Mental health has long been a neglected problem in global healthcare. The social and economic impacts of conditions affecting the mind are still underestimated. However, in recent years it is becoming more apparent that mental disorders are a growing global concern that is not to be trivialised. Considering the rising burden of psychiatric illnesses, there is a necessity of developing novel services and researching effective means of providing interventions to sufferers. Such novel services could include technology-based solutions already used in other healthcare applications but are yet to make their way into standard psychiatric practice. This thesis presents a study on how pervasive technology can be utilised to devise an “early warning” system for patients with bipolar disorder. The system, containing wearable and environmental sensors, would collect behavioural data and use it to inform the user about subtle changes that might indicate an upcoming episode. To test the feasibility of the concept a prototype system was devised, which was followed by trials including four healthy volunteers as well as a bipolar patient. The system included a number of sensory inputs including: accelerometer, light sensors, microphones, GPS tracking and motion detectors. The experiences from the trials led to a conclusion that a large number of sensors may result in incompliance from the users. Therefore, a separate investigation was launched into developing a methodology for detecting behavioural patterns in inputs possible to collect from a mobile phone alone. The premise being that a phone is an everyday use appliance and is likely to be carried and accepted by the patient. The trial revealed that monitoring GPS tracks and Bluetooth encounters has the potential of gaining an insight into a person’s social and behavioural patterns, which usually are strongly influenced by the course of bipolar disorder. Lessons learned during these proceedings amounted to a clearer concept of how a future personalised ambient monitoring system could improve the outcome of treatment of bipolar disorder as well as other psychiatric conditions.

The properties of cubic nitrides grown by molecular beam epitaxy (MBE) on GaAs (001) have been studied using optical and electrical techniques. The aim of these studies was the improvement of the growth techniques in order to improve the quality of grown nitrides intended for bulk substrate and optoelectronic device applications. We have also characterised hexagonal nanocolumn structures incorporating indium. Firstly, bulk films of cubic AlxGa1-xN with aluminium fractions (x) spanning the entire composition range were tested using time-integrated and time-resolved photoluminescence (PL) plus reflectivity measurements. Strong PL emission was recorded from the samples, with improved intensity for higher aluminium concentrations. Temperature dependent and time-resolved PL showed the increasing role of carrier localisation at larger AlN fractions. The reflectivity results showed a near-steady increase in the bandgap energy with increasing AlN content. Alternative interpretations that did and did not involve a transition from direct-gap to indirect-gap behaviour in cubic AlxGa1-xN were considered. We next looked at cubic AlxGa1-xN/GaN/AlxGa1-xN single quantum well (QW) structures with varying AlN content in the barrier regions. The PL studies indicated that carrier escape from the QWs and non-radiative recombination at layer interfaces were limiting factors for strong well emission. Higher AlN concentration in the barriers appeared to exacerbate these problems. The doping of cubic GaN with silicon (n-type) and magnesium (p-type) was also studied. For Mg-doped GaN, a strong blue band emission was noted in the PL spectrum, which became more intense at higher doping levels. The Mg-doped GaN layers had low conductivity and their mobility could not be measured due to strong compensation effects. The cubic film had similar time-resolved PL properties for the blue band emission compared to hexagonal Mg:GaN. These results suggested that the blue band was the result of recombination between a shallow Mg acceptor and deep donor, believed to be a complex including a nitrogen vacancy and an Mg atom. This complex was also associated with the compensation effect seen in the electrical measurements. With the Si-doped cubic GaN, we observed PL spectra that were consistent with other sources. Thicker layers of GaN:Si did not have measurable mobility. This was likely caused by the rough surface structure that was imaged using a scanning electron microscope. The thin layer had a very smooth surface in comparison. The mobility of sub-micron thickness layers with a carrier concentrations of n = 2.0×1018cm-3 and n = 9.0×1017cm-3 were μ = 3.9cm2/Vs and μ = 9.5cm2/Vs respectively. The mobility values and structural issues indicated that growth improvements were needed to reduce scattering defects. In addition to cubic structures, we have considered nanocolumn growth of InGaN and InN. InxGa1-xN nanocolumns were grown on Si (111) by MBE with a nominal indium concentration of x = 0.5. PL emission was obtained from samples grown at higher temperature, but overall intensity was low. A second set of samples, where nanocolumn growth was followed by growth of a continuous coalesced film exhibited much stronger PL emission, which was attributed to the elimination of a phase separated core-shell structure in the nanocolumns. Next, a coalesced InxGa1-xN structure with vertically varying indium fraction was characterised. PL readings showed evidence of successful concentration grading. Finally, the PL spectra of coalesced InN layers were recorded, for which a specialised infrared PL system needed to be used. The results highlighted how increased growth temperature and indium flux can improve PL properties. For the binary alloy however, coalescence growth can decrease PL intensity compared to the nanocolumns stage.

Dilute magnetic semiconductors are an important family of materials that have many potential applications in spintronics; (Ga,Mn)As, (In,Ga,Mn)As and (Ga,Mn)N are of major interest. This thesis investigates dierent aspects of these, using the synchrotron radiation techniques of x-ray magnetic circular dichroism (XMCD) and x-ray magnetic linear dichroism (XMLD), supported by superconducting quantum interference device (SQUID) magnetometry and magnetotransport measurements. A large anisotropic XMLD signal is observed for the Mn L-edge in (Ga,Mn)As. In unannealed (Ga,Mn)As, an apparently reduced Mn magnetic moment is commonly observed. It is thought to be related to compensation of both carriers and magnetic moment, caused by interstitial Mn. This issue is investigated using combined data from XMCD, XMLD and SQUID magnetometry. The findings suggest that substitutional and interstitial Mn form `non-magnetic' pairs which do not have a preferred spin orientation. (Ga,Mn)N is studied by x-ray absorption and field-dependent XMCD at the Mn L-edge. Two distinct Mn congurations are identified: Mn2+ is prevalent towards the surface with nearly paramagnetic behaviour, while a weakly ferromagnetic Mn2+/Mn3+ mixed valence exists within the bulk. The weak ferromagnetism, often observed in (Ga,Mn)N, is attributed to coupling between the impurities by the double exchange mechanism. Finally, XMCD is used to measure the orbital polarization of As 4p states of (III,Mn)As materials. These states correspond to those of the holes involved in the itinerant exchange interaction in ferromagnetic semiconductors. The coupling between the localized d states of the magnetic impurities and the valence band p states of the host is demonstrated by an anisotropy in the orbital moment of these states. This is experimental confirmation of the origin of the magnetocrystalline anisotropy in dilute magnetic semiconductors.

The general aim of the work detailed in this thesis is to improve the quality of electroencepholography (EEG) recordings acquired simultaneously with functional magnetic resonance imaging (fMRI) data. Simultaneous EEG-fMRI recordings offer significant advantages over the isolated use of each modality for measuring brain function. The high temporal resolution associated with EEG complements the high spatial resolution provided by fMRI. However, combining the two modalities can have significant effects on the overall data quality. The gradient artefact (GA), which is induced on the EEG cables by the time varying magnetic fields associated with fMRI sequences, can be particularly problematic to correct for in experiments containing any subject movement. In this thesis, two novel, movement-invariant methods are introduced for correcting the GA. The first method is named the gradient model fit (GMF) and relies upon the assumption that the GA can be modelled as a linear combination of basis components, where the relative weighting of each component varies dependent upon subject position. By modelling these underlying components, it is possible to characterise and remove the GA, which is particularly beneficial in the presence of subject movement. The second method named the difference model subtraction (DMS) relies on the assumption that the GA varies linearly for small changes in subject position. By modelling the change in GA for a basis set of likely head movements, it was shown to be possible to combine DMS with standard GA correction methods to improve the attenuation of the GA for data acquired during subject movement. Both methods showed a significant improvement over the existing GA correction techniques, particularly for experiments containing subject movement. These methods are therefore relevant to any experimenter interested in working with subject groups such as children or patients where movement is likely to occur.

This thesis contributes to the reliability and aging studies of supercapacitors for more efficient use in EV/HEV applications. This thesis demonstrates the effect of aging/failure in supercapacitor cells and module cells using accelerated tests employed to expedite the aging process. The tests, as explained below were categorized based on operational and environmental aging factors associated with supercapacitor failure in EV/HEV applications to; • Investigate supercapacitor cell performance at high temperature and constant voltage individual conditions, and also simultaneously (known as calendar test) • Investigate the effect of voltage balancing/equalization circuits on supercapacitor module cells’ performance during constant current cycling tests under certain environmental and electrical factors • Investigate supercapacitor module cells’ cycling performance in a lab-scale designed electrical DC programmable motor load system that emulates supercapacitor operational conditions in an EV/HEV application. The aging behaviors characterized by the three factors mentioned above are quantified in this thesis through the periodic monitoring of their electrical and electrochemical state of health with Electrochemical Impedance Spectroscopy, Cyclic Voltammetry, and Constant Current characterization tests. These tests help identity aging modes in supercapacitors, and it was observed that regardless of their aging factors; an increase in ESR and decrease of capacitance was determined. Although this information is required, the results from Electrochemical Impedance Spectroscopy (EIS) tests revealed more details distinctive to each aging factor. From this distinction, the aging mechanisms in relation to the aging factors, which causes the deterioration in the supercapacitor electrical performance, are identified and summarized as the following: 1. Loss of contact within supercapacitor electrode, given rise to the contact resistance due to the presence of high temperature as the main aging factor 2. Change of supercapacitor porous electrode emulating a charge transfer reaction thereby increasing its distributed resistance, caused by the effect of high voltage or cycling Mathematical models in the form of electrical equivalent circuits (EECs) distinctive of their aging factors are generated from EIS electrochemical behaviors to easily describe aging behaviors in supercapacitors. The EEC models developed using impedance modeling generated an initial model from dormant cells, which transitioned to aging models distinctive of their aging factors as soon as a 100% increase in ESR and/or an 80% decrease in capacitance is observed. The proposed EEC models were validated to show the dynamic interaction between aging of the supercapacitor cells on their electrical performance in both frequency and time domains. In summary, the EEC models encompass this thesis objective and as such considered the main contribution of this research work.

Over the past years there has been great progression in the field of micro- and nanomechanics with devices with higher and higher Q factors being created. This has been made possible thanks to a combination of advances in fabrication techniques and an increase in understanding as to what causes dissipation in nanometre scale structures. This understanding of dissipation mechanisms is still incomplete however. While lots of work has been done investigating mechanisms such as thermoelastic dissipation and dissipation due to two level systems (TLS) within the standard tunnelling model (STM) a full understanding has not been forthcoming. The increase in the quality of nanomechanical systems has allowed them to be coupled to optical or microwave cavities allowing the position of the mechanical system to be measured with near quantum limited accuracy. This thesis looks at both these streams of research within nanomechanics. It looks at the fabrication of silicon nitride torsional resonators that can have either their flexural or torsional modes preferentially actuated via a piezoelectric drive. It was found for a single paddle resonator that the room temperature Q factor of the flexural mode was 2870±70 and for the torsional mode was 5050±220. It was shown that while thermoelastic damping was reduced in the torsional mode it was still present meaning that we could not use the model for a simple beam to describe thermoelastic damping for a paddle resonator. The properties of an nanomechanical beam fabricated from a single crystal of aluminium were also investigated. It was found that at 1.5 K it had an unloaded Q factor of 36900 which is at least 2 times larger then any other group has reported. We also used our knowledge of high stress silicon nitride membranes to design a system that could couple an aluminium on silicon nitride membrane to a LCR circuit. Calculations show that this would have a coupling constant, g, of over 1000 putting it well within the regime where ground state cooling and quantum limited measurements are possible.

No description available.

Hard disk drives (HDDs) have been the dominant secondary memory device in computing for over 50 years, while more recently magnetoresistive random access memory (MRAM) has emerged as a candidate for primary computing memory. Both HDDs and MRAM store information in the polarity of a magnetic layer, which is written and read by non relativistic mechanisms. There is now gathering interesting in using relativistic mechanisms whose origins lie with spin-orbit coupling (SOC) for MRAM writing because of potential benefits in terms of scalability, device design, and efficiency. This thesis investigates the fundamental physics of SOC phenomena that can write (spin-orbit torque (SOT), Neel order SOT) or read (anisotropic magnetoresistance (AMR), magnetic gating) the magnetic state by the application of electrical current. These phenomena are studied in ferromagnetic and antiferromagnetic semiconducting materials that offer a relevant electrical conductivity for integration into commercial electronic devices. Effective magnetic fields which parametrise the SOT phenomenon are measured in the diluted magnetic semiconductor (Ga,Mn)As using a technique based upon experimental planar Hall effect measurements and analytical fitting with a free energy equation for coherent magnetization rotation. It is found that effective magnetic fields which originate from Dresselhaus SOC increase in magnitude with increasing temperature, whereas those originating from Rashba SO have no significant temperature dependence within experimental uncertainty. The size of the measured effective fields per unit of current density, as well as the ratio of Dresselhaus to Rashba effective field magnitudes averaged over all temperatures are comparable to previous experimental measurements. Sb-based diluted magnetic semiconductors (Ga,Mn (As0.9,Sb0.1) and (Ga,Mn)Sb are characterised by magnetic and transport measurements. The Curie temperature (Tc) of (Ga,Mn)(As0.9,Sb0.1) increases from 28K to 55K upon sample annealing. The Tc of as-grown (Ga,Mn)Sb is found to be 34K, and in contrast to (Ga,Mn)(As0.9,Sb0.1) does not change upon annealing, indicating a lack of interstitial Mn in (Ga,Mn)Sb. Field rotation transport measurements for current along various crystalline directions reveal significant crystalline and non crystalline contributions to the AMR of both as-grown and annealed (Ga,Mn)(As0.9,Sb0.1). An anomalous temperature dependence of the AMR of the annealed (Ga,Mn (As0.9,Sb0.1) sample for current along the [110] crystalline direction is accounted for by considering the relative sizes of the individual AMR contributions as a function of temperature. Results are shown of an attempt to vary the current flow through a non-magnetic GaAs/AlGaAs 2D electron gas (2DEG) by changing the magnetization orientation of an electrically insulated Fe gate layer. Such magnetic gating of electrical current is based upon the principle that, as a result of SOC, the electrochemical potential of a ferromagnet is anisotropic with respect to its magnetization orientation. The magnetic gating experiment proved to be unsuccessful due to an AMR-like signal arising in field rotation measurements of 2DEG samples both with and without the gate layer. The origins of this AMR-like signal are unknown, and it cannot not be accounted for by fitting analysis.

In this thesis we present a study of the fluctuations and noise which occur in a particular nanoelectrical device, the single electron transistor (SET). Electrical transport through the SET occurs through a combination of stochastic, incoherent tunnelling and coherent quantum oscillations, giving rise to a rich variety of transport processes. In the first section of the thesis, we look at the fluctuations in the electrical properties of a SET. We describe the SET as an open quantum system, and use this model to develop Born-Markov master equation descriptions of the dynamics close to three resonant transport processes: the Josephson quasiparticle resonance, the double Josephson quasiparticle resonance and the Cooper-pair resonances. We use these models to examine the noise properties of both the charge on the SET island and the current flowing through the SET. Quantum coherent oscillations of Cooper-pairs in the SET give rise to noise spectra which can be highly asymmetric in frequency. We give an explicit calculation of how an oscillator capacitively coupled to the SET island can be used to infer the quantum noise properties close to the Cooper-pair resonances. To calculate the current noise we develop a new technique, based on classical full counting statistics. We are able to use this technique to calculate the effect of the current fluctuations on an oscillator coupled to the current through the SET, the results of which are in good agreement with recent measurements. In the final part of the thesis we explore the coupled dynamics of a normal state SET capacitively coupled to a resonator in the presence of an external drive. The coupling between the electrical and mechanical degrees of freedom leads to interesting non-linear behaviour in the resonator. We are able to find regions where the resonator has two possible stable amplitudes of oscillation, which can lead to a bistability in the dynamics. We also look at the fluctuations in the energy of the system. We use numerical methods to simulate the dynamics of the system, and to obtain the probability distribution for the work done, whose form can be interpreted by the appropriate fluctuation relation.

This thesis describes the optimisation of the growth of bulk cubic GaN with low hexagonal content and with the intention of making it a commercial substrate for device applications. The optimised material was then applied for fabrication of cubic AlxGa1-xN/GaN based double barrier resonant tunnelling diode (DBRTD) devices. The devices with a clear negative differential resistance (NDR) and high reproducibility are demonstrated. In the early part of this project, we reported a study on cubic GaN material with the variation of III/V ratio, growth rate and wafer position. Using PL and XRD measurements, we found that all these factors influence the increase of hexagonal inclusions in cubic GaN, leading to poor quality in cubic nitride growth. This problem however is more significant when the thickness of cubic GaN is increased. From the calibration work, a ~50μm cubic GaN layer has been grown for the first time with a low average proportion of hexagonal inclusions of around 10% and just few percent at the interface of cubic GaN and GaAs substrate. Thus, the interface would be the most suitable surface for further growth. Next, we investigated the fundamental properties of cubic GaN using picosecond acoustic measurements. In this material, the sound velocity is found to be 6.9±0.1 kms-1, elastic constant = 285±8GPa and the refractive index at 400nm = 2.63±0.04. Comparison with hexagonal GaN films indicated that these parameter values differ considerably in different symmetry of GaN. These show the usefulness of our layers for determination of the basic properties of cubic GaN using a wide range of techniques. From the Hall transport measurement, the electrical properties of undoped cubic GaN samples depend on growth conditions and thickness. In this work, we successfully demonstrated p-type cubic GaN:Mn using C-doping and n-type behaviour from Si-doped cubic GaN. However, these samples have high electron density but low mobility as the residual impurities and intrinsic defects were found to be higher inside the samples. We extended the technology of growing cubic GaN to cubic AlxGa1-xN. A number of cubic AlxGa1-xN samples with different Al content, x were grown and characterised by PL measurement. We found that the hexagonal PL starts to dominate when x is increased, even for thin samples. This could be due to the problems of maintaining cubic AlxGa1-xN growth. It could also be due to the miscibility gap between AlN and GaN. More results and data are required to explain this behaviour. In this thesis, we demonstrated potential cubic GaN substrates for device applications for the first time. The study on bulk cubic GaN showed that the interface between the cubic GaN and the GaAs substrate has only few percent of hexagonal content. Thus, the surface that was in contact with GaAs is the most suitable surface for further processing and growth. Due to the effect of strain, As inclusions and defects were already formed on the surface. By polishing the surface for ~2 hours, these problems were minimised and the surface still had low hexagonal content. In this work, the first working InGaN LED device grown on a polished free-standing cubic GaN substrate has been demonstrated. Our polished cubic GaN substrates also improve the quality of the grown device as been measured by luminescence and I-V characteristics. In the last part of this thesis, we investigated the potential of cubic GaN for developing cubic AlxGa1-xN/GaN DBRTD devices. In the first stage of this work, the I-V characteristics of the cubic tunnel devices were calculated for various band offset, well width, barrier composition and barrier thickness parameters. From this work, optimal designs of cubic AlxGa1-xN/GaN tunnel diodes that could be fabricated and characterised were proposed. The result was then used as a starting point for the growth of cubic AlxGa1-xN/GaN DBRTD. A number of cubic AlxGa1-xN/GaN tunnel devices with different structural parameters were grown. Some devices showed a clear NDR effect but not all of them are reproducible due to breakdown of the device. This factor may also contribute to the irreproducibility of wurtzite (hexagonal) nitride based tunnel diodes in addition to the problem related to charge trapping.