The applicability and functionality of high frequency digital and millimetre wave circuits can be enhanced by the integration of sensor elements into the circuits. It is furthermore advantageous to utilise or modify the pre–existing fabrication process flow in creating this added functionality. This thesis describes a work on magnetic field sensors based on an InP/InGaAs heterojunction bipolar transistor (HBT) which has been fabricated to be compatible with high frequency epilayer structure and processes. In this work, the complete fabrication process for the HBT magnetic sensors has been developed, using standard, transferrable process modules. Ohmic contact metallisations were optimised and D.C. electrical characterisations are also reported upon. The effects of several surface treatments on device performance have been studied and characterised. Surface passivation using two distinct sulphur containing compounds of different phases was shown to enhance performance and an ion bombardment process was developed that degraded surface quality and increased surface leakage currents for enhanced sensor performance. In order to improve the sensitivity of an HBT to magnetic field 3–dimensional magnetic structures were designed to be incorporated onto the surface of the extrinsic base. This design process was informed by simulation of magnetic field profiles of the magnetic elements and fabrication processes were created that would allow for arbitrary 3–dimensional structures. The response to magnetic field applied both parallel and perpendicular to the normal of the wafer of an as–fabricated HBT was investigated. Two different emitter structures were compared, a simple square emitter and a multiple finger emitter, and the ability of the devices to resolve applied field angle was uncovered. The effects of device bias on the field response was also looked at and the optimal bias conditions determined. An analysis of the temperature variation of the magnetic field response was conducted with lower temperatures resulting in higher sensitivity to applied field. Finally, the response of an HBT with integrated 3–dimensional magnetic structures was investigated. A passivated device was found to be less sensitive to applied magnetic field and a device treated with ion bombardment to be more sensitive to magnetic field applied parallel to the normal. The signal to noise ratio for an HBT with integrated magnetic structures was 36.4 dB with an equivalent noise of 0.002 T. The maximum magnetic field strength sensitivity was 0.339 T^(−1) and the maximum magnetic field applied angle sensitivity was 0.119 rad^(−1). The maximum change in normalised D.C. current gain was 0.019. A mathematical description of the change in current gain caused by a given magnetic field applied at a given angle was also determined.

The field of Autonomous Underwater Vehicles (AUVs) has increased dramatically in size and scope over the past two decades. Application areas for AUVs are numerous and varied; from deep sea exploration, to pipeline surveillance to mine clearing. However, one limiting factor with the current technology is the duration of missions that can be undertaken and one contributing factor to this is the efficiency of the propulsion system, which is usually based on marine propellers. As fish are highly efficient swimmers greater propulsive efficiency may be possible by mimicking their fish tail propulsion system. The main concept behind this work was therefore to investigate whether a biomimetic fish-like propulsion system is a viable propulsion system for an underwater vehicle and to determine experimentally the efficiency benefits of using such a system. There have been numerous studies into biomimetic fish like propulsion systems and robotic fish in the past with many claims being made as to the benefits of a fish like propulsion system over conventional marine propulsion systems. These claims include increased efficiency and greater manoeuvrability. However, there is little published experimental data to characterise the propulsive efficiency of a fish like propulsive system. Also, very few direct experimental comparisons have been made between biomimetic and conventional propulsion systems. This work attempts to address these issues by directly comparing experimentally a biomimetic underwater propulsion system to a conventional propulsion system to allow for a better understanding of the potential benefits of the biomimetic system. This work is split into three parts. Firstly, the design and development of a novel prototype vehicle called the RoboSalmon is covered. This vehicle has a biomimetic tendon drive propulsion system which utilizes one servo motor for actuation and has a suite of onboard sensors and a data logger. The second part of this work focuses on the development of a mathematical model of the RoboSalmon vehicle to allow for a better understanding of the dynamics of the system. Simulation results from this model are compared to the experimental results and show good correlation. The final part of the work presents the experimental results obtained comparing the RoboSalmon prototype with the biomimetic tail system to the propeller and rudder system. These experiments include a study into the straight swimming performance, recoil motion, start up transients and power consumption. For forward swimming the maximum surge velocity of the RoboSalmon was 0.18ms-1 and at this velocity the biomimetic system was found to be more efficient than the propeller system. When manoeuvring the biomimetic system was found to have a significantly reduced turning radius. The thesis concludes with a discussion of the main findings from each aspect of the work, covering the benefits obtained from using the tendon drive system in terms of efficiencies and manoeuvring performance. The limitations of the system are also discussed and suggestions for further work are included.

The International Technology Roadmap for Semiconductors (ITRS) specifies that MOSFET logic devices are to be scaled to sub-10nm dimensions by the year 2020, with 32nm bulk devices ready for production and double-gate FinFET devices demonstrated down to 5nm channel lengths. Future device generations are expected to have lower channel doping in order to reduce variability in devices due to the discrete nature of the channel dopants. Accompanying the reduced channel doping is a corresponding increase in the screening length, which is even now comparable with the channel length. Under such conditions, Coulomb scattering mechanisms become increasingly complex as the scattering potential interacts with a larger proportion of the device. Ionized impurity scattering within the channel is known to be an important Coulombic scattering mechanism within MOSFETs. Those channel impurities located close to the heavily doped source and drain or both, will induce a polarisation charge within the source and drain. These polarisation charge effects are shown in this work to increase the net screening of the channel impurities, due to the inclusion of remote screening effects, and significantly decrease the scattering rate associated with ionized impurity scattering. Remote screening can potentially reduce the control by ionized channel impurities over channel transport properties, leading to an increased sub-threshold current. A potential model has been obtained that is based on an exact solution of Poisson’s equation for an ionized impurity located close to one or both of these highly doped contact regions. The model shows that remote screening effects are evident within a few channel screening lengths of the highly doped contact regions. The resultant scattering model developed from this potential, which is based on the Born approximation, is implemented within a Monte Carlo simulator and is applied to MOSFET device simulation. The newly developed ionized impurity scattering model, which allows for remote screening, is applied in the simulation of two representative MOSFET devices: the first device being a bulk MOSFET device developed for the 32nm technology generation; the second device is an Ultra-Thin-Body Double Gate (UTB DG) MOSFET developed for the forthcoming 22nm technology generation. Thorough investigative simulations show that for both the bulk MOSFET and the UTB DG MOSFET, that remote screening of channel impurities in these devices is not a controlling effect. These results prove that the current model for ionized impurity scattering employed in Monte Carlo simulations is sufficient to model devices scaled to at least the 22nm technology node, predicted to be in production in the year 2012.

It has been of major interest in recent research to produce faster optical processing for many telecommunications applications, as well as other applications of high performance optoelectronics. The combination of one-dimensional photonic crystal structures (PhC) and narrow photonic wire (PhW) waveguides in high refractive-index contrast materials such as silicon-on-insulator (SOI) is one of the main contenders for provision of various compact devices on a single chip. This development is due to the ability of silicon technology to support monolithic integration of optical interconnects and form fully functional photonic devices incorporated into CMOS chips. The high index contrast of the combination of a silicon core with a surrounding cladding of silica and/or air provides strong optical confinement, leading to the realization of more compact structures and small device volumes. In order to obtain a wide range of device functionality, the reduction of propagation losses in narrow wires is equally important, although there are still performance limitations determined by fabrication processes. Compact single-row PhC structures embedded in PhW waveguide micro-cavities could become essential components for wavelength selective devices, especially for possible application in WDM systems. The high quality factor, Q, and confinement of light in a small volume, V, are important for optical signal processing and filtering purposes, implying large Purcell factor values. In this thesis, one-dimensional photonic crystal/photonic wire micro-cavities have been designed and modeled using both 2D and 3D versions of the finite-difference time-domain (FDTD) approach. These devices were fabricated using electron beam lithography (EBL) and reactive ion etching (RIE) for patterning of the silicon layer. The device structures were characterized with TE polarized light, using a tunable laser covering the range from 1480 nm to 1585 nm. Single-row periodic hole-type PhC mirrors consisting of identical and equally spaced holes were embedded in 500 nm wire waveguides. Two PhC hole mirrors were separated with a cavity spacer varying from 400 nm to 500 nm in length to form a micro-cavity. In contrast, several different cavity arrangements were also successfully investigated, - i.e. extended cavity and coupled micro-cavity structures. The experimental results on photonic crystal/photonic wire micro-cavity structures have demonstrated that further enhancement of the quality-factor (Q-factor) - up to approximately 149,000 at wavelengths in the fibre telecommunications range is possible. The Q factor values and the useful transmission levels achieved are due, in particular, to the combination of both tapering within and outside the micro-cavity, with carefully designed hole diameters and non-periodic hole placement within the tapered sections. On the other hand, a large resonance quality factor of approximately 18,500, together with high normalized transmission of 85% through the use of tapering on both sides of the hole-type PhC mirrors that formed the micro-cavity, has been obtained. For the extended cavity case, the multiple resonances excited within the stop band, together with substantial tuning capability of the resonances obtained by varying the cavity length has been demonstrated, together with a Q-factor value of approximately 74,000 at the selected resonance frequency with a normalised transmission of 40%. In addition, the coupled micro-cavity structures considered in this thesis have formed the basic building block for designing multiple cavity structures where the combination of several cavities splits the selected single cavity resonance frequency into a number of resonances that depends directly on the number of cavities used in the design. The coupling strength between the resonators and the Free Spectral Range (FSR) between the split resonance frequencies of the coupled cavity combination were controlled via the use of different numbers of periodic hole structures – and through the use of different aperiodic hole taper arrangements between the two cavities in the middle section of the mirrors.

Wind power is growing rapidly around the world as a means of dealing with the world energy shortage and associated environmental problems. Ambitious plans concerning renewable energy applications around European countries require a reliable yet economic system to generate, collect and transmit electrical power from renewable resources. In populous Europe, collective offshore large-scale wind farms are efficient and have the potential to reach this sustainable goal. This means that an even more reliable collection and transmission system is sought. However, this relatively new area of offshore wind power generation lacks systematic fault transient analysis and operational experience to enhance further development. At the same time, appropriate fault protection schemes are required. This thesis focuses on the analysis of fault conditions and investigates effective fault ride-through and protection schemes in the electrical systems of wind farms, for both small-scale land and large-scale offshore systems. Two variable-speed generation systems are considered: doubly-fed induction generators (DFIGs) and permanent magnet synchronous generators (PMSGs) because of their popularity nowadays for wind turbines scaling to several-MW systems. The main content of the thesis is as follows. The protection issues of DFIGs are discussed, with a novel protection scheme proposed. Then the analysis of protection scheme options for the fully rated converter, direct-driven PMSGs are examined and performed with simulation comparisons. Further, the protection schemes for wind farm collection and transmission systems are studied in terms of voltage level, collection level  wind farm collection grids and high-voltage transmission systems for multi-terminal DC connected transmission systems, the so-called “Supergrid”. Throughout the thesis, theoretical analyses of fault transient performances are detailed with PSCAD/EMTDC simulation results for verification. Finally, the economic aspect for possible redundant design of wind farm electrical systems is investigated based on operational and economic statistics from an example wind farm project.

A spectrum of phenomena related to the reliability of AlGaN/GaN HEMTs are investigated in this thesis using numerical simulations. The focus is on trap related phenomena that lead to decrease in the power output and failure of devices, i.e. the current collapse and the device degradation. The current collapse phenomenon has been largely suppressed using SiN passivation, but there are gaps in the understanding of the process leading to this effect. Device degradation, on the other side, is a pending problem of current devices and an obstacle to wide penetration of the market. Calibration of I-V measurements of two devices is performed with high accuracy to provide a trustworthy starting point for modelling the phenomena of interest. Traditionally, in simulations of nitride based HEMTs, only direct piezoelectric effect is taken into account and the resulting interface charge is thence independent of the electric field. In this work, the impact of the electric field via the converse piezoelectric effect is taken into account and its impact on the bound charge and the drain current is studied, as a refinement of the simulation methodology. It is widely believed that the current collapse is caused by a virtual gate, i.e. electrons leaked to the surface of the device. We have found a charge distribution that reproduced the I-V measurement that shows current collapse, hence validating the concept of the virtual gate. While it was previously shown that the virtual gate has a similar impact on the I-V curve as is observed during the current collapse, we believe that this is for the first time that a wide range of gate and drain voltages was calibrated. High gate/drain voltage leading to permanent degradation was also investigated. The hypothesis that stress induced defects and dislocations might be responsible for the degradation was tested but not fully confirmed. Finally, the leakage of electrons thought to be responsible for formation of the virtual gate and the current collapse due to the Poole-Frenkel emission, is simulated in order to explain the surface charge distribution responsible for the current collapse and deduced in Chapter 5.

The aim of this thesis is to dielectrophoretically characterise sub-micron particles on the basis of their surface properties and to devise a DEP technique suitable for the fractionation and manipulation of particles on this scale. Polystyrene particles are modified by the attachment of biological ligands using various established localisation techniques and their DEP response observed using micro-electrodes with well defined high and low field regions, corresponding to a previously utilised design and modified in the course of this project for multiple sample handling. The results of these observations are modelled for the first time using a charge relaxation mechanism pertaining to a structured interfacial charge distribution and, through fitting the data to this model, fundamental parameters of the system - the surface conductance and electrokinetic charge - are predicted. The model viability is assessed with reference to both comparisons with alternative measurements and the technical limitations of the data fitting procedure, and corresponding surface charge transport mechanisms are discussed in the light of the DEP response following surface modification. Investigations are made into the possibility of a DEP based device suitable for the transport/fractionation of sub-micron particles. Given the essentially dissipative nature of sub-micro particle ensembles, a Brownian ratchet principle is chosen. A Brownian ratchet is a generic system wherein a net directional drive is effected by biasing Brownian diffusion on a periodically activated anisotropic structure. Without need of thermal gradients or net macroscopic forces Brownian ratchet pumps could be an interesting alternative in many microfluidic applications. Simulated fields and corresponding particle transport rates are compared for two basic electrode structures in order to assess their viability for use as DEP Brownian ratchets and a new design proposed, based on the simultaneous juxtaposition of positive and negative DEP forces. This device is built on the necessary scale using multi-layer fabrication techniques with a silicon elastomer moulded channel. The existence of stochastic transport on the device is investigated experimentally by means of processed video sequences and resulting possibilities for particle separation on the basis of size and surface properties inferred.

This thesis provides an investigation of the dynamics and control of turbogenerators from a multivariable control viewpoint. The multivariate control framework chosen -Individual Channel Analysis and Design- is particularly appropriate since it encapsulates the dynamical characteristics of the uncontrolled system with a view to exposing the potential and limitations for subsequent closed-loop control. The main contribution of the thesis is a complete new insight into why excitation/governor control with Power System Stabilisers (PSS) has been so successful for the control of turbogenerators connected to an infinite bus provided by the small-signal multivariable analysis framework, Individual Channel Analysis and Design. The multivariable analysis justifies treating the turbogenerator system as a pseudo- Single-Input Single-Output, (SISO) system where the governor loop is first closed and the exciter loop is treated as a SISO system for the prime purpose of rejecting voltage disturbances. The function of the PSS is identified as that of overcoming an awkward switch-back frequency-domain characteristic of the excitation channel so as to permit high-performance excitation channel bandwidths up to 10 rad/sec that otherwise could not be obtained. Thus, in addition to the control requirements of set point regulation of the terminal voltage and shaft speed, the PSS provides for a second control requirement of strong voltage disturbance rejection over the important frequency range of 0 to 10 rad/sec. The PSS control option is also assessed against other control options. Several other results concerning stability robustness to system uncertainties in different system configurations follow from the analysis in a transparent and immediate way.

Over the past 5 years there has been an increase in the number of applications that require devices that operate in the millimetre range (30-300GHz). This demand has driven research into " devices that will operate at frequencies above 100GHz. This performance has been achieved using two main technologies, the Heterojunction Bipolar Transistor (HBT) and the High Electron Mobility Transistor (HEMT). At present it is a HEMT device that holds the record for the highest operating frequency of any transistor. It is this technology that this project concentrates on. In order to fabricate devices that operate at these frequencies two methods are commonly employed. The first is to vary the material of the device, in particular, increasing the indium content of the channel. The second method is to reduce the physical dimensions of the transistors, including reducing the gate length of the device therefore reducing transit time and gate capacitance. Reducing the separation of the source-drain ohmic contacts or employing a self-aligned ohmic strategy reduces the associated parasitic resistances. This project will concentrate on the scaling of the gate length in addition to the reduction of parasitic resistances with the use of self-aligned ohmic contacts. This work includes the realisation of the first self-aligned 120nm T -Gate. GaAs pHEMT fabricated at the University of Glasgow. These devices required the development of two key technologies, the non-annealed ohmic contact and the succinic acid based selective wet etch. The self-aligned devices showed good RF performance with a ft of 150 GHz and a fmax of 180 GHz which compares favourable with results o~ 120nm GaAs pHEMTs previously fabricated at Glasgow. The investigation of gate length scaling to device performance included the development of two lithographic process capable of producing HEMT with a gate length of 50nm and 30nm respectively in addition to a method ~f sample preparation that allows these devices to be analysed using TEM techniques. This work has lead to the realisation of SOnm T -gate metamorphic HEMTs using a PMMAIcopolymer resist stack, these devices displayed an excellent yield, with over 95% of devices working. The uniformity of the gate process was also high with a threshold voltage of - 0.44SV with a standard deviation of O.OOSV. The devices demonstrated an .it of 330GHz and a fmax of 260GHz making these devices some of the fastest transistors that have ever been fabricated on a GaAs substrate. The second lithography process was developed to realise T -gates with a gate length of less than SOnm. This processed used a two stage "bi-lithography" process to minimise the effect of forward s7attering through the resist. The gate footprint was transferred into a Si02 gate by a dry etch process. This lithography process was integrated into a full process flow for lattice matched InP HEMTs Using this process, HEMTs were fabricated with a T-gate of 2Snm. This is the smallest T -gate device that has been fabricated at the University of Glasgow and is comparable with the smallest HEMT devices in the world.

No description available.