The object of this thesis is to form an understanding of the origin of the problems associated with proton-exchanged waveguides, and to investigate possible solutions. Chapter 1 gives a brief introduction to the properties of lithium niobate, and discusses the methods available for fabricating optical waveguides in the bulk material, with particular emphasis on waveguide fabrication by the proton-exchange process. Some of the devices which have been fabricated by proton-exchange are discussed. The problems associated with proton-exchanged waveguides are reviewed. Chapter 2 deals with the physical and chemical characterisation of proton-exchanged waveguides fabricated using neat benzoic acid melts. The extent of proton-exchange is determined as a function of fabrication time and temperature using optical waveguide prism-coupler measurements, infrared absorption spectroscopy, and atomic absorption spectroscopy. Chapter 3 is concerned with the problem of waveguide mode-index stability. Using a hydrogen isotopic-exchange reaction, the extent of which is obsrved via infrared absorption spectroscopy, information on the (room-temperature) mobility of protons within the guiding layer is obtained for waveguides fabricated using neat benzoic acid melts. The recently reported process of fabricating waveguides in lithium niobate by deuterium-exchange is investigated. The behaviour of proton-exchanged and deuterium-exchanged waveguides with respect to reaction with atmospheric water vapour is investigated, and the optical properties of deuterium-exchanged waveguides are studied. In Chapter 4, a study of annealed and dilute-melt proton-exchanged waveguides is presented. It is shown, using prism-coupler measurements and infrared absorption spectroscopy, that ennealed and dilute-melt waveguides can have very similar optical properties, depending on the amount of annealing and the lithium benzoate mole-fractions used. The extent of proton-exchange is determined with time (between 215oC and 235oC) for dilute-melt waveguides produced using lithium benzoate mole-fractions of up to 1.1%. Isotopic-exchange in annealed and dilute-melt waveguides is also investigated, both at room-temperature and at temperatures commonly used for annealing. A possible explanation for the poor optical properties of (neat-melt) proton-exchanged waveguides is given. Chapter 5 deals with a study of propagation losses (using the two-prism method) and the electro-optic effect in x- and z-cut proton-exchanged waveguides. Measurements of r33 (in proton-exchanged waveguides) and r22 (in titanium-indiffused waveguides) are carried out using an external interferometric method designed by the author. The results of Chapter 4 are used to establish a method by which losses below 0.5dB/cm and a substantially restored electro-optic effect can be achieved (using a combination of dilute-melt fabrication with post-exchange annealing). Prior to the waveguide measurements, the bulk electro-optic effect is investigated for congruent, incongruent, MgO-doped, and annealed (high-temperature) crystals. Finally, in Chapter 6, a summary of the thesis is presented, and suggestions for future work are given.

To avoid the problems associated with specifying the exact nature of the heat input from welding arcs, an analogue model is proposed which simulates the quasi-static thermal field produced around the isothermal contour of the molten weld pool boundary during the welding of thin plate. The design of an electrical analogue based directly on Rosenthal's equation (1) governing the quasistatic heat flow about a moving source is shown to be impractical although this approach identifies the physical significance of the two parameter ratios. To overcome the difficulties associated with the direct analogue, a simple transformation of Rosenthal's equation is employed and the design of an indirect or 0 field analogue of this transformed equation is developed. The details of the construction and commissioning of such an analogue are reported. The application of this analogue to studying the quasi-static thermal field is tested by comparing analogue predicted and experimentally measured temperature histories of points in the HAZ for a range of autogenous TIG melt runs on thin mild steel plate. The experimental results are obtained from a purpose built automatic welding rig which incorporates a facility for determining the shape of the molten weld pool during welding. The results from these comparative tests show a good agreement between predicted and measured temperature histories and the application of the 0 field analogue to studying the thermal field during welding is discussed.

A general study of the finite element approach and its application to structural analysis is conducted and methods of derivation of element properties are reviewed. It is found that the method has the advantage of generality of application and that the direct approach is easier to programme than the force method. A unified and simple formula is derived for the computation of element elastic and geometric stiffness matrices and is found to be much easier to apply than existing methods. The standard finite element displacement method is used to study overall buckling of beams, plates, and stiffened plates. The results show that the method can provide accurate answers as compared with the existing analytical approaches. The exact finite strip, the approximate finite strip, and the finite strip for local buckling analysis are reviewed and are found to economise greatly in computer time, needing less storage space because of their narrow band width as compared with the standard finite element method. In particular, the finite strip for local stability which uses a standard eigenvalue subroutine and is based on the concept of geometric stiffness matrices is used to study local buckling and buckling of plates and plates supported elastically by continuous elastic medium. The results obtained are shown to be very close to those obtained analytically, the effect of the elastic support being to increase greatly the buckling stress relative to the unsupported plates

The present dissertation develops, applies and evaluates a novel method for the representation and retrieval of patterns in musical data. The method supports the typical polyphonic patterns that one finds in music theory textbooks. Most current computational methods to musical patterns are restricted to monophony (one melody at a time). The Structured Polyphonic Patterns method (SPP) applies to the general case of polyphonic music, where many melodies may unfold concurrently. Pattern components are conjunctions of features which encode properties of musical events, or relations that they form with other events. Relations between events that overlap in time but are not simultaneous are supported, enabling patterns to express key temporal relations of polyphonic music. Patterns are formed by joining and layering pattern components into sequences (horizontal structures) and layers (vertical structures). A layer specifies voicing in an abstract way, and the exploration of different voice permutations is handled automatically. The SPP method also provides a mechanism for defining new features. We evaluate SPP by developing a small catalog of musicologically relevant queries and analyzing the results on four corpora: 185 chorale harmonizations by J.S. Bach, Mozart Symphony no. 40, a small set of piano pieces by Chopin, and a collection of folk songs containing more than 8000 pieces – in addition to its size, demonstrating the scalability of the method, that latter corpus is interesting as it shows that SPP is also usable for monophony. Examining several corpora allows us to establish that some polyphonic patterns constitute salient properties of a corpus: they are over-represented in one corpus by comparison to the others. In addition, the queries we develop demonstrate that the SPP method possesses sufficient expressiveness to capture important music-theoretic notions. At the same time, we show how the method is more restrictive than some existing polyphonic pattern representations, hence providing a better approximation of the expressive power required for polyphonic patterns. It is a better candidate representation for music data mining, a difficult problem that has received significant attention for the monophonic case, but limited attention for the more general polyphonic case.

Quantifying flow of information in a program involves calculating how much information (e.g. about secret inputs) can be leaked by observing the program's public outputs. Recently this field has attracted a lot of research interest, most of which makes use of Shannon's information theory, e.g. mutual information, conditional entropy, etc. Computability entails that any automated analysis of information is necessarily incomplete. Thus quantitative flow of analyses aim to compute upper bounds on the sizes of the flows in a program. Virtually all the current quantitative analyses treat program variables independently, which significantly limits the potential for deriving tight upper bounds. Our work is motivated by the intuition that knowledge of the dependencies between program variables should allow the derivation of more precise upper bounds on the size of flows, and that classical abstract interpretation provides an effective mechanism for determining such dependencies in the form of linear constraints. Our approach is then to view the problem as one of constrained optimization (maximum entropy), allowing us to apply the standard technique of Lagrange multiplier method. Application of this technique turns out to require development of some novel methods due to the essential use of non-linear (entropy) constraints, in conjunction with the linear dependency constraints. Using these methods we obtain more precise upper bounds on the size of information flows than is possible with existing analysis techniques.

Thapsigargin is a sesquiterpene lactone which was isolated from the plant Thapsia gargantica. It has been used in studies investigating intracellular calcium pools and calcium signalling as it causes a rise in intracellular calcium concentration. It inhibits the endoplasmic/sarcoplasmic reticulum family of calcium-ATPases and allows calcium to leak out of intracellular stores followed by a sustained influx of extracellular calcium. Porcine lymphocytes are used throughout this thesis due to their ready availability and as they have been used in this laboratory for many years as a model for human lymphocyte activation. In this study, the effects of thapsigargin on the activation of porcine and human peripheral blood mononuclear cells in vitro was studied and compared with those of the calcium ionophore, ionomycin and an alternative endoplasmic/sarcoplasmic reticulum calcium-ATPase inhibitor, 2,5-di-(tert-butyl)-1,4-benzohydroquinone. The drugs were also used in combination with the phorbol ester, phorbol myristate acetate, used to activate protein kinase C, and the immunosuppressive drug, FK506, which inhibits calcium dependent T cell activation. Thapsigargin was found to raise intracellular calcium concentration in both porcine and human cells in a concentration-dependent manner but was more effective in human than in porcine cells. In porcine cells, thapsigargin raised intracellular calcium concentration to a similar level as the mitogenic lectin conconavalin A but not to as high a level as ionomycin or 2,5-Di-(tent-butyl)-1,4-benzohydroquinone. In human cells, thapsigargin was as effective as both ionotnycin and 2,5-Di-(tert butyl)-1,4-benzohydroquinone. Neither phorbol myristate acetate nor FK506 had any effect on the rise in cytoplasmic calcium concentration induced by thapsigargin in porcine cells. As expected, all three of the drugs used to raise intracellular calcium concentration in combination with phorbol myristate acetate were found to be mitogenic for both human and porcine cells. They induced DNA synthesis at 48 hours post stimulation in a concentration dependent manner. Thapsigargin was found to induce DNA synthesis at a lower cytoplasmic calcium concentration than either of the alternative calcium mobilisers in porcine cells. FK506 inhibited the DNA synthesis induced by ionomycin/ phorbol myristate acetate and 2,5-Di-(tert-butyl)-1,4-benzohydroquinone/phorbol myristate acetate in both human and porcine cells and DNA synthesis induced by thapsigargin/phorbol myristate acetate in human cells. However, FK506 enhanced DNA synthesis in porcine cells that had been exposed to an antagonistic combination of thapsigargin and phorbol myristate acetate. This result was surprising as FK506 is expected to inhibit calciumdependent activation via the inhibition of the calcium-dependent protein phosphatase, calcineurin. On further investigation it was found that the porcine cells incubated with thapsigargin, phorbol myristate acetate and FK506 were able to proliferate in the absence of measurable levels of the cytokine interleukin 2. Furthermore, calcineurin activation was inhibited in these cells. It was also found that combinations of thapsigargin, phorbol myristate acetate and/or FK506 did not affect activation of some of the protein kinases believed to be involved in T cell activation in any different way from the alternative mitogenic combinations used in this study. The effects of thapsigargin and other mitogens on apoptosis in porcine and human peripheral blood mononuclear cells were also investigated. lt was found that a very high proportion of resting porcine cells in culture were apoptotic and that there were fewer apoptotic cells in proliferating cultures. There were fewer apoptotic cells in resting human cell cultures and all combinations of mitogenic or immunosuppressive drugs increased the proportion of apoptotic human cells. In addition, DNA synthesis was occurring in only a small proportion of the population of stimulated human cells regardless of the mitogen(s) used. The data presented show that the effects of thapsigargin on cytoplasmic calcium concentration and/or the effects of phorbol myristate acetate on protein kinase C activation cannot entirely account for the results and that the activation of porcine peripheral blood mononuclear cells does not represent an entirely accurate model of human lymphocyte activation.

No description available.

No description available.

No description available.

No description available.