This thesis presents the design and fabrication techniques for miniature microwave bandpass filters using multilayer liquid crystal polymer (LCP) technology. As a multilayer technology for microwave devices, LCP is of low cost and light weight. It also has excellent electrical properties across a wide frequency range. These characteristics make it promising for the development of next generation microwave devices for applications across commercial, defence and civil sectors. However, very limited work has been found in the open literature to apply this technology to the design of miniature bandpass filters, especially at low microwave frequencies. In addition, the reported work shows lack of fabrication techniques, which limits the size reduction of multilayer LCP devices. To address these problems, this thesis develops advanced fabrication techniques for sophisticated LCP structures, such as multilayer capacitors, via connections and cavities. These techniques are then used to support the design of novel miniature bandpass filters for wideband and narrowband applications. For the design of miniature wideband bandpass filters, a cascaded approach, which combines highpass and lowpass filters, is presented first to provide a flexible design solution. This is followed by another novel ultra-wideband bandpass filter which produces extra transmission zeroes with minimum number of elements. It does not only have high performance but also a compact structure for high yield fabrication. For narrowband applications, two types of advanced coupled-resonator filters are developed. One type produces a very good selectivity at the upper passband edge, and its spurious-free stopband is extremely wide and of high interference attenuation. The other type, based on novel mixed-couplings approaches developed in this thesis, provides a solution to produce almost the same response as the coupling matrix prototype. This type is used to generate arbitrarily-located transmission zeroes. All designs presented in this thesis are simulated using CAD design tools and then validated by measurements of fabricated samples. Good agreements between simulations and measurements are shown in the thesis.

The thesis is concerned with an investigation of several aspects of the short-circuit currents which arise in turbogenerators. Of particular interest is the phenomenon that one or more of the armature currents may not pass through zero for a considerable period following a fault at the armature terminals, so that the generator breaker may face an extremely arduous duty. Rather than employing the familiar 2-axis analysis, the numerical study in the thesis is based on a general-purpose digital phase model of the generator, since this is ideally suited to the problems investigated.

During the last decade the use of 8162 as gate dielectric layers in complementary metal oxide semiconductor (CMOS) microelectronic devices has become increasingly problematic due to leakage resulting from the electron tunnelling with gate oxide thickness approaching 1 nm. Approaches to deal with these problems have focused on increasing the dielectric constant (k) of the material, initially though nitridation of the oxide layer and more recently the application of high-A: materials such as Hf based dielectrics. The work described in this thesis concerns the physical characterisation of thin high-A: multilayered samples using medium energy ion scattering (MEIS). A MEIS computer simulation model was applied and adapted to enable the interpretation of depth profiles from MEIS energy spectra. Forming part of an EU collaborative project, results obtained were compared to those of X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), and X-ray fluoresence (XRF) to provide a better overall understanding of the characteristics of the layers. Nanometre thin SiC>2 layers nitrided using a novel plasma nitridation technique were investigated, demonstrated the nitridation of, and yielded the N distributions in the SiC>2 samples as well as demonstrating plasma damage. An improved k value was found, leading to an increased equivalent oxide (EOT) thickness. Studies of HfO2 and HfSiOx nanolayers, both with and without subjection to a decoupled plasma nitridation (DPN) process were carried out, characterising the layer structures with an accuracy of 0.1 nm in excellent agreement with the additional techniques. Crystallisation of the HfO2 layers, but not of the HfSiOx layers, after DPN was demonstrated. A high-A; metal gate Si/SiO2/HfO2/Al2O3/TiN stack was also investigated and Hf/Al interdiffusion demonstrated upon annealing. Finally Si/TiN/STO layers grown using different stoichiometric recipes, with and without a rapid thermal anneal at 650°C for 15s, were analysed. Layer structures were again determined with sub-nm resolution and diffusion between the Sr and Ti layers was observed after annealing. The high level of agreement between the depth profiles derived from the MEIS energy spectra, the growth parameters and the results from additional techniques has demonstrated the capability of MEIS in combination with spectrum simulation for the accurate analysis of these demanding ultra thin layer structures.

Materials used in technology, from the development of nano-electronics to the efficient conversion of solar energy to electrical power are under constant optimisation. The factors that govern the size of field effect transistors, such as leakage current through the gate or the on-state power consumption, and those that impact upon the efficiency of optoelectronic devices such as non-radiative recombination and device aging arising from defect migration are specific to the elements and compounds deployed in these devices. Defects, whether intentionally incorporated such as electrical doping, or unintentional contamination such as oxygen in silicon, may have a qualitative impact upon the materials properties (e.g. conductivity), or limits in processing that need to be accommodated in both the manufacture of devices, and in their operation. Device miniaturisation for CPUs and photovoltaic efficiency improvements are two areas developing in parallel for which a detailed understanding of the composition and evolution of defects, dopants and impurities are particularly important. In the project presented in this thesis, first principles density functional calcu- lations within a supercell approach have been performed, with particular focus on selected impurities in a range of technologically relevant materials, where the impu- rities are likely to be incorporated due to their presence in the growth environment. The host materials are divided into those relevant for nano-electronic devices and high capacitance structures, being the perovskite titanates (SrTiO3, BaTiO3 and PbTiO3), and material that has a role in the energy sector in photovoltaic cells, being cadmium telluride (CdTe). In the case of the titanate films, it is important to recognise that thin films are often grown using organic precursors, and therefore carbon contamination is of key importance. Although it is generally assumed that carbon is incorporated in such materials in the form of carbonates, the results of the calculations performed for this study challenge this, showing that a distribution of multiple sites occurs, depending upon growth conditions and the Fermi level. Observable calculated for these struc- tures, including their electrical properties and vibrational modes are presented to aid the identification of the carbon sites in future experimental studies. Critically, car- bon substitution at the Sr, Ba or Pb site leads to electrical effects not present for Ti substitution. The similarities and differences between the three titanates are reviewed. For CdTe, it has long been understood that oxygen is a common impurity grown in thin-films, occurring in high concentrations across a range of growth methods. Vibra- tional modes observed in experiment at 1096.78 and 1108.35 cm−1 have been variously assigned to oxygen-containing point defects, but most recently to SO2 molecules dis- solved in the lattice. However, the precise structure and location of these centres, as well as the electrical properties of the defect, are yet to be determined. As with the analysis of carbon in the titanates, density-functional simulations of SO2 in various locations in CdTe show that several possible structures are low in energy, with the equilibrium form depending upon growth conditions and the Fermi-level. It is shown that a plausible candidate for the vibrational centre is an interstitial species, based upon the frequencies and isotopic splittings.

To date, investigations concerning single or small groups of molecules for molecular electronics have largely focussed upon the transfer of electrons from donor to acceptor (electrodes or redox centres), through a single molecular pathway (bridging ligand or linear structure). Herein is described the synthesis and electrochemical properties of some branched and macrocyclic compounds containing {FeCp 2 } and {Ru(dppe) 2 } centres, with phenylene ethynylene backbones and pyridyl-termini for surface binding (Chapters 2 and 5 , respectively). Such systems provide two well-defined molecular pathways for electron transfer, facilitating the study of quantum interference effects and other phenomena resulting from concurrent, and ultimately convergent, electron transport. In addition, the large-scale 'oxidative purification' of 1,1?-diiodoferrocene ( 1 ) is presented (Chapter 2) , its usefulness as a starting material being further enhanced by an investigation into the optimisation of its (typically low-moderate yielding) reactions with terminal alkynes under Sonogashira cross-coupling conditions (Chapter 3) . As part of this study, attempts to synthesise 1,1?-bis[(4-thioacetylphenyl)ethynyl]ferrocene from 1 and 4-ethynylphenylthioacetate unexpectedly produced a cyclic trimer (15) of the latter. This observation afforded an explanation as to why the Fc?C?C?C6H4?SAc motif may not be formed using this approach, and resulted in the realisation of a new route to ?-phenylthioketones in general. The linear complex Ru(dppe)2(C?C?C5H4N)2 (17) has also been prepared - a synthesis complicated by the bifunctional nature of the 4-ethynylpyridine ligand (Chapter 4). It is hoped that future conductance studies of this and extended structures will complement recent work by others on analogous complexes with isonitrile and thiolate-termini. Experiences with protecting the pyridyl nitrogen of 4-ethynylpyridine, and the moderate success of employing N-methyl-4- ethynylpyridinium triflate (19) as a ligand, are detailed. Finally, 'mixed-valence' complexes of the type [{M(L) n}2(?-BL)] (?-BL = bridging ligand) are discussed and evaluated in light of recently observed correlations between their measured ?E1/20 values (the difference between sequential redox events, ?E1/20 = E20 - E10) and IR triple bond frequencies of analogous {M(L) n}?R (R = C?O, ?C?N?C?C?Ph) complexes (a indicator of electron density at {M(L) n}) (Chapter 6). Trends in individual contributions to the free energy of comproportionation are explored for complexes of the type [{M(L) n}2(?-C?C?C?C)].

Micro-electro-mechanical systems (MEMS) technology allows the fabrication of small mechanical systems in silicon using standard micro-fabrication pro- cesses. MEMS techniques have found wide acceptance in such devices as ac- celerometers, micro-mirrors, resonators, probes, and micro-tweezers to name a few. Though small linear motions are common in MEMS applications, few devices exhibit reliable rotary motion. This work explores several methods of fabricating rotary bearings using micro-balls as the support mechanism. Micro-ball bearings have several advantages over other MEMS bearing tech- nologies in that they provide robust mechanical support, require no external control systems, and basic designs require very few fabrication steps. Ball cages or retainers are common in macro-scale bearings, providing uniform spacing between the balls. Several cage designs are proposed and explored in this work: a radial ball bearing with an integrated ball cage, a dual-row style cage, and ve unique cage geometries integrated into silicon micro-turbines (SMTs.) Also, an example of a curved or angular contact race- way is presented as an example of this type of raceway geometry in MEMS devices. Each is presented with a discussion of the design considerations and fabrication process. This is followed by a characterization of the performance of each design. These studies found that the integrated cage in the radial ball bearing performs well at speeds ranging up to 20 000RPM. Minimal wear was ob- served after 6 hours of continuous testing. However, the solder bond in the cage was a common failure point in these devices, limiting the reliability and longevity. The dual groove style cage was designed to eliminate the solder bond. However, the higher frictional forces between the ball and the cage in this design resulted in higher losses during operation. Taking into account the higher losses and the added complexity of the design, it seems unlikely that this approach would be appropriate for further study. However, the design does represent a novel approach for releasing multi-wafer rotary structures and is presented here as example of this technique. Testing of the cage de- signs for the SMTs indicated that a full ring design (a full annulus with holes for the balls) performed the best of the 5 cage geometries. However, these devices do not perform as well as cage-less designs for high speed applications due to higher ctional forces and increased raceway wear at the interface be- tween the ball and the raceway edge. Finally, the curved raceway has shown excellent performance up to 2500RPM with normal loads up to 40mN in tribometer testing. SMTs with this raceway desing were also tested for over 10 million revolutions and at speeds over 70 000RPM. The test results for all of the bearings designs presented here show that the devices exhibit stable operation at low to moderately high speeds.

Large-area electronics (LAE) manufacturing has been a key focus of both academic and industrial research, especially within the last decade. The growing interest is born out of the possibility of adding attractive properties (flexibility, light weight or minimal thickness) at low cost to well-established technologies, such as photovoltaics, displays, sensors or enabling the realisation of emerging technologies such as wearable devices and the Internet of Things. As such there has been great progress in the development of materials specifically designed to be employed in solution processed (plastic) electronics, including organic, transparent metal oxide and nanoscale semiconductors, as well as progress in the deposition methods of these materials using low-cost high-throughput printing techniques, such as gravure printing, inkjet printing, and roll-to-roll vacuum deposition. Meanwhile, industry innovation driven by Moore's law has pushed conventional silicon-based electronic components to the nanoscale. The processes developed for LAE must strive to reach these dimensions. Given that the complex and expensive patterning techniques employed by the semiconductor industry so far are not compatible with LAE, there is clearly a need to develop large-area high throughput nanofabrication techniques. This thesis presents progress in adhesion lithography (a-Lith), a nanogap electrode fabrication process that can be applied over large areas on arbitrary substrates. A-Lith is a self-alignment process based on the alteration of surface energies of a starting metal electrode which allows the removal of any overlap of a secondary metal electrode. Importantly, it is an inexpensive, scalable and high throughput technique, and, especially if combined with low temperature deposition of the active material, it is fundamentally compatible with large-area fabrication of nanoscale electronic devices on flexible (plastic) substrates. Herein, I present routes towards process optimisation with a focus on gap size reduction and yield maximisation. Asymmetric gaps with sizes below 10 nm and yields of > 90 % for hundreds of electrode pairs generated on a single substrate are demonstrated. These large width electrode nanogaps represent the highest aspect ratio nanogaps (up to 108) fabricated to date. As a next step, arrays of Schottky nanodiodes are fabricated by deposition of a suitable semiconductor from solution into the nanogap structures. Of principal interest is the wide bandgap transparent semiconductor, zinc oxide (ZnO). Lateral ZnO Schottky diodes show outstanding characteristics, with on-off ratios of up to 106 and forward current values up to 10 mA for obtained upon combining a-Lith with low-temperature solution processing. These unique devices are further investigated for application in rectifier circuits, and in particular for potential use in radio frequency identification (RFID) tag technology. The ZnO diodes are found to surpass the 13.56 MHz frequency bernchmark used in commercial applications and approach the ultra-high frequency (UHF) band (hundreds of megahertz), outperforming current state of the art printed diodes. Solution processed fullerene (C60) is also shown to approach the UHF band in this co-planar device configuration, highlighting the viability of a-Lith for enabling large-area flexible radio frequency nanoelectronics. Finally, resistive switching memory device arrays based on a-Lith patterned nanogap aluminium symmetric electrodes are demonstrated for the first time. These devices are based either on empty aluminium nanogap electrodes, or with the gap filled with a solution-processed semiconductor, the latter being ZnO, the semiconducting polymer poly(9,9-dioctylfluorene-alt-benzothiadiazole) (F8BT) or carbon nanotube/polyfluorene blends. The switching mechanism, retention time and switching speed are investigated and compared with published data. The fabrication of arrays of these devices illustrates the potential of a-Lith as a simple technique for the realisation of large-area high-density memory applications.

Since the development of efficient THz sources, THz-TDS has been proved to be a promising tool to probe directly the intermolecular modes, rotational motion and intermolecular vibrations of molecules in a variety of chemicals capable of extracting useful spectroscopic information. In this thesis, an on-chip spectroscopy system based on coplanar waveguide (CPW) technology has been designed, optimised and tested, in order to probe spectral features of overlaid polycrystalline materials. As proof of principle, this system was used to recover the THz spectra of α-lactose monohydrate, observing spectral features at ~ 0.53 THz and ~ 1.37 THz. A significant frequency shift in the 1.37 THz feature was observed when the on-chip spectroscopy measurements were performed over a variable temperature range of (~ 6-293 K). Spectral features obtained from the on-chip system were also compared to those obtained from a free-space THz-TDS system to highlight the benefits of using an on-chip system over free space THz-TDS. A theoretical model developed using Ansoft HFSS tool was then used to optimise device design parameters in the second generation of CPW devices. In doing so, the bandwidth of the system was enhanced from ~ 0.42 to 1.6 THz and a much higher frequency resolution of (~ 2 GHz) was obtained compared to that of the first-generation CPW devices (~ 55 GHz) with the modified device design of second generation devices. Branching waveguide systems (THz Y-splitter and coupler) were also simulated, fabricated and measured in order to investigate THz pulse splitting in branching waveguides. These systems allow the measurement of both a sampled and reference pulse. During these measurements, the splitting of the THz pulses propagating in an on-chip THz system was also demonstrated for the first time.

The primary intention of this research work was to evaluate a topology for a sub-harmonic down conversion mixer exploiting the fourth harmonic of a LO signal. Designs were evaluated by simulation at 640GHz and 320GHz with the aim of exploring the potential of a RTD based down-converter at 640GHz, in the 580-750GHz atmospheric window, with an intermediate frequency signal in the range around 2GHz by mixing with the fourth harmonic of a 159.5GHz LO signal. Related design studies were undertaken at 320GHz which gave a simulated single side band (SSB) conversion loss of 5.7dB, and with a LO power requirement of less than -9.5dBm which vindicated the principle, as far as the design stage is concerned, of using RTDs as the non-linear mixing element, where the layer design can be tailored to favour very low pump powers. The other, related, target of the current PhD work was to also explore the potential for high LO drive level mixers and their up-conversion efficiencies using the same novel devices, i.e. RTDs, but with a different layer design, better suited to support high pump powers in this instance. For achieving the latter goal, two different sub-harmonic up-conversion mixers employing a single RTD and using the second harmonic of an LO signal were designed and evaluated at two different frequencies. The first mixer design was aimed at 180 GHz providing -7.5dBm of output power while the second one should work at 110GHz showing output power in the range of -4dBm, and was used to initially evaluate the approach and which could, in principle, be later fabricated and measured. All these down and up-conversion mixers were carefully designed using ADS and HFSS and evaluated using two different technologies, microstrip and Grounded Coplanar Waveguide (GCPW), and both compared with a nearest Schottky diode based approaches, and also their physical mask was produced in anticipation of a later fabrication stage.

No description available.