this line via the Purcell effect was investigated, but only with limited success. Different proposals for future work are presented in this regard. 3. The third approach is deposition of a thin film of an erbium disilicate on top of a PhC cavity. The erbium emission is enhanced by the PhC cavity. Using this method, an optically pumped light source emitting at 1.54 µm and operating at room temperature is demonstrated. A practical application of silicon light source developed in this project in gas sensing is also demonstrated. As a first step, I show refractive index sensing, which is a simple application for our source and demonstrates its capabilities, especially relating to the lack of fiber coupling schemes. I also discuss several proposals for extending applications into on-chip biological sensing.

Quantum dot structures of InAs(Sb)/InGaAs/InP designed as easy to fabricate, low cost mid-IR emitting lasers, have been spectroscopically characterised using temperature and power dependent photoluminescence. These structures have been simulated using a truncated pyramid structure in the Nextnano software package. The results show that the observed experimental data is the result of a bimodal dot distribution in both samples. In the InAs case, the bimodal behaviour is the result of varying width dots (35nm and 38.5nm). In the InAsSb case the dot groups were calculated to contain ~10% and zero antimony, indicating difficulties during the growth process. Additionally the InAs dots were found to have a dominant radiative recombination process, while the InAsSb dots were found to be affected by a defect related recombination process. It is suggested this is a result of increased defects formed by the larger lattice mismatch. InAs/InAsSb superlattice structures have potential as mercury cadmium telluride (MCT) alternative mid-IR photo-detectors, and are predicted to not suffer from Ga-related defect recombination as other superlattice structures. High pressure techniques and modelling were used to probe the defect level in these structures. High pressure, low temperature photoluminescence experiments were performed using the sapphire ball cell to move the conduction band minima up in energy until overlap with the predicted defect level state was achieved. This resulted in a decrease in the measured integrated intensity of the sample due to carriers recombining via the defect states. Additionally power dependent measurements at high and low pressure were performed and an observed shift from radiative to defect dominated recombination was observed. This provides the first experimental evidence of a defect level positioned above the conduction band edge. This means that SRH recombination in the forbidden band gap will not be a contributing factor to the dark currents in InAs/InAsSb superlattice photo-detectors showing their promise for low dark current mid-IR detectors.

We study the low-temperature physics arising in models of a strongly correlated, tunnel-coupled double quantum dot (DQD), particularly the two-impurity Anderson model (2AIM) and the two-impurity Kondo model (2IKM), employing a combination of physical arguments and the Numerical Renormalisation Group. These models exhibit a rich range of Kondo physics. In the regime with essentially one electron on each dot, there is a competition between the Kondo effect and the interdot exchange interaction. This competition gives rise to a quantum phase transition (QPT) between local singlet and Kondo singlet phases in the 2IKM, which becomes a continuous crossover in the 2AIM as a result of the interlead charge transfer present. The 2IKM is known to exhibit two-channel Kondo (2CK) physics at the QPT, and we investigate whether this is also the case for the 2AIM at the crossover. We find that while in principle 2CK physics can be observed in the 2AIM, extremely low temperatures are required, such that it is unlikely that 2CK physics will be observed in an experimental DQD system in the near future. We have studied the effect of a magnetic field on the 2AIM and the 2IKM, finding that both the zero-field QPT in the 2IKM and the zero-field crossover in the 2AIM, persist to finite field. This presents the possibility of observing 2CK physics in an experimental DQD at finite field, but we find that the temperatures required to do so are extremely low. We show that longer even-numbered chains of spins also exhibit QPTs at finite field, and argue that a 2N-spin chain should undergo N QPTs as field is increased (starting deep in the local singlet phase at zero field). We have also carried out a joint theoretical-experimental study of a carbon nanotube based DQD, in collaboration with Dr. Mark Buitelaar et al. The agreement between experimental and theoretical results is good, and the experiments are able to access the crossover present in the 2AIM at finite field. Furthermore, the experiments show the wide range of physics exhibited by DQD systems, and illustrate the utility of such systems in probing correlated electron physics.

This thesis presents intraband optical studies of self-assembled quantum dots (QDs) in the few terahertz (THz) range and optical THz frequency mixing in the same samples. Magneto-transmission measurements of QDs in the few-terahertz range are described. The polaronic nature of the excitations of the system was studied by tracking the intraband transition energy as a function of magnetic field up to 28T. The first direct measurement of anticrossings in the I-LO phonon energy range are reported and evidence of polarons consisting of an electron bound with an InAs-like optical phonon in addition to phonon-like polarons is presented. Degenerate pump-probe spectroscopy was used to study intraband energy relaxation for transition energies below the LO phonon energy. Post-growth annealing was used to reduce the energy level separation of the samples to below the typical as-grown transition energy of 40 - 60me V. It was observed that as the intraband transition energy reduced, the relaxation times increased, exhibiting very strong energy dependence. It was found that at an energy of ~ 30meV (7THz) the relaxation times were of the order of 2ps, whilst at an energy of ~ 14meV (3.4THz) the relaxation times increased to 1.5ns. These results were explained using a microscopic polaron relaxation model (developed by colleagues at the ENS, Paris), in which decay occurs via phonon anharmonicity, with very good agreement found between experiment and theory. Optical sideband generation via frequency mixing of a near infrared (NIR) laser beam and a THz beam using interband and intraband resonances in QDs is described.

This thesis scrutinizes the condensed matter physics phenomenon known as the fractional quantum Hall effect (FQHE), in particular fractional quantum Hall effects occurring in multicomponent systems. Broadly speaking, the FQHE can be defined as a many-electron quantum phenomenon, driven by strong interactions, that occurs in two-dimensional electron gasses in the presence of a perpendicular external magnetic field (and it is also predicted to occur for any two-dimensional particles, such as confined cold atoms, in an external gauge field). Multicomponent systems are systems where the constituent particles (such as electrons or cold atoms) possess internal degrees of freedom, for instance a spin or valley index. These internal degrees of freedom are often overlooked when modeling the FQHE. Taking into account the multicomponent degree of freedom yields an abundance of possibilities for the intellection of new types of so-called “topological phases of matter”, which are ubiquitously associated with the FQHE. In this thesis several different cases are considered. The first topic discussed herein is a study of phase transitions that can take place between FQHE phases with different net values of their multicomponent degrees of freedom. Examples are phase transitions between phases of different uniform net spin polarization, tunable as a function of certain system parameters. Some significant technical refinements are made to a previous model and comparisons are made with a variety of different experiments. The results are relevant for multicomponent FQHEs occurring in GaAs,AlAs and SiGe semiconductor systems where the electronic structure is confined to two dimensions, as well as in two-dimensional materials such as graphene. The second topic discussed herein is the introduction of the multiparticle multicomponent pseudopotential formalism. This methodology is oriented towards considerably expanding an existing framework for the construction of exactly solvable FQHE models by parameterizing multicomponent interactions. The final topic is the first example application of this new formalism to the construction of an exactly solvable FQHE model.

Magneto optical studies of colloidal PbS nanocrystals (NCs) have been undertaken to determine their fundamental properties. Measurements including absorption, photoluminescence (PL) and PL lifetime decay are presented along with their dependence upon temperature, magnetic field, and magnetic dopant concentration for undoped and doped PbS NCs. Temperature dependence of undoped PbS NCs, recorded from 300 K down to 3 K, displays a Stoke shift increasing from ~75 meV to -125 meV which is fitted using a three-level rate equation model, supported by PL lifetime decay measurements, that indicate energy separation of ~6.0+/-0.3 meV between the two optically active levels within PbS NCs. Magneto optical studies of undoped PbS NCs at low temperature using a field sweep from -7 Tesla to 7 Tesla are presented. Analysis of the magneto-PL data yields a degree of circular polarization (DCP) of ~33% at 7 T and 2 K. Further analysis predicts an excitonic g-factor, gex for the ~4 nm diameter PbS NCs of ~0.54 by taking account of random orientation of PbS NCs. Using this value of gex the expected Zeeman splitting at 7T, AEZeeman, is calculated to be ~0.22 meV. Optical studies of PbS NCs with TTF/TCNQ molecule showed modification of the PL spectra and PL lifetime. It is proposed that the quenching effect on the PL of PbS-TTF in the range of 900 nm to 1300 nm is due to a charge transfer mechanism. The PL obtained from PbS-TCNQ solutions display a second emission peak centred at ~700 nm which is directly related to the TCNQ concentration. A study of the optical properties of Mn-implanted PbS NCs was undertaken and compared with that of undoped PbS NCs. The PL spectra of all Mn-implanted PbS NCs showed a significant red shift of the PL peak compared to undoped PbS NCs. Based on fitting of a Brillouin function to the difference in the temperature dependent Stokes shift between the Mn-implanted and undoped PbS NCs an effective exchange field of ~81 T is predicted.

In this thesis I will investigate photonic structures that can be utilised to entangle the electron spin of a self-assembled quantum dot with a photon. Charged. quantum dots, whose emitted photons are trapped or guided by the photonic structure, can be utilised as quantum memories and form the basic building block of the interface. It will be shown, using a dyadic Green's function method, that photonic structures need to fulfil! two criteria in order to implement a spin-to-photon interface. Criterion one is the support of a degenerate mode, criterion two requires the quantum dot to couple equally to the two orthogonal modes. Only a subset of photonic structures fulfill these requirements. Two structures that fulfill these criteria are investigated in more detail using the dyadic Green's function method developed, and I show that both structures enable in-plane spin readout, which had not been demonstrated yet. The first, a cross configuration of two nanowire waveguides, couples a spin state to different waveguides of the cross configuration. By recombining two of the four arms and interfering the outputs spin readout is made possible. It will be shown that the device has two working regimes for spin readout depending on the quantum dot position. The structure works well in both regimes. The second structure, a photonic crystal waveguide, enables in-plane spin readout by positioning the quantum dot on a circular polarisation singularity resulting in directional coupling of different spin states. Subsequently, I will show that this structure is able to generate spin-photon and photon-photon entanglement by reflecting photons of the waveguide.

A consistent trend in the semiconductor industry has been the increase of embedded functionality in new designs. As a result, the verification process today requires significant resources to cope with these increasingly complex designs. In order to alleviate the problem, industrialists and academics have proposed and improved on many formal, simulation-based and hybrid verification techniques. To dale, none of the approaches proposed have been ab le to present a convincing argument warranting their unconditional adoption by the industry. In an attempt to further automate design verification (DV), especially in simulation-based and hybrid approaches, machine learning (ML) techniques have been exploited to close the loop between coverage feedback and test generation ; a process also known as coverage directed test generation (COG). Although most techniques in the literature are reported to help in constructing minimal tests that exercise most, if not the entire design under verification, a question remains on their practical usefulness when applied in real-world industry-level verification environments. The aim of this work was to answer the following questions: I. What would constitute a good ML-COG solution? What would be its characteristics? 2. 00 existing ML-CDG technique(s) scale to industrial designs and verification environments? 3. Can we develop an ML-based system that can attempt functional coverage balancing? This work answers these questions having gathered requirements and capabilities from earlier academic work and having filtered them through an industrial perspective on usefulness and practicality. The main metrics used to evaluate these were effectiveness in terms of coverage achieved and effort in terms of computation time. A coverage closure effective and easy to use genetic programming technique has been applied on an industrial level verification project and the poor results obtained show that the particular technique does not scale well. Linear regress ion has been attempted for feature extraction as part of a larger and novel stochastic ML-CDG model. The results on the capability of these techniques were again below expectations thus showing the ineffectiveness of these algorithms on larger datasets. Finally, learning classifier systems, specifically XCS, have been used to discover the cause-effect relationships between test generator biases and coverage. The results obtained pointed to a problem with the learning mechanism in XCS, and a misconception held by academics on its capabilities. Though XCS at its current state is not an immediately exploitable ML~CDG technique, it shows the necessary potential for later adoption once the problem discovered here is resolved through further research. The outcome of this research was the realisation that the contemporary ML methodologies that have been experimented with fall short of expectations when dealing with industry-level simulation-based digital design verification. In addition, it was discovered that design verification constitutes a problem area that can stress these techniques to their limits and can therefore indicate areas for further improvement and academic research.

In recent years CdZnTe has been identified as an attractive alternative semiconductor material for the fabrication of radiation detectors. However, the poor hole transport properties have forced researchers to turn towards pixelated detectors in order to utilise the small pixel effect and minimise the contribution of holes in the output signal. This project focuses on the study of charge sharing and how the performance of a pixelated detector can be improved with the addition of a steering electrode. Silvaco ATLAS, a simulation software package, is used in this work along with the PIXIE ASIC to study these effects theoretically and experimentally. Two different detectors are used our measurements, which include general spectroscopy and collimated X-ray beam experiments. The simulations showed that the use of a steering electrode changes the electric field and potential distributions, which should reduce any charge sharing or charge loss. It is important however, for the steering electrode to have the right bias based on the anode geometry and also not to be shorted to an anode pixel. If the latter occurs then there is significant charge loss through the steering electrode. The two detectors demonstrated good energy resolution for the largest of the four arrays included in the PIXIE anode pattern. It was also shown that the amount of charge sharing is increased with decreasing pixel size and pitch. The performance of pixelated detectors is affected by pixel gain matching and the cathode bias, which determines the charge cloud diffusion and charge collection efficiency. The collimated X-ray beam experiments showed that pixelated detectors suffer from charge loss in the inter-pixel gap and that the cathode bias plays a bigger role in the amount of charge sharing than the incident radiation energy. Finally, our results show that care needs to be taken when devices are bonded to the ASIC, because when the steering electrode is shorted to an anode pixel the performance of the detector drops significantly and the amount of charge loss in the inter-pixel gap can be as high as 80%.

Nanostructured metal films were electrodeposited through the hexagonal lyotropic liquid crystalline phase (HI). The mesoporous structure consists of porous channels (a few nm in diameter) arranged in an hexagonal array. These mesoporous metal films exhibit high surface areas supplied by the concave surface within the pores. The properties of these mesoporous materials have been investigated to gain an insight on the mesoporous structure. Cyclic voltammetry in acid of HI mesoporous Pt is similar to polycrystalline Pt made up of low index Pt facets. However CO stripping voltammetry shows differences between the HI mesoporous Pt and polished Pt electrodes. The CO stripping voltammogram for the HI mesoporous Pt electrode exhibits a CO oxidation pre-wave and CO oxidation at lower overpotentials. These differences result from the presence of trough sites corresponding to the intersection of two pore walls within the mesoporous structure. The adsorption of foreign atoms Bi and Ge on HI mesoporousPt was investigated to identify the different crystalline Pt facets. The features of the voltammetric profiles recorded in acid revealed the absence of large (111) domains and the presence of (100) terraces sites. CO stripping voltammetry for HI mesoporous Pt modified with Bi suggests the presence of a CO-Bi mixed ad layer. However, the absence of the aforementioned pre-wave was attributed to the adsorption of Bi on the trough sites thus causing inefficiency in oxygen transfer for CO oxidation. In contrast, the significant pre-wave observed for HI mesoporous Pt-Ge leads to an enhancement for CO oxidation. HI mesoporous metal films grown on microelectrodes show good stability of the measurement of hydrogen peroxide. HI mesoporous Rh with a variety of film thicknesses was extensively studied over a wide range of hydrogen peroxide concentrations. A kinetic model is proposed to describe the diffusion of hydrogen peroxide and the surface reaction in the pore. The accessibility of the ionic liquid BMIM-PF6 in the pores was investigated to assess the performance of mesoporous electrodes in supercapacitors. HI mesoporous Pt of diverse pore size and polished Pt electrodes were characterised in 1 M sulphuric acid and then tested in BMIM-PF6. The double layer capacitances were higher for the HI mesoporous Pt films thanks to their internal surface area leading to the confirmation that the ionic liquid penetrates into the pores. The analysis of electrochemical impedance spectroscopy shows that the results fit a transmission line model and provides useful parameters for the characterisation of the nanostructured Pt film in BMIM-PF6. 11.

This thesis describes the optical and electrical properties of colloidal PbS nanocrystals synthesized in aqueous solution and comprising different capping ligands and/or matrices. Post-synthesis thermal annealing of thiol-capped PbS colloidal quantum dots (QDs) is discussed and shown to provide a novel means of tailoring the morphological and electronic properties of the QDs. Two different regimes are reported: at low annealing temperatures (< 80oC), the annealing provides a simple strategy for controlling and narrowing the QD size distribution and photoluminescence emission. At higher annealing temperatures, the displacement of the thiol-ligands promotes the fusion of nearby quantum dots thus leading to interconnected nanocrystals. A study of the circularly polarized magneto- photoluminescence of colloidal PbS nanocrystals under the influence of a magnetic field up to 30T is presented. A semiclassical model for the population of polarized excitons is used to account for the measured magnetic field and temperature dependence of the degree of circular polarization of the QD photoluminescence. The g-factor, gX, of the exciton and its dependence on the QD size is reported for the first time. The value of gX increases from 0.1 to 0.3 at low temperature with decreasing the nanocrystal diameter from 9 to 4 nm. The transport properties of a PbS QDs thin film deposited between two electrodes are investigated. This study reveals a non-linear dependence of the current on the applied bias. At low temperature (T<100 K), the conduction is limited by the charging energy (~20 meV) of the quantum dot. The fabrication of a solid-state device based on porous TiO2 impregnated with PbS QDs is discussed. The photovoltaic response of the device in the visible and near infrared wavelength range is obtained by exploiting the Schottky junction that forms at the interface between the PbS/TiO2 film and a metallic contact.

Accurate control and addressability of quantum devices will come with the promise of improvement in a wide variety of theoretical and applied fields, such as chemistry, condensed matter physics, theoretical computer science, foundational physics, communications, metrology and others. Decoherence of quantum states and the loss of quantum systems have adverse effects and deter a satisfactory usage of quantum devices. This is the main problem to be overcome, which is the goal of quantum fault tolerance. In this thesis we present a series of works that contribute to some of the fields mentioned above, in the direction of fighting decoherence and loss. These works fall in two categories: on one hand, we looked at computer architectures which can be used to combat errors, using techniques of quantum error correcting codes. In a first project we found decoherence and loss probability thresholds below which quantum computing is provably possible. We assumed a very particular error model tailored specially to quantum dots as single photon sources and linear optics. Subsequently we looked at the problem of loss, both of heralded and unheralded, and devised some ways to fight it. The framework under which this work was done was used to develop theory which is currently being tested in a quantum optics experimental group and will be reported in an article later this year. On the other hand, we studied how the error probability can be reduced at the physical level, thanks exclusively to the properties of the system in which information is stored, as opposed to making use of quantum codes. We looked at a particular superconducting circuit, which is potentially very well protected against some types of decoherence. In particular, we observed that the interaction with the environment become weaker for certain values of the circuit external parameters.

Silicon-Germanium alloy heterostructures offer the most viable opportunity to integrate electronics with optoelectronic devices for widespread commercial application. Indeed Germanium rich devices may be designed for application around 1.5 m by preying on the direct-gap energy of 890meV. Low power optical modulators operating, under the quantum confined Stark effect, at wavelength bands used in 3rd generation fibre optic communication channels are developed in this thesis from a theoretical perspective. An investigation into strained Germanium rich quantum well structures was performed, revealing information about sub-band dispersion, joint density of states and absorption coefficient using the double group formulation of k . p theory. Using zone centre eigenstates as symmetrised half integer basis functions transforming according to irreps of the double group, the spin orbit interaction is incorporated into the unperturbed Hamiltonian. Along with semi-empirical input parameters available in the literature, dispersion in bulk Silicon and Germanium reveals information about hole effective masses and indirect conduction band minima in broad agreement with experimental data. In accordance with degenerate perturbation theory; effective mass Hamiltonians, with an arbitrary quantisation axis through a canonical transformation, are constructed through a series of matrix multiplications. Retaining operator ordering allows numerical modelling of heterostructures grown on arbitrary growth planes with appropriate boundary conditions across an abrupt interface under the envelope function framework. In this thesis, the effect on the transition energy, hh1-e1, by the choice of growth plane in a quantum well heterostructure is investigated.

CMOS technology scaling improves the speed and functionality of microprocessors by reducing the size of transistors. Static power dissipation also increases as a result of scaling however, and has been identified as a limiting factor in technology scaling. As current technology approaches that limit, techniques are required both at the technology-level and in the architecture design to reduce sub-threshold leakage, which accounts for the majority of static power dissipation. This thesis presents an approach to predict the idle periods of execution units at runtime and power-gate them during these periods to eliminate their static power leakage. We exploit similar execution characteristics across loop iterations to build a prediction of the units required to execute an entire loop from the units used over the first few iterations. The utilisation of each execution unit is monitored for each iteration, and thresholds are used to determine which units should be power-gated for the remainder of the loop. Three techniques are presented: Loop-Directed Mothballing (LDM), Extended Loop-Directed Mothballing (ELDM) and schedule balancing. LDM power-gates execution units only during innermost loops, which are simple to detect at runtime. ELDM extends this method to all loops using loop entry and exit information gathered offline. The balancing scheduler is developed to balance the types of instruction issued each cycle, to encourage reuse of execution units and make unnecessary units easier to detect. Extensive simulation using traces of 16 benchmarks from the SPEC CPU2006 suite demonstrates that LDM reduces the energy-delay product of our simulated superscalar processor by 10.3%. For traces with a low proportion of executed instructions inside innermost loops, ELDM improves the energy-delay product by up to 13% by allowing the technique to be applied to other loops in the trace. Employing schedule balancing with ELDM achieves similar savings, and simplifies the hardware required to make predictions.

The k . p perturbation method, initially developed for bulk systems has subsequently been generalised to model heterostructures using the envelope function approximation. This semi-empirical modelling technique provides vital information on the electronic and optical properties of complicated material systems that form heterostructures. It is known that k . p theory can be formulated in two ways; either through use of single or double group basis functions, with the implication that spin orbit interaction is either treated as a perturbation in parallel with the k . p term, or as part of the unperturbed Hamiltonian. The critical difference between the two approaches is that under the former, single group selection rules are considered in the evaluation of the k-dependent Hamiltonian, and the subsequent treatment of spin orbit interaction as a perturbation places restrictions on the adapted double group bases. Under the latter approach, double group selection rules are considered, resulting in an explicit change in the k-dependent Hamiltonian for multiband models of 14-band models or higher, and implicit changes to material parameters describing lower band models. A key result, is the ability of the double group formulated k . π theory to properly account for the experimentally measured spin orbit band and conduction band effective masses. This in particular leads to a change in the bulk valence band dispersion relation. In heterostructures, double group-effective mass equations result in changes to the operator ordered Hamiltonian, and the use of double group material parameters impacts the confinement energies and in-plane dispersion of subbands in a quantum well. An investigation is made into the impact of these changes on the electronic properties of Ge/SiGe heterostructures, which have applications as optical modulator devices.

Quantum dots are semiconductor nanocrystals with size-dependent optical properties that result from their nanoscale dimensions. These materials are emerging as simpler and more sensitive alternatives to traditional fluorescent small molecules and radioactive reporters in biomarker assays. Quantum dot emission can be modulated via proximal binding of organic dyes or gold nanoparticles, which can form the basis of a sensor. Their multivalency and ease of functionalisation allow for the attachment of multiple biosensing ligands, boosting the detection sensitivity. Quantum dots with different emission wavelengths can be excited simultaneously and distinguished from one another spectrally, a property which can be used for multiplexing. Enzymes mediate the chemical modification of proteins thereby controlling the signalling cascades that regulate cell behaviour. Hence, aberrant activity of enzymes can be associated with the onset of disease. Clinical tests typically determine the total concentration of enzyme in a sample without regard to quantification of activity. In this thesis, the development of generic activity-dependent tests for acetyltransferases, kinases and proteases is described. Enzyme activity is reported via decoration of quantum dots with enzyme substrate peptides and subsequent binding of FRET acceptor dye-labelled antibodies, which mediate changes in quantum dot emission spectra. Using this platform, p300 histone acetyltransferase was detected with a limit of detection comparable to that of radiolabelling assays. Modifications of the platform to detect serine and tyrosine phosphorylation were investigated. The phosphotyrosine assay was combined with a gold nanoparticle-quantum dot assay for the detection of a kinase/protease biomarker pair relevant for the determination of breast cancer prognosis. The modular nature of this assay design allowed for the detection of different classes of enzymes singly and simultaneously, representing a generic platform for high-throughput enzyme screening in rapid disease diagnosis and drug discovery.

A series of novel semiconducting co-polymers based on known, fused aromatic monomers (dibenzosilole, dithienosilole and dithienopyrrole) were synthesized for use in organic field effect transistor (OFET) devices. The fused monomers were copolymerized with triarylamine and thienothiophene units to give a series of six donor polymers. Their electronic energy levels were probed by UV-Vis and photoelectron spectroscopy, and predictions regarding their performance were drawn from this information. Two polymers were tested in OFETs and conclusions were made concerning the relationship between structure and performance. Four donor-acceptor type polymers were synthesized using dithienopyrrole as the electron rich unit. This monomer was copolymerized with benzothiadiazole, difluorobenzothiadiazole, thienopyrrolodione and 1,1’-bithienopyrrolodione. These low band gap polymers were tested as the donor polymer in polymer:fullerene BHJ devices. A novel monomer with long branched alkyl substituents, based on the extended fused-ring system indacenodithiophene (IDT), was synthesized. The nitrogen analogue of this monomer (NIDT) was of interest as its performance could be compared to the carbon and silicon versions of IDT. A route to NIDT was devised and then it was polymerized with four electron poor monomers. The energy levels were estimated by UV-Vis and photoelectron spectroscopy, followed by testing of the materials in OFET and OPV devices. Those with the lowest lying HOMO levels performed best in OPVs. Another series of NIDT containing polymers were synthesized, designed specifically for use in OFET devices. The aim was to produce polymers with highly planar backbones which would form ordered domains and enhance hole transport. An NIDT monomer was synthesized with linear hexadecyl substituents in order to improve π-stacking. The linear NIDT monomer was then copolymerized with thiophene, thienothiophene, thienopyrrolodione and diketopyrrolopyrrole. It was found that the diketopyrrolopyrrole containing polymer, performed best in OFETs and the molecular weight played a large role in how a polymer performed.

This thesis presents research into spin-transport in Carbon nanotube quantum dots. Sputtered Permalloy electrodes designed with shape anisotropy were used to contact Carbon nanotubes grown by chemicalvapour deposition in lateral spin-valve structures. The magnetoresistance of these spin-valves were measured at low-temperatures and as a function of the charge state of the quantum dots. Two conductance regimes were measured in a Carbon nanotube spinvalve with Permalloy nucleation pads. At high bias outside of the coulomb blockade regime a ~ 10% magnetoresistance was measured that is analogous to giant-magnetoresistance, in that it is due to spindependent scattering at the ferromagnet-Carbon nanotube interfaces. At lower bias the device enters the coulomb blockade regime and the magnetoresistance observed develops a different structure, over a larger field range, together with the development of an offset in conductance between saturations. The maximum value of this MR was MR ~ 245% and it was attributed to changes in the induced charge on the quantum dot. By modifying the design of the Permalloy electrodes, a single domain state at the point of contact of the Carbon nanotube was achieved. A well defined anti-parallel state of the Permalloy electrodes, with associated changes in the conduction of the devices was observed, yet the conductance offset remained, with a maximum MR of ~ 60%. The positions of the coulomb peaks were measured during magnetic reversal of the electrodes, showing the change in induced charge on the quantum dot, with a maximum MR ~ 350%. Predictions of device transport based on the magneto-coulomb effect and spin-dependent interfacial phase shifts were compared to experimental results and found to not fit the observed behaviour. This led to the conclusion that changes in the charge state of the quantum dots must be due to a fixed spin-quantisation axis intrinsic to the Carbon nanotube.

The phosphate modified tellurite (PT) glasses were investigated for their potential applications of bulk glass and PLD films as glass-on-semiconductor integrated thin films for engineering a semiconductor pump source integrated waveguide amplification medium which can operate at a wavelength of 1500nm. High quality PT bulk glasses were used as PLD target for deposition of high quality thin films on the silica buffered GaAs substrate by using optimized deposition parameters; target-to- substrate distance, processing gas pressure, deposition temperature and laser energy. The propagation loss of the deposited films was measured as low as 2.3 dB/cm at a wavelength of 633 nm. The Er3+: 4I13I24I15/2 optical transition lifetime response was measured as 2.5 ms. However, a dramatic increasing in the measured lifetime value was recorded by increasing the deposition temperature. Although, deposited PT films shows crack free structure (in the 200nm range), the material transfer from the bulk target to the deposited films was observed to be non-stoichiometric transfer for rare-earth elements, and for host materials a stoichiometric transfer of the materials was observed. The effects of changing the ambient gas composition on the surface morphology of the deposited PLD film was investigated and it was found that using Oxygen-Helium gas mixture produce of a ceramic-like structure in deposited thin films. The ultrafast laser micromachining was used for thin film waveguide fabrication. However, strip loaded waveguide structures were carried out using conventional photolithography followed by reactive ion etching of Si3N4. Buried waveguide structure were fabricated using ultrafast laser micromachining technique in the bulk PT glasses, and maximum gain of 3.2 dB was recorded at the erbium peak for a 21 mm long waveguide using pump wavelength of 976 nm. However, for the ridge waveguide structure a non-inverted signal gain response with internal gain peak of 2.69 dB at 1565 nm for a 11 mm long waveguide was recorded in the deposited film on the silica substrate. Glass-on- semiconductor integrated waveguide structure was fabricated using Si3N4 as strip layer, and satisfactory ASE response to the 976 pump light was recorded.

Defects usually have a very large influence on the semiconductor material properties and hence on fabricated electronic devices. The nature and properties of defects in semiconducting materials can be investigated by applying electrical characterization techniques such as thermal admittance spectroscopy (TAS), deep level transient spectroscopy (DLTS) and high resolution Laplace-DLTS measurements. This dissertation presents the electrical characterisation of two different wide bandgap semiconducting materials (polycrystalline diamond and GaN) which have both recently attracted a great deal of attention because of their potential applications in the fields of power electronics and optoelectronics. Raman spectroscopy, I-V and C-V measurements were carried out as supporting experiments for the above investigations. The first part of this work focuses on studying the effect of B concentration on the electronic states in polycrystalline diamond thin films grown on silicon by the hot filament chemical vapour deposition method. A combination of high-resolution LDLTS and direct-capture cross-section measurements was used to investigate whether the deep electronic states present in the layers originated from point or extended defects. There was good agreement between data on deep electronic levels obtained from DLTS and TAS experiments. A number of hole traps have been detected; the majority of these levels show an unusual dependence of the DLTS signal on the fill pulse duration which is interpreted as possibly the levels are part of extended defects within the grain boundaries. In contrast, a defect level found in a more highly doped film, with an activation energy of -0.37 eV, exhibited behaviour characteristic of an isolated point defect, which we attribute to B-related centres in the bulk diamond, away from the dislocations. The second part of this thesis presents electrical measurements carried out at temperatures up to 450 K in order to study the electronic states associated with Mg in Mg-doped GaN films grown on sapphire by metalorganic vapour phase epitaxy, and to determine how these are affected by the threading dislocation density (TDD). Two different buffer layer schemes between the film and the sapphire substrate were used, giving rise to different TDDs in the GaN. Admittance spectroscopy of the films finds a single impurity-related acceptor level. It is observed in theses experiments that admittance spectroscopy detects no traps that can be attributed to extended defects, despite the fact that the dislocations are well-known to be active recombination centres. This unexpected finding is discussed in detail.

In this thesis I will discuss the results of optical spectroscopy measurements in relation to the study of quantum well and multiple quantum well structures made up of group III-nitride semiconductors. This will include the results of photoluminesence(PL), reflectance, and time resolved photoluminesence spectroscopy experiments as well as the results of time correlated single photon counting (TCSPC) experiments. I have investigated the droop in efficiency observed in InGaN/GaN quantum wells as the current density or excitation power density is increased. I will show that at the onset of the efficiency droop a simultaneous reduction in the S-shape temperature dependence of the peak emission energy shift is observed. These results suggest that one of the mechanisms behind the efficiency droop is the saturation of localised states as the density of carriers is increased. I also discuss the carrier dynamics in non-polar GaN/AlGaN quantum wells intersected by basal plane stacking faults (BSF)s. In order to describe the form of the results of TCSPC results I put forward a model for the transfer of carriers from the quantum well into the regions of the quantum well intersected by BSFs. Finally I show the results of an investigation into the features observed in photoluminesence excitation measurements by N Hylton et al [1, 2] near the GaN band gap. These features are caused by a red shift in the peak emission energy using the results of PL measurements made as a function of excitation photon energy it is possible to resolve the lowest energy feature observed by Hylton et al into two features. Two possible mechanisms are then discussed for how the excitation of carriers at set energies above the GaN band gap results in a change to the peak energy of the quantum well emission.

Dilute magnetic semiconductors (DMSs) are attractive. They are candidate materials for applications in novel spintronic devices. Because of the giant Zeemaneect in the paramagnetic state, a magnetic eld can be used to manipulate the spin and charge of carriers in DMSs. One possibility is to exploit the nonhomogeneous magnetic elds due to superconductors. In this thesis, the heterostructures of the planar DMS and superconductors in dierent geometries and superconducting states are investigated to understand the electronic structure of electrons in the DMS. The combination of a superconducting disk in the Meissner state and the planar DMS is studied using both simple and realistic models of the magnetic eld associated with the disk. The giant Zeeman interaction is found to substantially inuence the energies of magnetically conned states in the adjacent DMS. In the simple model eld, the giant Zeeman energy acts as an extra conning potential and results in spin dependent electron states exhibiting dierent spatial distributions, while the more realistic model eld results in conned states exhibiting a variety of mixed spin characters. The hybrid of a superconducting lm in a superconducting vortex state and the DMS is then explored. The concentrated magnetic eld due to an isolated vortex is shown to trap strongly spin polarised electron states. In the case of an Abrikosov lattice of vortices, interactions between vortex-bound states result in a band structure which can be controlled by the magnitude of an external uniform magnetic eld. It is found that the numerical band structures obtained using a basis of Landau states dier from those previously reported, leading to the development of a tight-binding theory to conrm their corrections. Another hybrid investigated is a square superconductor above the DMS. In this case, the arrangement of vortices is distorted by the boundary of the sample, leading to the possibility of multivortex state and/or giant vortex states. It is discovered that the magnetic eld due to the former state induces \molecular" electron states in the DMS, while that due to the latter state induces electron states with increased spatial distribution. Tight-binding theory is again used to describe the observed energy levels and the interactions between electron states induced by the magnetic elds due to separated vortices in the multivortex state.

Precessional dynamics are exploited in the operation of high frequency magnetic devices such as magnetic disk drives, non reciprocal microwave devices and spin transfer oscillators. The trajectory of the precession and its damping are of crucial importance. This thesis presents the characterisation of a variety of magnetic thin film structures performed with a range of phase sensitive techniques. It is possible to obtain new insight by utilising the chemical and site specificity of X-ray Magnetic Circular Dichroism (XMCD) to isolate the precession in different chemical species or at distinct sites in the crystal structure of a chosen material. X-ray Ferromagnetic Resonance (XFMR) combines XMCD and Ferromagnetic Resonance (FMR) phenomena in a technique capable of measuring the FMR response of an alloy or multilayer with both chemical and site specificity. To complement the XFMR technique a low temperature Time-Resolved Magneto Optical Kerr Effect (TR-MOKE) setup has been developed. This allowed for the characterisation of samples at temperatures in the range 4 K to room temperature. A frequency swept Vector Network Analyser FMR (VNA-FMR) setup was developed to allow for a fast method for determining the resonance condition and damping of a range of ferromagnetic thin film samples. In addition a TR-X-ray Photoemission Electron Microscopy (TR-XPEEM) setup has been established which allows images to be obtained with magnetic contrast. The combination of the above techniques has lead to studies on rare earth capped spin valve free layers and the measurement of spin pumping in industrially relevant spin valves.

This thesis presents theoretical studies of the optical and electronic properties of defects in diamond and of the mechanisms of doping graphene. The birefringence of the four petalled defect commonly observed in CVD diamond is explained by four linear arrays of dislocations along ⟨110⟩ directions with ⟨110⟩ Burgers vectors. Such an arrangement of dislocations reproduces the extension and the features of the birefringence patterns observed experimentally. Density functional theory via the AIMPRO code was used to study the electronic and optical properties of different nitrogen-related point defects in diamond. It was found that the zero-phonon luminescence line of the NV− defects can split in the presence of a surface or other NV− defects. Since VNH and VN2 are expected to have similar optical properties, the optical transi- tions for VN2 were used to correct the transitions for VNH calculated by local density approximation. The absorption band at 2.38 eV (520 nm) observed in CVD diamond is then attributed to an internal transition of VNH. The weak zero-phonon line and broad vibronic sidebands for VN− and VN−2 and its absence for VNH− is explained by the large structural change when the defect is excited. Finally, different mechanisms for doping graphene were considered. The calculations predict the electropositive metals, such as Ti and Cr, act as donors, while molecules with strong electron affinity, such as F4-TCNQ, act as acceptors in graphene. An unexpected mechanism of doping graphene was shown by Au which dopes bilayer graphene but not single layer. In the presence of water, electrochemical reactions on the graphene can also lead to p or n-type doping.

The work in this doctoral thesis is mainly concerned with the detection of volatile organic vapours (analytes) using organic field-effect transistors (OFETs) as transducers, in some cases using a ‘sensitiser layer’ on top of the devices to improve their response to certain analytes; some work has also been carried out using a gold nano-particle chemi-resistor to detect amine vapour and the development of an aqueous sensing system is also discussed. It was found that the porphyrins PtOEP (platinum (II) octaethyl porphyrin) and PtEP-I (Etioporphyrin-I) could be used as organic semiconductors and that PtOEP was sensitive to isopropanol (IPA) and acetone vapours; PtOEP was also used to successfully sensitise a pentacene OFET to ethylene vapour at low ppm concentrations. Pentacene OFETs were found to be sensitive to octylamine (an amine), ethylethanoate (an ester), formamide (an amide) and ethylene (an alkene); through the use of a 2:1 molar ratio blend of the calixarene calix[8]arene (calixarene 2) and the porphyrin 5,10,15,20-Tetrakis (3,4-bis (2-ethylhexyloxy) phenyl)-21H,23H-porphyrinato cobalt (II) (Co-EHO) as a sensitiser layer, it was possible to introduce sensitivity to both octanal (an aldehyde) and octan-2-one (a ketone) into a pentacene OFET; the calixarene: 5,17-(34-nitrobenzylideneamino)-11,23-di-tert-butyl-25,27-diethoxycarbonyl-methyleneoxy-26,28dihydroxycalix[4]arene (calixarene 1) was also be used to improve OFET recovery after exposure to ethylethanoate and formamide, but some sensor response was lost in the process. The n-type organic semiconductor PDI8-CN2 (N,N’-bis (n-octyl)- dicyanoperylene-3,4:9,10-bis(dicarboximide)) was found to be sensitive to octylamine vapour, but the nature of its response seems to indicate some kind of amine base-doping mechanism is at work within the device, analogous to the acid doping possible with p-type semiconductors. Gold nano-particles were found to be sensitive to octylamine vapour as the amine group has an affinity for gold and coats the nano-particles, increasing the resistance of the nano-particle film. Creating a water gated P3HT (poly(3-hexylthiophene-2,5-diyl)) OFET without the electro-chemical doping normally experienced by such devices was found to be possible through the use of a calixarene 1 barrier layer, paving the way for the development of an aqueous sensing system.

III-V semiconductor alloys possess interesting electronic properties that can be engineered and optimised for highly-efficient light absorption and emission in optoelectronic devices. Although GaAs is by far the most mature and cost-effective III-V semiconductor material, its use as a starting material for optoelectronic devices is limited by the feasibility of growing high-quality single crystals with different lattice parameters on GaAs substrates. In this thesis, p-i-n diodes comprised of two types of GaAs-based alloys with bandgap energies in the near and short-wave infrared (SWIR) range for photo-detection applications, are investigated experimentally. All of the samples are grown by molecular beam epitaxy (MBE) technique. The main focus of this thesis is on the first type of alloy, GaInNAs, in which the addition of a small amount of substitutional N atoms into GaAs results in an anomalously large energy band gap reduction, while the addition of In compensates the tensile strain introduced by the small N atoms to achieve lattice matching to GaAs. The second alloy, GaInAsSb, is conventionally grown on GaSb substrate, but epitaxial growth on GaAs is possible via a special growth mode, which enables a very thin GaSb epilayer to be grown directly on GaAs with minimum or absence of threading dislocations. As both of the material systems investigated in this work are known for various growth-related problems, attempts were made, whenever possible, to infer the electronic properties of the devices irrespective of the growth process. Growth-related defects in the dilute nitride GaInNAs alloy system are known to affect optoelectronic device performance with issues such as short minority-carrier diffusion lengths and high background doping densities. Despite these difficulties, interest into the dilute nitride alloys has continued to grow for over two decades, as evident by a recent breakthrough in high-efficiency, multi-junction solar cells utilising 1 eV GaInNAs absorber layers. Hence the thesis has to first deal with the effects of growth-related defects on the optoelectronic performance of GaInNAs devices. Investigation of the photovoltaic characteristics of GaInNAs diodes are useful in revealing issues affecting the efficiency and detectivity of photodetectors and solar cells, such as those associated with dark currents, quantum efficiency, minority carrier diffusion lengths, and lifetime. In addition, the effect of Sb, which is commonly used as a surfactant during the growth process of Ga(In)NAs, is investigated. The unique electronic band-structure properties in GaInNAs semiconductors are investigated through experimental measurements of the avalanche breakdown voltage, multiplication and excess noise in GaInNAs photodiodes with a range of N compositions below 4%. The results reveal evidence that both the electron and hole ionisation coefficients in GaInNAs with 3.8% N and 10% In are suppressed by up to two orders of magnitude at low fields, compared with GaAs. At high fields, analysis of the avalanche breakdown behavior suggests that electrons impact ionise at energies above the upper subband, E+(k). Consequently, the electron and hole ionisation coefficients are not so significantly disparate as previously thought, which has important implications for avalanche photodiode applications. Finally, the detectivity performance of GaInAsSb SWIR p-i-n photodiodes grown on GaAs substrate via a special growth mode is investigated. The study reveals that very smooth epilayer surface morphology and robust, high-detectivity GaInAsSb photodiodes can be realised on GaAs despite the presence of threading dislocations. These photodiodes yield an estimated maximum detectivity of mid 1010 cm Hz1/2 W-1 at room temperature, with a zero-bias differential-resistance area product of 260 O cmand a peak responsivity of 0.8 A/W at 2 µm. Although not having the best crystal quality at present, the performance of the GaAs-based GaInAsSb photodetectors is comparable to, if not marginally better, than similar GaSb-based devices reported to date in the literature. This further substantiates the potential of extending the detection wavelength of GaAs-based, high-operating­-temperature photodetectors to the short and mid-wave infrared range.

This thesis describes experiments on the initialisation, coherent control, and readout of a single hole spin trapped in a self assembled InGaAs semiconductor quantum dot. High fidelity initialisation of a hole spin state is achieved by the fast ionisation of a spin polarised neutral exciton under an applied electric field and in a Faraday geometry magnetic field. The preparation of a coherent superposition state is demonstrated by observing the precession of the hole spin about a Voigt geometry magnetic field. The hole spin dephasing time is deduced from the decay of the spin contrast. Coherent optical rotation of the hole spin state about the z-axis is demonstrated using the geometric phase shift induced by a picosecond laser pulse. By combining the precession of the spin about the x-axis, and optical rotations about the z-axis, full quantum control of a hole spin is demonstrated over the surface of the Bloch sphere. This is an important prerequisite for the use of a hole spin as a qubit for quantum information processing applications.

A unique organic/inorganic nanocomposite of lead sulphide (PbS) quantum dots (QDs) embedded in substituted metal-free phthalocyanine (C6H2Pc) has been prepared by a simple and low-cost method. The preparation procedure consists of exposure of a thin spun film of non-peripherally octa-hexyl lead phthalocyanine to hydrogen sulphide atmosphere. The formation of the PbS QDs has been verified using X-ray diffraction and transmission electron microscopy techniques. From the transmission electron microscopic measurements, the average size of the PbS QDs is found to be 4.5 nm, which is smaller than the exciton Bohr radius. Independent Xray diffraction and optical absorption studies provide supportive evidence for the size of QDs. Quantum confinement gives rise to a clear blue shift in the absorption spectrum with respect to the bulk PbS. The QDs band gap has been estimated to be 1.95 eV from Tauc's law and the frontier energy levels of the PbS QDs has been derived. About two orders of magnitude increase in ohmic conductivity, from 6.0×10−12 for C6H2Pc to 3.1×10−10 for the nanocomposite, is observed by steady-state electrical measurements in sandwich structure between indium tin oxide and aluminium. Temperature-dependence of the electrical conduction is studied aimed to calculate the activation energy and determine the type of conductivity. The incorporation of the PbS QDs decreases the activation energy by about 0.5 eV at temperatures higher than 240 K. It is found that the Poole-Frenkel mechanism is in good consistency with the superlinear electrical behaviour of the nanocomposite. The frequency response of alternating current (AC) conduction is found to obey the universal power-law. The cryogenic study of AC conduction reveals that the correlated barrier hopping (CBH) model closely fits to the experimental data at temperatures below 240 K. The parameters obtained by fitting the CBH model point out that the hopping process cannot take place directly between neighbouring PbS QDs but involves the localised states within the matrix.

Power electronics are an enabling technology and play a critical role in the establishment of an environmentally-friendly and sustainable low carbon economy. The electrification of passenger vehicles is one way of achieving this goal. It is well acknowledged that Electric vehicles (EVs) have inherent advantages over the conventional internal combustion engine (ICE) vehicles owing to the absence of emissions, high efficiency, and quiet and smooth operation. Over the last 20 years, EVs have improved significantly in their system integration, dynamic performance and cost. It has attracted much attention in research communities as well as in the market. In 2011 electric vehicle sales were estimated to reach about 20,000 units worldwide, increasing to more than 500,000 units by 2015 and 1.3 million by 2020 which accounts for 1.8 per cent of the total number of passenger vehicles expected to be sold that year. In general, electric vehicles use electric motors for traction drive, power converters for energy transfer and control, and batteries, fuel cells, ultracapacitors, or flywheels for energy storage. These are the core elements of the electric power drive train and thus are desired to provide high reliability over the lifetime of the vehicle. One of the vulnerable components in an electric power drive train is the IGBT switching devices in an inverter. During the operation, IGBT power modules will experience high mechanical and thermal stresses which lead to bond wire lift-off and solder joint fatigue faults. Theses stresses can lead to malfunctions of the IGBT power modules. A short-circuit or open-circuit in any of the power modules may result in an instantaneous loss of traction power, which is dangerous for the driver and other road users. These reliability issues are very complex in their nature and demand for the development of analytical models and experimental validation. This work is set out to develop an online measurement technique for health monitoring of IGBT and freewheeling diodes inside the power modules. The technique can provide an early warning prior to a power device failure. Bond wire lift-off and solder fatigue are the two most frequently occurred faults in power electronic modules. The former increases the forward voltage drop across the terminals of the power device while the latter increase the thermal resistance of the solder layers. As a result, bond wire lift-off can be detected by a highly sensitive and fast operating in-situ monitoring circuit. Solder joint fatigue is detected by measuring the thermal impedance of the power modules. This thesis focuses on the design and optimisation of the in-situ health monitoring circuit in an attempt to reducing noise, temperature variations and measurement uncertainties. Experimental work is carried out on a set of various IGBT power modules that have been modified to account for different testing requirements. Then the lifetime of the power module can be estimated on this basis. The proposed health monitoring system can be integrated into the existing IGBT driver circuits and can also be applied to other applications such as industrial drives, aerospace and renewable energy.

New approaches to metal-oxide-semiconductor field effect transistor (MOSFET) engineering emerge in order to keep up with the electronics market demands. Two main candidates for the next few generations of Moore’s law are planar ultra-thin body and buried oxide (UTBB) devices and three-dimensional FinFETs. Due to miniature dimensions and new materials with low thermal conductivity, performance of advanced MOSFETs is affected by self-heating and substrate effects. Self-heating results in an increase of the device temperature which causes mobility reduction, compromised reliability and signal delays. The substrate effect is a parasitic source and drain coupling which leads to frequency-dependent analogue behaviour. Both effects manifest themselves in the output conductance variation with frequency and impact analogue as well as digital performance. In this thesis self-heating and substrate effects in FinFETs and UTBB devices are characterised, discussed and compared. The results are used to identify trade-offs in device performance, geometry and thermal properties. Methods how to optimise the device geometry or biasing conditions in order to minimise the parasitic effects are suggested. To identify the most suitable technique for self-heating characterisation in advanced semiconductor devices, different methods of thermal characterisation (time and frequency domain) were experimentally compared and evaluated alongside an analytical model. RF and two different pulsed I-V techniques were initially applied to partially depleted silicon-on-insulator (PDSOI) devices. The pulsed I-V hot chuck method showed good agreement with the RF technique in the PDSOI devices. However, subsequent analysis demonstrated that for more advanced technologies the time domain methods can underestimate self-heating. This is due to the reduction of the thermal time constants into the nanosecond range and limitations of the pulsed I-V set-up. The reduction is related to the major increase of the surface to volume ratio in advanced MOSFETs. Consequently the work showed that the thermal properties of advanced semiconductor devices must be characterised within the frequency domain. For UTBB devices with 7-8 nm Si body and 10 nm ultra-thin buried oxide (BOX) the analogue performance degradation caused by the substrate effects can be stronger than the analogue performance degradation caused by self-heating. However, the substrate effects can be effectively reduced if the substrate doping beneath the buried ii oxide is adjusted using a ground plane. In the MHz – GHz frequency range the intrinsic voltage gain variation is reduced ~6 times when a device is biased in saturation if a ground plane is implemented compared with a device without a ground plane. UTBB devices with 25 nm BOX were compared with UTBB devices with 10 nm BOX. It was found that the buried oxide thinning from 25 nm to 10 nm is not critical from the thermal point of view as other heat evacuation paths (e.g. source and drain) start to play a role. Thermal and substrate effects in FinFETs were also analysed. It was experimentally shown that FinFET thermal properties depend on the device geometry. The thermal resistance of FinFETs strongly varies with the fin width and number of parallel fins, whereas the fin spacing is less critical. The results suggest that there are trade-offs between thermal properties and integration density, electrostatic control and design complexity, since these aspects depend on device geometry. The high frequency substrate effects were found to be effectively reduced in devices with sub-100 nm wide fins.

Systems have been designed and synthesized using CMOS technology for many years, with improvements in the fabrication process allowing designs to be scaled onto smaller areas with relative ease. The introduction of nano scale CMOS technologies has ended this time of simple scaling, as variations within the silicon now dramatically affect circuit performance and manufacturing yield. These random physical variations cannot be removed from the manufacturing process, requiring that their affects are modelled, predicted and accommodated within the design process. This thesis presents an investigation into the challenges of including these affects within the design process, with a review of the recent research conducted in incorporating variability within timing analysis tools. The conclusion from the literature review is that an accurate, efficient and transparent method of predicting the impact of statistical process variations on the performance of a circuit has not yet been created and adopted by the IC design industry. The investigation begins with the modelling of transistor based statistical process variations at the standard cell level, where it is determined that simple statistical models do not accurately reflect the extremes in performance, and can provide overly pessimistic predictions. The techniques of Monte Carlo Cell Characterisation (MCCC) and Monte Carlo Static Timing Analysis (MCSTA) are introduced as more suitable approaches, which accurately reflect the performance of circuits as modelled by Monte Carlo SPICE simulations, with far less pessimism than the traditional method of Corner Analysis or even modern Statistical Static Timing Analysis. The final section of this thesis focuses on practical implementations of MCSTA, where the sample sizes required to accurately predict circuit behaviour (to within 1% of SPICE)can be reduced to as few as ten, using simple statistical sampling techniques.

In this thesis we investigate a system of three tunnel-coupled quantum dots, arranged in a triangular geometry and attached to a single metallic conduction band, using both analytic and semi-analytic methods and the numerical renormalisation group technique. This is the simplest coupled quantum dot system to exhibit frustration. We study three different models of the triple quantum dot device: a mirror symmetric arrangement of dots in which only one dot is connected to the conduction band, a triple quantum dot system in which only one dot is connected to the conduction band without a plane of mirror symmetry and a mirror symmetric arrangement of quantum dots in which all three dots are coupled to the conduction band. We study these models over a wide range of parameter space, and in both the presence and absence of a magnetic field. Both antiferromagnetic and ferromagnetic Kondo effects are observed, and in all three models we find that the system contains at least two phases, and so a number of quantum phase transitions may be observed, associated in some cases with significant changes in the low temperature conductance through the triple quantum dot device. In addition to zero-field Kondo physics, a number of field induced Kondo effects are also observed. Both first order quantum phase transitions and Kosterlitz-Thouless phase transitions are observed. We use both symmetry arguments and low energy effective models which we derive to explain and understand both the position of and type of phase boundary that is observed in each case, and perturbative methods are used to accurately predict Kondo temperatures for a wide range of systems.

III-V semiconductor nanostructures are widely used in optoelectronic devices (e.g. lasers and detectors) in the visible (0.4-0.8 μm), near-infrared (0.8-3 μm), mid-infrared (3-5 μm) and far-infrared (> 8 μm) wavelength ranges, with great potential for high performance and high temperature operation. As well as simple designs, complex structures incorporating low dimensional components (e.g. quantum wells and quantum dots) are not unusual. Often, the optical and electronic characteristics of these structures are altered significantly as compared to bulk material. As a prerequisite to design for different applications, the study of their electronic and optical properties is essential. With the increasing computational power of modern personal computers, computational modelling becomes viable and more efficient. Indeed, it has become routine to follow (or to precede) experimental studies with computational modelling of good interpretive and predictive power. Combined with experimental studies, this is a powerful tool to provide insight into new devices. This research work is primarily based on calculations of the electronic band structure of various semiconductor nanostructures, followed by modelling of optical transitions and optical spectra. All numerical calculations use a cost effective computational method. The applicability of the model to ultra-thin structures of short period InAs/GaSb superlattices is investigated. The work is then extended to study complex quantum-dot-in-well structures. Finally, the attempt to extract the structural parameters of quantum dots by a combination of modelling and optical spectroscopy is presented.

III-V semiconductors doped with Mn atoms have been studied recently for their potential applications in spintronic devices. These materials are referred to as diluted magnetic semiconductors (DMS) or ferromagnetic semiconductors. The lattice of these materials has Mn atoms (transition metal) and due to their half filled shell carries a spin and also donates carriers, "p-type" in case of Mn doping to the semiconductor lattice. Ferromagnetism is mediated by exchange interactions between localized d electrons and p-type carriers. A high transition temperature (in fact above room temperature) called Curie point, is necessary for these materials to have a practical value. The transition temperature Tc depends on the densities of both Mn impurity and the resultant hole concentration. A low density of defects, a high crystal quality and a high concentration of Mn atoms are desirable to achieve a high Tc. Additionally, cubic (zinc blende) structure offers several advantages over the hexagonal structure normally obtained for III-V semiconductors e.g. a higher crystal symmetry, smaller effective masses, higher carrier mobility, and significantly a higher doping density-which is expected to give a higher value of Tc. Both GaMnAs and GaMnN have been the materials of choice and have been studied intensely. The problems however, are related to fabricated technique, crystal quality and achieving a high doping concentration. The materials exhibit a solubility limit beyond which doping concentration are difficult to achieve. This thesis describes the investigation of the electronic properties of the zinc blend GaMnAs and GaMnN films grown at low temperature by molecular beam epitaxy (MBE) for varies concentration of Mn doping over the temperature range 15-400 K. Metal/c-GaAs: Mn/c-GaAs:Si p n junctions were studied by the I-V, C-V and C-F methods over the temperature range 15-400 K. It was found that I-V-T data could be interpreted on the basis of a back to back diode model. The c-GaAs:Mn made a Schottky contact with the metal and a p n junction with c-GaAs:Si. It was found that for the forward bias, where metal was biased negative with respect to the GaAs:Mn, the 1- v -T data could also be analysed on the basis of a Schottky behaviour. Here ideality factor n increased and the barrier height Φb decreased with temperature. This was seen to be a consequence of TFE through the metal/GaAs barrier. This was borne out by the Richardson's plots [ln (ls/T2) versus 103/nT] which exhibited straight lines. The slope indicated the barrier heights, which ranged between 1.1-1.4 eV. The backward diode on the other hand gave a barrier height of 1.4 eV, which was compatible with the proposed model. C-V-T measurements could also be accounted for, at least qualitatively, on the basis of the proposed back to back diode model. The device structure of c-GaMnN was planar. Both the Ohmic and Schottky contact were in the same plane on the top surface. I-V measurements were performed over the temperature range 100-480 K. I-V-T was explained in the terms of the Schottky model. The current was limited, however, by the Poole-Frenkel type centres in the bulk which behaved as an insulator (high resistivity semiconductor). The C-V–T was satisfactorily accounted for on the basis of MIS/MOS model. Low temperature grown films had poor material quality which made any correlation between samples difficult.

Research presented in this thesis focuses on the optical characterisation of InGaN/GaN quantum well (QW) and quantum dot (QD) structures and ScGaNepi layers, supported by microscopy results from the University of Cambridge. Reported in the first part of this thesis are the optical properties of sets of high In fraction (~25%) multiple QW structures designed to emit in the green part of the spectrum. Sets of InGaN/GaN QW structures were investigated which were grown using traditional methods but by varying QW growth temperature. These samples were found to have very broad photoluminescence linewidths and 1-10% room temperature IQE. Changes to the growth procedures had little effect on the improvement of the samples' luminescence properties. It was found that the first two wells of these samples were thicker and/or contained more indium than subsequent wells grown in the stack. The added thickness and/or indium content of these InGaN/GaN QWs resulted in lower energy emission than the rest of the QW stack, broadening the photoluminescence linewidth and decreasing the IQE. A modification of the growth procedures was developed to ensure that the individual QWs have as similar properties as possible so that the contribution to the PL linewidth due to well-to-well variations was significantly reduced. These modified procedures were used to produce new sets of InGaN/GaN 10 QW structures. These new structures showed a marked increase (~30%) in IQE and a significant decrease in PL linewidth. Furthermore, a set of 1, 3, 5 and 10 QW structures grown under modified growth procedures were investigated to determine the optimum number of QWs needed. It was found that 3 QWs provided a significant improvement to the IQE over the 1 QW samples; however, no additional improvements were realised by growing additional QWs. The second part of this thesis explores the use of macroscopic optical spectroscopy methods to study the properties of QDs. InGaN QDs are typically studied using spatially resolved techniques which allow for the study of individual dots in a structure. A sample which was known to contain InGaN QDs was investigated; however it was determined that macroscopic spectroscopic techniques were unable to determine the existence of QDs in the structure. Difficulties creating highly efficient green or near-UV light emitting InGaN semiconductors have lead to an interest in alternative material structures. One suggested alternative is to replace indium with the Group IIIB transition metal, scandium. The final part of this thesis explores a series of ~ 260 μm thick MBE grown ScGaN layers on 500 μm thick MOCVD-grown GaN templates. It was found that these materials, believed to contain up to 8% scandium, emit a broad spectrum violet luminescence. This broad luminescence spectrum resolves into multiple, narrower features with either increased substrate temperature during growth or by decreasing the scandium effusion cell temperature. The absorption spectrum, on the other hand, only shows evidence of GaN and samples grown at the highest substrate temperature revealed an exciton absorption feature near the band edge of GaN, a sign of increased crystal quality. These results have lead to questions as to whether the violet luminescence is due to the ternary alloy ScGaN or rather from shallow, radiative defects in the material.

We study correlated two-level quantum impurity models coupled to a metallic conduction band in the hope of gaining insight into the physics of nanoscale quantum dot systems. We focus on the possibility of formation of a spin-1 impurity local moment which, on coupling to the band, generates an underscreened (USC) singular Fermi liquid state. By employing physical arguments and the numerical renormalization group (NRG) technique, we analyse such systems in detail examining in particular both the thermodynamic and dynamic properties, including the differential conductance. The quantum phase transitions occurring between the USC phase and a more ordinary Fermi liquid (FL) phase are analysed in detail. They are generically found to be of Kosterlitz-Thouless type; exceptions occur along lines of high symmetry where first-order transitions are found. A `Friedel-Luttinger sum rule' is derived and, together with a generalization of Luttinger's theorem to the USC phase, is used to obtain general results for the $T=0$ zero-bias conductance --- it is expressed solely in terms of the number of electrons present on the impurity and applicable in both the USC and FL phases. Relatedly, dynamical signatures of the quantum phase transition show two broad classes of behaviour corresponding to the collapse of either a resonance or antiresonance in the single-particle density of states. Evidence of both of these behaviours is seen in experimental devices. We study also the effect of a local magnetic field on both single- and two-level quantum impurities. In the former case we attempt to resolve some points of contention that remain in the literature. Specifically we show that the position of the maximum in the spin resolved density of states (and related peaks in the differential conductance) is not linear in the applied field, showing a more complicated form than a simple `Zeeman splitting'. The analytic result for the low-field asymptote is recovered. For two-level impurities we illustrate the manner in which the USC state is destroyed: due to two cancelling effects an abrupt change in the zero-bias conductance does not occur as one might expect. Comparison with experiment is made in both cases and used to interpret experimental findings in a manner contrary to previous suggestions. We find that experiments are very rarely in the limit of strong impurity-host coupling. Further, features in the differential conductance as a function of bias voltage should not be simply interpreted in terms of isolated quantum dot states. The many-body nature of such systems is crucially important to their observed properties.

The growth optimisation and nuclear spinl eddects in Ill-V semiconductor quantum dots are investigated and described in this thesis, The two main methods of epitaxial growt.h (molecular beam epitaxy and metal-orgnic vapour phase epitaxy) are considered, QuantuIIl dot growt h within material system of InP/GaInP is studied. with the effect on the resulting quantum dot charactristics such as spatial and spectral density investigated with respect to growth parameters, This provides an optimised low density of quantum dots for single-dot studies. The nuclear spin behaviour of single quantum dots is investigated through optically- detected nuclear magnetic resonance experiments. A radio frequency electro-magnetic field is applied to the sample during the experiment, and the competition between this route of nuclear spin depolarisation and the nuclear spin polarisation dUE' to an incident flux of spin polarised photons is observed and investigated. The effectivc Knight and Overhauser fields are found and a suitable mathematical model is shown to support the observed phenomena and conclusions, In addition, the effects 011 the emission behaviour of IuGaAs. CuAs and GaAsiAlGaAs quantum dots upon addition of low density dopant atoms such as manganese and iron is t hen studied, It is concluded that such quant um clots doped with single manganese atoms are unlikely to be produced spontaneously in strained material systems but in strain-free material systems an unexpected effect is observed in μPL power dependent emission spectra whereby each line consists of two or three contributions

III- V semiconductor materials are the foundation of many modern electronic and optical devices. In particular, Ill-V semiconductor quantum dots (QDs) are successfully utilized in many applications especially where high optical output (laser diodes) is required. With the aim to improve the performance further, the nanostructure and the growth mechanisms of layer and dot structures used for transistors or laser diodes have to be explored. In this work, a JEOL JEM-2010F field-emission gun transmission electron microscope is applied to investigate the crystallography and chemical composition of InAs or InP QDs at near atomic level using conventional transmission election microscopy (TEM), high resolution TEM, scanning transmission electron microscopy (STEM) and energy-dispersive X-ray spectroscopy (EDXS). The bulk of the thesis concerns the study of InAs/GaAs QD samples, which were grown by molecular beam epitaxy (MBE). An analysis of InAs/GaAs QDs deposited on top of InGaAs or GaAsSb buffer layers is presented. It was found that the GaAsSb buffer, compared to the conventional InGaAs buffer, relaxes more of the strain by introducing misfit dislocations at its lower interface and provides a smoother surface for the subsequent growth. Therefore, the QDs above GaAsSb were fully strained without any dislocations, whereas the strain was partially relaxed by dislocations in the QDs above the InGaAs buffer. The effects of thermal annealing on InAs/GaAs QDs covered with an InGaAs strain-reducing layer were also studied. The influence of the annealing temperature on the degree of interdiffusion between In and Ga atoms was analysed by investigating the change in the size of the QDs. A new technique was used to deduce the indium concentration from series of annular dark-field (ADF) STEM measurements at different collection angles. During annealing, the QDs mainly become flatter and wider. Carbon doping of part of the GaAs barriers has been experimentally found to stabilise both the intensity and the energy of optical emission from those InAs/GaAs QDs upon annealing. ADF Z-contrast STEM imaging suggests the out-diffusion of indium from the QDs is compensated by lateral segregation of indium from the InGaAs layers, which is stronger in the C- doped sample compared to the undoped. An InP QD sample grown by metalorganic vapour phase epitaxy (MOVPE) was successfully utilized to generate 200 femtosecond pulses at ~800 nm wavelength with 2 GHz repetition rates. This sample has been investigated and compared with three other InP QD samples grown with different growth temperatures (640 QC or 720°C), growth rates (0.38 or 0.76 nmlsec) or growth interrupts (0, 10 or 20 sec) using STEM and EDXS. The results propose that InP QDs can be formed by MOVPE at relatively low growth temperatures of - 640 QC and QD formation is favoured by slow growth with growth interrupts.

The aim of this project was to establish the arc discharge in water method as a valid method of producing metal oxide nanoparticles (NPs) with dielectric properties. A vertical arc discharge system was designed and fabricated and several diagnostic tools were attached to the system in order to monitor in situ the propagation of the plasma. Electrical measurements, high speed imaging, optical emission spectroscopy and electrode mass reduction measurements were employed to three different anode materials; aluminum and tantalum rods as well as a mixture of copper-tantalum compressed powder grains. The cathode of the DC arc was in all cases was a carbon rod. The product of the discharge was collected in powder form and examined by means of electron microscopy, x-ray diffraction (XRD), x-ray photoelectron spectroscopy (XPS) and broadband dielectric spectroscopy (BDS). The experiments showed that, during discharge, the physical processes occurring between the electrodes, force the anode material to evaporate in the atomic level. While atomization takes place, the water molecules surrounding the plasma region gets vaporized due to the localized high temperature resulting in the division of oxygen and hydrogen atoms. Metal and oxygen atoms bond, becoming the seeds for nanoparticle formation. The particle growth stops when quenching occurs due to the continuous condensation and expansion of the plasma. High-Resolution Transmission Electron Microscopy (HR-TEM) examination revealed that spherical crystalline nanoparticles are formed, with an average size of 40 nm when using either aluminum or tantalum rod. XRD and XPS analyses concluded that high purity Ah03 and Ta20s NPs are produced by using the arc discharge method respectively. HR-TEM was also employed to the nano-product of the composite anode showing a peculiar core-shell NP structure (mean size 20 nm). Scanning Transmission Electron Microscopy (STEM) was used in High Angle Annular Dark Field (HAADF) imaging and Energy Dispersive X-ray Spectroscopy (EDX) modes to identify the atomic arrangement resulting in the formation of an oxidized copper core with an oxidized tantalum shell. The nano-powder was tested via BDS showing a capacitive behavior. Throughout this work it was been proven that arc discharge in water is a cost- effective and easy to implement method of preparing dielectric NPs. A methodology for studying both the process as well as the product has been described. High purity Ah03 and Ta20s have been successfully produced. CuO- Ta20s core-shell NPs were synthesized for the first time and were characterized structurally and electrically. Adjustments in order to improve the efficiency of the system were proposed and new ideas for the formation of composite metal oxides have emerged.

This dissertation attempts to understand how energy transfer in a molecular wire and a spherical organic assembly are affected by molecular structure. The molecular wire is a DNA-hybrid structure composed of a strand of thymine bases appended by a cyanine dye. Hydrogen bonded to each base is a naphthalene-derivative molecule. Using time-integrated photoluminescence and time-correlated single photon counting measurements, energy transfer from the naphthalene donors to the cyanine acceptors was confirmed, and its dependence on temperature and DNA-template length investigated. Donor-thymine bonding was disrupted at temperatures above about 25 degrees Celcius resulting in poor donor template decoration and low rates of energy transfer. Increasing numbers of donors attach to the scaffold, forming an orderly array, as the template length increases due to the stabilising effects of the donor-donor pi-stacking interactions. Conversely, modelled energy transfer rates fall as the scaffold length increases because of the longer donor-acceptor distances involved. Therefore, the energy transfer rate was greatest for a template built from 30 thymines. The spherical organic assemblies (nanoparticles) are formed by fast injection of a small volume of molecularly dissolved fluorene-derivative amphiphilic molecules into a polar solvent. The amphiphilic molecules contained either a naphthalene (donor) or a benzothiadiazole (acceptor) core. The donor-acceptor mixed nanoparticles resemble an amorphous polymer film and were modelled as such using the Foerster resonance energy transfer theory. The Foerster radii extracted from the measurements depends intricately on the donor-acceptor spectral overlap and distance. The latter effect was controlled by the stacking interactions between the molecules. Altering the morphology of the structural units is the key to optimising energy transfer in molecular structures. To achieve efficient organic molecule-based devices, the importance of this property needs to be fully appreciated and effectively exploited.

Advances in semiconductor technology have been driven by the continuous demands of market forces for IC products with higher performance and greater functionality per unit area. To date industry has addressed these demands, principally, by scaling down device dimensions. However, several unintended consequences have undermined the benefits obtained from the advances in technology, firstly, the growing impact of process variations on interconnectivity delay, aggravated by the increase in the amount of interconnectivity as circuit complexity increases. Overall, the difficulty of establishing delay parameters in a circuit is adversely impacting on the attainment of the timing closure for a design. Secondly, the increase in the susceptibility of the circuits , even at ground level, to the effects of soft errors due to the reduction in supply voltages and nodal capacitances, together with the increase in the number of nodes in a circuit as the functionality per unit area increases. The aim of this research has been to model and analyse the reliability of logic circuits with regard to the impact of process variations and soft errors, and to finds ways to minimise these effects using different process technologies such as fully depleted silicon on insulator (FDSOI) and partially depleted silicon on insulator (PDSOI) technologies, together with the implementation of different circuit architectures. In view of the increased susceptibility of logic elements to the effects of process variations and soft errors as device geometries are reduced, a logic element which is not only widely used but also typical to asynchronous design is the Muller C-element, which can be realised in a number of different circuit configurations. The robustness of various C-element configurations implemented in different technologies with regard to the effects of process variations and soft errors was examined using the design of the experiment (DoE) and response surface (RSM) techniques. It was found that the circuits based on SOI technology were more robust compared with bulk silicon technology. On the other hand, from the circuit architecture perspective, the differential logic implementations of C-element were found to be more resilient to the effects of process variation and soft errors in comparison with the other C-element implementations investigated.

The aggressive scaling of CMOS process technology has been driving the rapid growth of the semiconductor industry for more than three decades. In recent years, the performance gains enabled by CMOS scaling have been increasingly challenged by highlyparasitic on-chip interconnects as wire parasitics do not scale at the same pace. Emerging 3D integration technologies based on vertical through-silicon vias (TSVs) promise a solution to the interconnect performance bottleneck, along with reduced fabrication cost and heterogeneous integration. As TSVs are a relatively recent interconnect technology, innovative test structures are required to evaluate and optimise the process, as well as extract parameters for the generation of design rules and models. From the circuit designer’s perspective, critical TSV characteristics are its parasitic capacitance, and thermomechanical stress distribution. This work proposes new test structures for extracting these characteristics. The structures were fabricated on a 65nm 3D process and used for the evaluation of that technology. Furthermore, as TSVs are implemented in large, densely interconnected 3D-system-on-chips (SoCs), the TSV parasitic capacitance may become an important source of energy dissipation. Typical low-power techniques based on voltage scaling can be used, though this represents a technical challenge in modern technology nodes. In this work, a novel TSV interconnection scheme is proposed based on reversible computing, which shows frequencydependent energy dissipation. The scheme is analysed using theoretical modelling, while a demonstrator IC was designed based on the developed theory and fabricated on a 130nm 3D process.

Colloidal semiconductor photocatalytic materials such as Ti02• α-Fe203 and Sn02 have been used in laboratory-scale pollution abatement systems, reducing I oxidising both organic and inorganic impurities / toxins to chemically useful and / or biologically harmless compounds. For Ti02 there are potentially two distinct photoinduced phenomena: the first is the photocatalytic redox phenomena described above. The second involves photolytically induced high wettability or superhydrophilicity. The longer the surface is illuminated with IN light, the smaller the contact angle formed between the air - water - TiO2 interfaces until, in some cases the angle approaches zero. We have studied Photo-induced SuperHydrophilicity (PISH) on mesoporous TiO2(M-Ti02) films produced from reverse micellar solutions of titanium isopropoxide. M-Ti02 films have been deposited on quartz by spin coating and shown to clearly exhibit phot(}-induced wettability when irradiated with ultra-band gap light. Using a new continuous contact angle measurement technique. the time dependence of the photo-induced wettability changes on the film have been evaluated through the spreading of a sessile water drop. Results obtained at 100% RH, 293 K showed that drops on M-Ti02 exhibited a photo-induced "stick-slip" behaviour. The thermodynamic driving force for this photo-induced stick-slip has been attributed to the departure of the system from capillary equilibrium as the equilibrium contact angle of the drop, θo, decreased with illumination time. Based on this model, a simple theoretical description of photo-induced stick-slip has been derived and used to calculate a value of the potential energy barrier opposing the onset of movement of the triple line, U = 6.26 X 10-6 J m-1. This is the first time that U has been quantified for a surface with photo-induced wenability. Micro gravimetric studies of M-Ti02 layers in humid atmospheres show that, especially at RH > 95%, illumination of the M-Ti02 surface leads to a photo-induced watcr adsorption I condensation at the semiconductor / gas interface. Studies of this photo-induced water condensation on M-Ti02 coated crystals as a function of crystal age / experiment run number indicates that there are two possible modes of condensation - in-pore / in-layer condensation and condensation over the outer surface of the layer.

This project investigates dilute nitrides such as InAsN and InGaAsN with a view to the fabrication of optoelectronic sources and detectors operating in the mid-infrared (2-5 um) spectral range which has many practical applications. Samples of both bulk epitaxial layers and state-of-the-art nanostructures have been grown by liquid phase epitaxy (LPE) and molecular beam epitaxy (MBE). The effect of N incorporation on the material quantum efficiency has been studied using temperature dependent photo- and electroluminescence spectroscopy, high resolution x-ray diffraction and other techniques. InAsN and InGaAsN bulk epilayers were grown by liquid phase epitaxy under standard and neutral solvent growth techniques to investigate the feasibility of high quality material production. Photoluminescence (PL) showed that the dominant peak in the majority of the material was from a defect related transition which quickly quenched leaving just band - band recombination. Raman spectroscopy was employed to identify the types of defect levels present within the material, the strange behaviour of the PL from InGaAsN material was determined to be from a doubly-ionised defect involving a combination of In vacancies and higher order nitrogen complexes. High quality InAsN layers were grown from neutral solvent growth with an XRD full-width-half-maximum (FWHM) of 17.5 arc-seconds obtained from 0.3% N incorporation; this however was only observed in thin layers of the dilute nitride believed to be due to the rapid depletion of N within the growth melt. High quality InAsN was grown onto GaAs substrates using molecular beam epitaxy; PL from this material persisted up to room temperature with a final emission wavelength of 4 urn and nitrogen incorporation of 1%. The temperature dependence of the material was found to be superior to that of similar material grown onto InAs substrates and shown to have an 'S' shaped behaviour at low temperatures originating from tail states caused by inhomogeneous nitrogen incorporation typical of all nitrogen inclusive materials. A comparison with LPE material was then carried out. PL measurements showed that the LPE material had comparable emission intensities with narrower FWHM for low N content samples but MBE proved more favourable for higher N content material with the maximum N inclusion from LPE being 0.5%. The addition of Sb to InAsN multi quantum wells was then studied with the view to device fabrication. The addition of Sb was found to improve both lattice quality and PL intensities while reducing the overall strain of the material. Interpretation of the 4 K PL was shown to be intense with the observation of recombination originating from both the first heavy hole and light hole consistent with a type I band alignment. A model to determine the band alignment of the InAsSbN multiple quantum wells (MQWs) was derived from a single-band Schrodinger solver and found to be in good agreement with the experimental results. Finally InAsSbN light emitting diodes (LEDs) were fabricated and shown to exhibit strong electroluminescence reaching room temperature with final wavelengths of 3.7 urn and the presence of hydrocarbon absorption in the spectrum reveals that this material has potential for gas detection. Output powers of over 3 μW under 100 mA drive current at 50% quasi- continuous operation were obtained leading to an internal efficiency of 0.63%, an improvement over InAsSb and InSb quantum dot LEDs. These prototype devices show promise for type I dilute nitride materials operating in the mid infrared region.

Zinc oxide is a transparent semiconductor with a direct wide band-gap 3.4 eV and large exciton binding-energy of 60 meV, that combine to make ZnO a promising material for possible applications such as optoelectronic devices, lasers and light emitting diodes. Recently, the difficulty in obtaining high quality p-type ZnO has attracted much attention. Considerable effort has been made to obtain p-type ZnO by doping with the group- V elements N, P, As, and Sb, with the anticipation of replacing oxygen atoms in the ZnO lattice. However, experimentally these dopants can produce both p-type and n-type conductivity. Here the results of first principles density functional theory calculations performed using the AIMPRO code are presented. By evaluating the relative energies of substitution on the oxygen and zinc sub-lattices, it is possible to predict the most likely forms of doping centres that might be achieved depending both upon the dopant species and whether the ZnO is grown under oxygen or zinc rich conditions. As a general trend, it is found that dopants tend to be stabilised in environments where covalent bonds with oxygen can be formed, such as substitution on the zinc sub-lattice. The doping properties of the group-V elements can be best understood by not considering the dopant atoms individually, but as a part of an atomic group such as phosphate and nitrate ions either substituting for host atoms, or lying in interstitial sites. The preferential formation of dopant-oxygen bonds leads to a revision of the zinc-vacancy based model for p-type doping (such as P-(VZn)2 complexes) to structures involving interstitial oxygen.

A synthetic route for the general synthesis of Ce(IV) alkoxides is described utilising a reaction with eerie ammonium nitrate and tert-butoxide to create the in situ intermediate [Ce(OBut)4] by means of salt metathesis. A further alkoxy- exchange reaction with the desired alkoxide resulted in the formation of alkoxide complexes [Ce(L)4] (L = mmp, dmae, dmap, dmop, dmomp). The compounds [Ce(mmp)4], [Ce(dmap)4] and [Ce(dmop)4] were characterised crystallographically. [Cetdmae)»] was shown to decompose over time and gave a crystal structure of the cluster [Ce14(dmae)22(N03)2O14(OH)4]. DmomH was found to be an unsuitable ligand for the formation of a homoleptic complex. A range of heteroleptic complexes were also synthesised by reaction of 2 equivalents of an alkoxide (dmop, dmom, dmomp) or β-diketonate ligand (thd, dbm) with the complexes [Ce(rnmp)4] and [Ce(dmap)4]. The thd ligand was found to be too strong a donor to make a heteroleptic complex resulting exclusively in the formation of [Ce(thd)4]. These reactions yielded the complexes [Ce(mmp)(L)2] (L=dbm, dmom, dmomp) and [Ce(dmap)2(dbm)2]. The reaction between [Ce(mmp)4] and 2 equivalents of dmop gave the dimer [Ce2(mmp)2(dmop)4O] which was characterised crystallographically. [Ce( mmp )4] and [Ce(dmap)4] were tested for use as precursors for the liquid- injection MOCVD and ALD of Ce02 thin films on a Si(100) substrate. The results were compared with the known precursors [Ce(thd)4] and [Ce(OC(CH3)2Pr')4], [Ce(thd)4] being the most commonly used precursor for Ce02 thin film growth. Both new precursors were found to be more volatile and have a temperature window of 200-600 QC for MOCVD. [Ce(dmap)4] showed growth at temperatures as low as 100 QC in ALD though was susceptible to decomposition via CVD processes at 250 QC. [Ce(mmp)4] was self-limiting over the temperature range of 150-350 QC. A range of heteroleptic Ti(IV) alkoxide complexes were also synthesised by reaction of either [Ti(OBut)4] or [Ti(OPri)4] with 2 equivalents of the desired alkoxide in an alkoxy exchange reaction to give complexes with the formula [Ti(OR)2(L)2] (R = But, Pri, L = dmae, mmp, dmop, dmom). [Ti(OPri)2(dmop)2] and [Ti(OBut)2( dmop )2] were characterised crystallographically. These compounds were used as precursors for the MOCVD of anatase Ti02 nanostructures on Si(100) and F -doped Sn02 substrates. [Ti(OPri)2( dmae )2] provided the optimum conditions of growth on both substrates at 450 QC for 180 minutes giving full coverage and uniform nanostructures. The precursor also interestingly grew nanorods at 600 QC on Si(100).