In the course of the work on the interactions of electric and magnetic fields with biological materials both living and dead, 'it was noticed that published dielectrophoretic yield curves for biological cells showed unexplained deviations in the region of 2kHz which is the proton magnetic resonance frequency in a typical laboratory ambient magnetic field. The exact value is 2.13 kHz for a field of 50 UT. The results of preliminary dielectrophoretic and dielectric measurements show features at frequencies which correspond to the NMR condition for that value of steady magnetic field in which the measurements were made. Very sharp dielectric loss peaks were found corresponding to 1H, 31 p, 23Na, 35C1 amd 39K resonances. The electron spin resonance also showed up in the dielectric loss. The onset plus NMR conditions of these resonances commences at the value of the steady magnetic field strength such that one quantum of magnetic flux (2.07 x 10-15 Wb) would link the cross-sectional area of a single biological cell or pair of cells. Approximately 1.0 gauss or 0.5 gauss (100iT or 50pT) respectively in the case of 5um diameter yeast cell. Growing cultures of the yeast cells in fields which satisfy the proton NMR conditions as a function of temperature results in a slight reduction in the mean generation time (MGT). Comparison of this with the dielectric constant and loss increments at the corresponding temperatures showed that when the MGT is least, the cells are dividing most rapidly and the dielectric increments are greatest. Steps in the voltage-current characteristic of a pearlchain of yeast cells were found to occur for a few minutes around the time of cytokinesis as they were observed under phase contrast microscope. The cells prepared for synchronous division were collected by dielectrophoresis into a one micron gap between two point electrodes mounted on a microscope slide. Steps were observed about 3 to 4 hours later at ambient temperature. The mean generation time is 4 hours. An emission of a radio frequency signal from yeast cells in the region of 7 MHz and in the range of 50 MHz - 80 MHz were found to occur about a mean generation time after starting the incubation of the cells for synchronous growth. All these effects were only observed on the live yeast cells and never observed in experiments using killed yeast cells under the conditions otherwise same in all aspects.

The development of low cost, efficient photovoltaic devices is a major technological challenge which demands suitable materials and fabrication processes. Thin film polycrystalline heterojunction solar cells appear to be most appropriate with respect to cost and ease of manufacture, and it is anticipated that the next generation of photovoltaic devices will be based entirely on thin film technologies. Copper based ternary and multinary compounds are well established as exceptional semiconductors with potential applications in the fields of solar cells for both terrestrial and space applications, infra-red detectors, light emitting diodes etc. The chalcopyrite forms of these compounds have large absorption coefficients and exhibit superior radiation resistance. Among these compounds, CuInSe2 (CIS) and CuIn1-xGaSe2 (CIGS) have raised the most interest and recent thin film heterojunction photovoltaic devices based on these materials have achieved efficiencies of the order 15.5% and 16.9% respectively. The higher efficiencies realised in CIGS based devices is due to the fact that the band gap of the material can be adjusted to wards the optimum value (1.45eV) by the partial substitution of gallium for indium. In this work, thin films of both CIS and CIGS were deposited onto glass substrates by flash evaporation of the respective pre-reacted source materials. The substrate temperature was varied between room temperature and 200 degrees C. Two types of evaporation sources, a flat tungsten strip and a molybdenum twin chimney were used. The effect of the growth conditions on the film properties was observed. The structural, compositional and electro-optical properties were studied using a variety of analytical techniques including x-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive analysis with x-ray (EDAX), x-ray fluorescence (XRF), Rutherford backscattering spectroscopy (RBS), four point and thennal probe techniques, photoconductivity (PC) and photoacoustic spectroscopy (PAS). The as-grown films were found to have a columnar structure and a strong preferred orientation with the <112> plane parallel to the substrate. Results from EDAX, XRF and RBS indicated that the as-grown films were slightly deficient in selenium, otherwise the composition was comparable with that of the starting polycrystalline material. Electrical measurements revealed both n- and p-type conductivities with resistivity values in the range 10-2 to 106 gcm. The as-grown films were subsequently processed under several sets of conditions including vacuum, selenium, inert and forming gas ambients at different temperature and times. A two stage post-deposition heat treatment of the films was developed to improve the composition and crystal structure and to optimise the electro-optical properties. It was observed that the first annealing stage (in a selenium ambient) produced an excellent improvement in the composition of the film. An increase in the film grain size (to > 2pm) was observed when the films were subsequently annealed in a forming gas ambient. Significant improvements were also observed in the optical properties. The as-grown and annealed films were analysed using the PAS technique which revealed the existence of several donor and acceptor states originating from intrinsic defect levels. The results were compared with those obtained from single crystals. Photoconductivity measurements were also performed on the as-grown thin films.

Techn. University, Diss., 2005--München.

In this dissertation some advanced methods for extracting sources from single and multichannel data are developed and utilized in biomedical applications. It is assumed that the sources of interest have periodic structure and therefore, the periodicity is exploited in various forms. The proposed methods can even be used for the cases where the signals have hidden periodicities, i.e., the periodic behaviour is not detectable from their time representation or even Fourier transform of the signal. For the case of single channel recordings a method based on singular spectrum anal ysis (SSA) of the signal is proposed. The proposed method is utilized in localizing heart sounds in respiratory signals, which is an essential pre-processing step in most of the heart sound cancellation methods. Artificially mixed and real respiratory signals are used for evaluating the method. It is shown that the performance of the proposed method is superior to those of the other methods in terms of false detection. More over, the execution time is significantly lower than that of the method ranked second in performance. For multichannel data, the problem is tackled using two approaches. First, it is assumed that the sources are periodic and the statistical characteristics of periodic sources are exploited in developing a method to effectively choose the appropriate delays in which the diagonalization takes place. In the second approach it is assumed that the sources of interest are cyclostationary. Necessary and sufficient conditions for extractability of the sources are mathematically proved and the extraction algorithms are proposed. Ballistocardiogram (BCG) artifact is considered as the sum of a number of independent cyclostationary components having the same cycle frequency. The proposed method, called cyclostationary source extraction (CSE), is able to extract these components without much destructive effect on the background electroencephalogram (EEG)

Magnetostriction is a source of vibration and acoustic noise of electrical machines and it can be highly anisotropic even in non-oriented electrical steel. Understanding of magnetostriction under magnetisation and stress conditions present in stator core laminations can help predict the core vibration and radiated noise. Anisotropy of magnetostriction of a 0.50 mm thick non-oriented steel investigated in Epstein strips cut at angles to the rolling direction was much higher than the anisotropy of its magnetic and elastic properties because magnetostriction arises directly from magnetic domain processes. Magnetostriction of a disc sample of the 0.50 mm thick steel was measured under ID and 2D magnetisation and compared with that of a 0.35 mm thick steel with different anisotropy level. A 2D magnetostriction model and an analytical simple domain model were used to explain the experimental results. 2D magnetostriction is dependent on the magnetostrictive anisotropy and the ratios of the transverse to longitudinal magnetostriction. AC magnetostriction measured in the disc samples was larger than in the Epstein strips due to the form effect. An induction motor model core was constructed from the 0.50 mm thick steel for measurements of localised flux density and deformation. Core deformation due to Maxwell forces was calculated. Magnetostriction and specific power loss of the core material under magnetisation conditions present in the core was measured. The localised loss in the stator teeth, tooth roots and back iron differed from their average value by 52%, 19% and 36% due to the magnetic anisotropy. Magnetostriction was estimated to be about 55% and 80% of the radial deformation at the tooth root and back iron regions respectively. Stator teeth deformed asymmetrically and the magnitude of the space harmonics increased due to the magnetostrictive anisotropy. The measurement results inferred that 2D magnetostriction can be predicted from the magnetostrictive anisotropy and vice versa. Also, core deformation and vibration of large machines, where segmented stator core laminations are used, can be estimated analytically with the knowledge of 2D magnetostriction of the core material.

The development and performance of a transistor scanned solid state ultrasonic image converter are described and the results of investigations into various applicable signal processing techniques to produce visible pictures are presented, The general requirements of an ultrasonic imaging system are considered and this is followed by a critical survey of the more sensitive ultrasonic image conversion methods. The operation and basic performance of the transistor scanned converter are discussed and its resolution and sensitivity investigated , resulting in an apparent practical limit to the resolution in. the image plane of ⅓ mm, and an estimated limiting sensitivity of approximately 10-13 W cm-2. Possible methods for the fabrication of a multielement converter are suggested and discussed. Three signal processing techniques are investigated and compared; ;that using an acoustic lens in a conventional camera arrangement, an on-line computer which effectively performs electrically the function of a lens, and a scanned array Doppler system on which only a feasibility study was carried out. It is argued that if flexibility is important then the computer "lens" is the most fitting method, and results for both near field focusing and far field beam deflection are presented) although at the moment.in 1-dimension only , A relatively simple extension to the program will make 2-dimensional image reconstruction possible.. The Doppler system described provides a simple electrical method of determining the range and bearing of far field targets and results adequately confirm the developed theory.

Microwave measurement of the properties of thin film dielectrics and lumped elements present difficulties because of the small volume of material involved. In this investigation both capacitively loaded coaxial cavities and capacitively loaded microstrip resonators were used. The dielectric properties of thin films were measured using symmetrical capacitively loaded coaxial cavities. The theoretical approaches are given for the design of the cavity and for the measurement techniques using (a) the VSWR method and (b) the perturbation method. The practical construction problems were overcome and highly stable. (mechanically as well as electrically) cavities were obtained. The measured electrical parameters of the cavities were very close to the theoretical values. Thin dielectric films were prepared using an electron beam evaporation source and an rfg sputtering unit. The properties of these films were measured using the capacitively loaded coaxial cavity. The thin film properties were correlated with the thin film deposition conditions. The dielectric properties of substrates used in the construction of mcrostrip transmission lines were measured on a sampling basis using a waveguide cavity technique. It was shown to be a most useful method • Capacitively loaded microstrip resonators were used to measure lumped elements at microwave frequencies. But' first the microstrip properties had to be measured. Very accurate techniques were introduced for measuring the effective dielectric constant and the capacitance of I a gap.in the strip of a microstrip, over a range of frequencies. Alumina, sapphire and quartz substrates were used in these measurements. The construction of .these resonators using thin film deposition and photolithographic techniques was·presented. It is considered that ·the use of these measurement techniques will contribute significantly to further advances in the field of m.i.c's.

This thesis explores suitable materials for energy stores for stationary applications, specifically a prototype hydrogen store, domestic thermal store operating between 25-100 C and a moderate thermal store for a concentrated solar power (CSP) plant operating at 400 C. The approach incorporated a unique coating technique to deliver prototype hydrogen and thermal storage media, where the coating could offer commercial advantages, for example, in the form of hydride activation and enhanced kinetics during successive cycling. The highly reversible Mg-MgH2 system is particularly promising for thermal storage, obtaining an enthalpy of reaction of 74.5 kJ/mol H2 that translates to a thermal energy capacity of approximately 2800 kJ/kg of MgH2. Nevertheless, magnesium is hindered by slow activation and poor kinetics of (de)hydrogenation, even when approaching temperatures ideal for concentrated solar power applications (in the region 400 C). Elevated temperature cycling studies were performed on commercial atomised Mg powder with magnetron sputtered catalysts (chromium, iron, vanadium and stainless steel) applied to their surfaces; the aim of which was to fabricate hydrogen storage materials that possess (de)hydrogenation characteristics equal to or even bettering their nanocrystalline equivalents, yet in a potentially economic and scalable manner. Following 50 cycles at 400 C, the coatings were found to have little to no positive impact on the behaviour of the atomised Mg powders. In addition, for both uncoated and coated samples the effects of an activation process at 400 C are matched by cycling the material 5 times from the outset, after which identical behaviour is observed during subsequent cycles. At 350 C, the benefits of catalyst coatings on the hydrogen storage properties of atomised Mg powders are evident during activation and successive cycling up to 90 times. The material undergoes different microstructural evolution during cycling when in the presence of a surface catalyst, causing an enhancement of the `nucleation and growth' stage of (de)hydrogenation. This was attributed to particle reorientation dominating particle sintering, whereas the opposite occurs for the uncoated material. For the domestic thermal and prototype hydrogen stores a selection of AB and AB2 intermetallic hydrides enhanced through catalysis or thermodynamic modification were investigated. TiFe produced via powder atomisation obtained thermodynamic properties (dehydrogenation H = 28.9 kJ/mol H2 and S = 105 J/K.mol H2) in line with published results. The minor substitution of Ni into TiFe1-xNix resulted in different hydrogenation characteristics to TiFe, for example, TiFe0:96Ni0:04 possessed a dehydrogenation of H = 29.9 kJ/mol H2 and S = 107 J/K.mol H2. Discrepancies between maximum achieved and theoretical capacities were observed for both atomised TiFe and TiFe0:96Ni0:04 and a range of possible contributing factors are discussed. A minor addition of Pd (1.17 wt.%) magnetron sputtered to the surface of TiFe0:96Ni0:04 enabled successful room temperature hydrogenation with no activation treatment required. Characterisation (SEM and TEM) confirmed it is not necessary to have complete Pd coverage in the form of a uniform coating and XPS was utilised to derive a theory for the activation mechanism. The AB2 alloy comparison between the commercially available Hydralloy C5 and in house fabricated Ti0:9Zr0:2Mn1:5V0:2Cr0:3 showed that Hydralloy C5 was the most promising alloy for the hydrogen store application with the higher working capacity (ca. 0.96 wt.%) in the pressure range of 4-15 bar at 22 C, despite Ti0:9Zr0:2Mn1:5V0:2Cr0:3 obtaining a higher maximum storage capacity (1.82 wt.%). The hydrogenation kinetics of both alloys were studied with corresponding activation energies and hydrogen diffusion coefficients determined. The kinetics of hydrogenation for both alloys is sufficiently fast that only the heat transfer of the storage system is the rate limiting parameter for hydrogen exchange for most technical applications.

Recent progresses of the aviation industry toward the More Electric Aircraft have increased the demand for high performance Electro-Mechanical Actuators. In this context, extensive research is being conducted for the design of high torque density electrical machines able to meet the high reliability standards required. For low-rate duty applications (e.g. landing gear extension, retraction and steering; flight control surfaces), this can be achieved also by exploiting the characteristic that the motor does not reach a thermal equilibrium. In this work, the principal aim is to investigate the main limits in the design of low-rate duty, high torque density electrical machines and to propose methods for the improvement of the performance and reliability of such machines. This is achieved through detailed analysis of the motor design issues, of its thermal performance, as well as of the most critical ageing phenomena during operations. A structured design procedure for surface mounted permanent magnet machines is presented. A novel thermal model which features high accuracy and low computational cost is presented. A novel winding insulation lifetime consumption model which relates the winding time to failure to its temperature profile during operations is proposed. As vessel to address the thesis’ objectives, a fault tolerant electrical machine, which is an integral part of an actuator for the extension and retraction of a helicopter landing gear, is designed, built and experimentally tested. The tests are aimed at validating the design procedure and the thermal model accuracy. A series of accelerated ageing tests is conducted on samples of the motor windings, which serve to analyse the insulation degradation processes under different stress levels and to validate the proposed lifetime consumption model.

The main objective of this thesis was the development of polarisation selective gratings in silicon-on-insulator (SOI) technology. These devices can find numerous applications in the design of highly performing optical filters and, more in general, in all those devices that require on-chip manipulation of the polarisation state. The development of these devices was preceded by the optimisation of several fabrication processes, such as lithography and dry etching, and the re-design of a number of key components such as inverse polymer tapers and metallic heaters for thermal tuning. This activity culminated into a very robust process flow for SOI devices, with repeatable propagation losses in the order of 1 dB/cm, heaters with a very high tuning efficiency of 12 mW per π phase shift and 2dB and 1dB waveguide-to-optical fibre coupling losses for the TE and TM polarised mode, respectively. The grating designs developed in this thesis consisted of periodic holes etched onto the top surface of the silicon optical waveguide. Such geometry overlaps strongly with the TM polarised mode only and does not introduce additional losses to the TE mode. The benefit and the additional functionalities provided by the top grating geometry was assessed on two different polarisation sensitive devices. The first consisted in a microring resonator with integrated gratings for the emission of optical vortex beams, for which the top gratings provided a route to engineer the polarisation state of the emitted beam. The second device was a Bragg grating filter, where the top grating allowed the demonstration of extinction ratio values as high as 60dB by filtering the residual TM mode generated by the strong polarisation scattering.