Resistive switching random access memory devices have attracted considerable attention due to exhibiting fast programming, non-destructive readout, low power-consumption, high-density integration, and low fabrication-cost. Resistive switching has been observed in a wide range of materials but the underpinning mechanisms still have not been understood completely. This thesis presents a study of the leakage current and resistive switching mechanisms of SrTiO3 metal-insulator-metal devices fabricated using atomic layer deposition and pulse laser deposition techniques. First, the conduction mechanisms in SrTiO3 are investigated. The leakage current characteristics are highly sensitive to the polarity and magnitude of applied voltage bias, punctuated by sharp increases at high field. The characteristics are also asymmetric with bias and the negative to positive current crossover point always occurs at a negative voltage bias. A model comprising thermionic field emission and tunnelling phenomena is proposed to explain ii the dependence of leakage current upon the device parameters quantitatively. SrTiO3 also demonstrates bipolar switching behaviour where the current-density versus voltage (J-V) characteristics show asymmetry at all temperatures examined, with resistive switching behaviour observed at elevated temperatures. The asymmetry is explained by the relative lack of electron traps at one electrode, which is determined from the symmetric J-V curve obtained at room temperature due to the redistribution of the dominant electrical defects in the film. Evidence is presented for a model of resistive switching that originates from defect diffusion (possibly oxygen vacancies) at high temperatures. Finally, a peculiar resistive switching behaviour was observed in pulse laser deposited SrTiO3. This switching depends on both the amplitude and polarity of the applied voltage, and cannot be described as either bipolar or unipolar resistive switching. This behaviour is termed antipolar due to the opposite polarity of the set voltage relative to the previous reset voltage. The proposed model based on electron injection by tunnelling at interfaces and a Poole-Frenkel mechanism through the bulk is extended to explain the antipolar resistive switching behaviour. This model is quantified by use of a simple mathematical equation to simulate the experimental results.

In this work we explore compositional methods for design of digital circuits with the aim of improving existing methodoligies for desigh reuse. We address compositionality techniques looking from both structural and behavioural perspectives. First we consider the existing method of handshake circuit optimisation via control path resynthesis using Petri nets, an approach using structural composition. In that approach labelled Petri net parallel composition plays an important role and we introduce an improvement to the parallel composition algorithm, reducing the number of redundant places in the resulting Petri net representations. The proposed algorithm applies to labelled Petri nets in general and can be applied outside of the handshake circuit optimisation use case. Next we look at the conditional partial order graph (CPOG) formalism, an approach that allows for a convenient representation of systems consisting of multiple alternative system behaviours, a phenomenon we call behavioural composition. We generalise the notion of CPOG and identify an algebraic structure on a more general notion of parameterised graph. This allows us to do equivalence-preserving manipulation of graphs in symbolic form, which simplifies specification and reasoning about systems defined in this way, as displayed by two case studies. As a third contribution we build upon the previous work of CPOG synthesis used to generate binary encoding of microcontroller instruction sets and design the corresponding instruction decoder logic. The proposed CPOG synthesis technique solves the optimisation problem for the general case, reducing it to Boolean satisfiability problem and uses existing SAT solving tools to obtain the result.

No description available.

In this thesis, quinone derivative molecules supported on a Cu(110) surface are studied using scanning tunnelling microscopy (STM). The experimentally investigated system is based on the bistable nature of these compounds, and so the work is introduced in the wider context of molecular electronics (Chapter 1). The theory and experimental techniques are also described (Chapters 2 and 3). In Chapter 4 the switching behaviour of azophenine (AP) and azotolyline (AT) is characterised using STM imaging and spectroscopy, and is demonstrated to be based on a hydrogen tautomerisation reaction. The activation energy for switching is quantified by measurement of the rate of switching as a function of varying bias voltage, and the process is determined to be stimulated by inelastic electron tunnelling. The reaction pathway is also revealed using theoretical modelling. Chapter 5 focusses on the condensed phase of AP on the Cu(110) surface. The switching behaviour is found to be largely quenched in the self- assembled phase, so statistical analyses of the AP-AP and AP-Cu interactions are conducted in order to try to explain this. Chapter 6 returns to the study of isolated AP molecules and investigates the spatial dependence of the switch with respect to the location of electronic excitation. It is shown that the final state of the molecule can be accurately selected by exciting specific functional groups within the molecule. This control originates from the non- degenerate reaction pathways for the sequential transfer of the two tautomeric protons. The work is then discussed in terms of future outlook and potential applications for bistable molecules.

The purpose of this study was to develop new FSS based microwave absorber designs to minimise the physical thickness, increase the bandwidth and provide radar backscatter suppression that is independent of the wave polarization at large incident angles. A new low cost, accurate and rapid printing technique is employed to pattern the periodic arrays with the precise surface resistance required for each of the FSS elements to optimize the performance of this class of absorber. The electromagnetic behaviour of five new FSS based structures, two standalone arrays, and three absorber arrangements, have been studied using CST Microwave Studio software. The FSS structures consist of two closely spaced arrays of rings with the conductor split at one or two locations to provide independent control of the resonances. By careful design these are shown to exhibit coincident spectral transmission responses in the TE and TM plane. Based on this design methodology, a very thin 4-layer metal backed resistively loaded rectangular loop FSS absorber which works from 0° - 22.5° is shown to give a wide band performance that is independent of the orientation of the impinging signals. To reduce the manufacturing complexity, a single layer FSS absorber which operates at 45° incidence has been designed to give a polarisation independent performance by employing an array of rectangular split loops with discrete pairs of resistive elements of unequal value inserted at the midpoint of the four sides. A major increase in bandwidth is obtained from a single layer FSS absorber which is composed of an array of nested hexagonal loops. Moreover the use of the same surface resistance for all four elements in the unit cell is shown to significantly simplify the construction of the structure which was designed to provide radar cloaking from 0° to 45° incidence. A new manufacturing strategy is presented, where the required surface resistances are obtained by employing an ink-jet printer to simultaneously pattern the FSS elements on the substrate and digitally control the dot density of the nano silver ink and aqueous vehicle mixture.

The key objective of this study was to investigate and develop various types of ground plane architectures for wideband planar spiral antennas. A two-arm, four-turn Archimedean spiral, which is designed to operate between 3- 10 GHz, and five new and innovative cavity arrangements have been studied using CST Microwave Studio® software. Each of the structures is used as means to provide transformation from bidirectional to unidirectional radiation and simultaneously increase the gain of the antenna, whilst exhibiting a good impedance match at the operating frequencies. A simple flat metal plate is modified to reduce modal contamination by inserting slots carefully arranged in a radial pattern to disrupt the current that flows on the surface of the reflector. More complex stepped ground plane and FSS cavity arrangements are designed to operate in conjunction with the spiral over the frequency range 3-10 GHz without the need to mechanically reposition the reflector to optimize the performance at each frequency. The stepped cavity is composed of eight metal rings each positioned λ/4 below the corresponding active region of the spiral, whilst the latter arrangement consists of a two-layer FSS and a metal plate which reflect signals in the upper (7- 10 GHz) and lower (3-6 GHz) frequency band, respectively. Two High Impedance Surfaces (HIS) designs are studied and shown to provide a more compact cavity compared to the previous arrangements; 1) selective loading by placing dissimilar HIS below the 3 and 6 GHz active region of the spiral, and 2) a uniformly distributed multi-resonant HIS for a tri-band WLAN antenna. Simulated and measured radiation patterns and key performance metrics are shown to be in good agreement for all five ground plane architectures. Following this, the results of a pilot study are presented to demonstrate that a reconfigurable HIS ground plane based on electronically tunable liquid crystals, can provide either a monopulse Sum (∑) or a Difference (∆) shaped beam by dynamically switching the permittivity of the tunable substrate between two states

The joint time-frequency (TF) domain provides a convenient platform for signal analysis by involving the dimension of time in the frequency representation of a signal. A straightforward way to acquire localized knowledge about the frequency content of the signal at different times is to perform the Fourier transform over short-time intervals rather than processing the whole signal at once. The resulting TF representation is the short-time Fourier transform (STFT), which remains to date the most widely used method for the analysis of signals whose spectral content varies with time. Recent application examples of the STFT and its variants – e.g. the squared magnitude of the STFT known as the spectrogram – include signal denoising, instantaneous frequency estimation, and speech recognition. In this thesis, we first address the main limitation of the trade-off between time and frequency resolution for the TF analysis by proposing a novel adaptation procedure which properly adjusts the size of the analysis window over time. Our proposed approach achieves a high resolution TF representation, and can compare favorably with alternative time-adaptive spectrograms as well as with advanced quadratic representations. Second, we propose a new scheme for the time-frequency adaptation of the STFT in order to automatically determine the size and the phase of the analysis window at each time and frequency instant. This way, we can further improve the resolution of the conventional as well as the time-adaptive spectrograms. Finally, we focus on denoising non-stationary signals in the STFT domain. We introduced an optimized TF mask in the STFT domain, which is based on the concept of the multi-window spectrogram. Experimentation has shown that the introduced approach can effectively recover distorted signals based on a small set of representative examples of the noisy observation and the desired signal.

This thesis describes the design and the study of implementing microwave power limiters with Gallium Arsenide (GaAs) pseudomorphic high electron mobility transistors (PHEMTs). Circuits were fabricated in microwave integrated circuits (MICs) and monolithic microwave integrated circuits (MMICs) format, tested for performance at 2GHz and 10GHz respectively. The majority of microwave power limiters offered in the current market are based on the p-i-n or PIN diode for protection. Generally in the RF industries, very high power applications still utilize solid-state devices such as T-R tubes for power limiters due to their electrical robustness. However, with the ever growing urge for product improvements and the maturity of MMICs, the lack of integration of the PIN diodes with MMICs calls for an investigation into alternate approaches in power limiter design. Many switching applications in MMICs are implemented using pHEMTs in both series and shunt orientation. The Schottky diode is the favoured choice for microwave power detection purposes. The microwave power limiter approach suggested in this thesis utilizes the pHEMTs as a voltage controlled attenuator in conjunction with a Schottky diode based detector as the control voltage supplier. The MICs or hybrids have been designed using Avago's general purpose pHEMT ATF-35143 as the major element in the voltage controlled switch attenuator sub circuit. The voltage controlled attenuator is supplied with a control voltage from a detector or combination of detector and external bias, depending on the switch's configuration. Avago's Schottky diode HSMS-8101 was used as the detector diode in the power detecting sub circuit. The MMICs were designed using the FD05 O.251lm pHEMT process design kit from Filtronic Compound semiconductor PLC, Newton Aycliffe, Durham, United Kingdom. The pHEMT epi-layers were grown by molecular beam epitaxy on 150mm semi-insulating substrates, featuring O.51lm gate length and 50llm gate width. It is found that larger device provides lower loss but lower isolation. The Schottky diode used in fabricated MMICs is a single gate and 20llm wide device. In the design approach adopted here, the junction of the voltage controlled attenuator and detector is realized using aT-junction. Other methods used include the proximity coupler and capacitive power divider network. Each coupling method finds merit depending on the switch configuration of the attenuator. Eight types of MIC limiters were made, including the single series switch, double series switch, single shunt switch, double shunt switch, series-shunt switch, shunt-series switch, n-switch and tee-switch limiters. The lowest insertion loss achieved was -O.34dB while the isolation is 8dB, delivered by the single shunt switch, measured at 2GHz. The best isolation measured was 20dB, seen in the tee-switch and the double series switch limiters. From the fabricated MMIC limiters, 3.03dB insertion loss and 15dB isolation is provided by the single series switch limiter. While the double series switch limiter is seen to deliver 5.25dB insertion loss and 25dB isolation. Both MMIC limiters exhibit the same limiting threshold level at 10dBm on X-band.

Resonant tunnelling diode (RTD) oscillators are considered to be one of the most promising solid-state terahertz sources which can operate at room temperature. The main limitation of RTD oscillators up to now is their low output power. For the published terahertz (THz) RTD oscillators, the output power is in the range of micro-Watts. This thesis describes systematic work on RTD device modelling, and on the design, fabrication and measurement of high power monolithic microwave integrated circuit (MMIC) RTD oscillators. The RTD device consists of a narrow bandgap layer (quantum well) sandwiched between two thin wide bandgap layers (barriers). When the device is biased, electrons with kinetic energy lower than the barriers may tunnel through the double barrier-quantum well (DBQW) structure, and the device exhibits a negative differential resistance (NDR) in this case. To investigate this phenomenon, a new numerical model based on quantum mechanics was developed. The model involves self-consistent solving of the Schrodinger equation until the quasi-eigen energy state converges. This model is expected to serve to optimize the RTD device structure for millimetre-wave and terahertz applications. Besides RTD device modelling, the fabrication process for single devices and for MMIC RTD oscillator circuits was developed and optimized on this project. Optical lithography, wet/dry etching and metallization were the main fabrication techniques utilized. For device sizes of a few square microns, the fabrication process required the development of new steps, i.e, via opening through polyimide. The fabrication process was optimized and high yield was obtained. On this project, one of the challenges was to realize RTD oscillators in the form of MMICs, aiming at about 100 GHz with milli-Watts output power, in accordance with a recently proposed power combining circuit topology. To accomplish such an oscillator, proper design of passive components was essential. On this project, these components included 50 coplanar waveguides (CPW), shorted CPW stubs, metal-insulator-metal (MIM) capacitors and thin _lm resistors. The design procedure for these components is described, and their performance characterized by DC or scattering parameter measurements as appropriate. Two types of MMIC RTD oscillator layouts were designed, fabricated and characterized. Details are described in this thesis. Measurement results showed that for the fabricated 75 GHz oscillator, the output power obtained was -0.2 dBm (0.96 mW), and for the 86 GHz oscillator, the measured output power was -4.6 dBm (0.35 mW), both of which, to the author's knowledge, were the highest power for published indium phosphide (InP)-based RTD oscillators operating in the W-band frequency range.

No description available.