For direct cycle gas cooled high temperature reactor designs, operating conditions may be limited as a result of excessive maintenance dose rates caused by the 110mAg source term on the turbine. The accurate prediction of silver fission and activation products’ release during normal operation is required to ensure regulatory compliance and economic viability of planned power plants. Fuel qualification programs should provide satisfactory results to ensure correct analyses, but will however not be available for many years. In the meantime data from the German fuel development program may be utilized. Traditionally diffusion models were used to derive transport parameters from limited irradiation testing of fuel materials and components. Best estimates for all applicable German fuel irradiation tests with defendable uncertainty ranges were never derived. However, diffusion theory and current parameters cannot account for all irradiation and heat-up test results, and for some tests, it appears unacceptably conservative. Other transport mechanisms have been suggested and alternative calculation models are being considered. In this thesis the relevant German material and irradiation tests were evaluated with the current PBMR metallic fission product release calculation model. Transport through all the fuel materials and components and from the sphere to the coolant gas was considered and best possible models and parameters were suggested. For the transport of silver through the SiC layer an alternative suggested model called the Molecular Vapour Transport Release (MVR) Model was evaluated against the traditional diffusion model. From this evaluation it was shown that classical diffusion modelling was still a viable model to predict silver transport in SiC. The MVR model was found to be a feasible model as well. However, due to the much larger verification and validation effort required, it was decided to use the diffusion model until such time that experimental results become available that might elucidate the exact physical transport model. The evaluation also showed that the diffusion model used must be quantified in a detailed evaluation of all applicable irradiation tests. A study of all German irradiation tests was previously performed and the applicable irradiation tests were identified. A detailed evaluation of these irradiation tests were performed with an updated diffusion model. New transport and material parameters were derived in this detailed evaluation and compared with existing values. An evaluation of some heat-up tests of irradiated fuel spheres was performed to assess the range for which the newly derived transport parameters are valid. The different models with their old and newly derived parameters were used to analyse sample PBMR cores. Recommendations were made to the suitability of the different models and parameters for future PBMR silver fission and activation product analyses. / Thesis (DPhil)--University of Pretoria, 2009. / Physics / unrestricted

Die Telomere sind spezialisierte DNA-Protein-Komplexe, die sich an den Enden der Chromosomen der eukaryotischen Zellen befinden. Die Telomerase ist ein Ribonukleoprotein, welches für die vollständige Replikation der Telomere bei den meisten Eukaryoten verantwortlich ist. Die katalytische Untereinheit des Enzyms (hTERT beim Menschen) besitzt Reverse-Transkriptase-Aktivität, und nutzt eine integrierte RNA (hTR beim Menschen) als Template, um Telomer-Wiederholungssequenzen an den Enden der Chromosomen zu synthetisieren. Die Telomerase ist in den meisten normalen humanen somatischen Zellen unterdrückt. In den meisten Krebszellen jedoch, stellt die Reaktivierung der Telomerase zur Beibehaltung der Telomerlänge eine Voraussetzung für deren unbegrenztes Wachstumspotential dar. Im Rahmen dieser Arbeit sollte ein neuer, einfacher und selektiver Assay für den Nachweis der humanen Telomerase entwickelt werden. In dem neuen Assay sollten die beiden Kernkomponenten des Enzyms, die Protein-Untereinheit und die RNA, die Targets sein. Der Test ist in seiner Grundstruktur wie folgt aufgebaut : 1. Immobilisierung der Telomerase über die hTERT an eine Festphase, beschichtet mit Phosphorothioat-modifizierten (PS) Oligonukleotiden oder Heparin. Zusammen mit der Telomerase werden bei diesem Schritt die Heparin-bindenden Proteine, die in der Probe enthalten sind, an die Festphase gebunden. 2. Spezifischer Nachweis der hTR. Zur Detektion der hTR wird ein Oligonukleotid-Ligations-Assay (OLA) oder eine Reverse-Transkriptase-PCR (RT-PCR) eingesetzt. In der optimierten Endversion wurde zur Immobilisierung des Enzyms eine Festphase, beschichtet mit PS-Oligonukleotiden, verwendet. Die hTR wurde mittels RT-PCR nachgewiesen. Mit dem neuen Assay wurden erfolgreich 75 Tumorzellen detektiert. / Telomeres are specialized DNA-Protein structures located at the ends of linear eukaryotic chromosomes. Telomerase is a ribonucleoprotein, which is responsible for the complete replication of the telomeres in most eukaryotes. The catalytic reverse transcriptase protein subunit (hTERT in humans) of the nucleoprotein uses an integral RNA (hTR in humans) as a template for the addition of telomeric repeat sequences to the ends of chromosomes. Telomerase is repressed in most normal human somatic cells, while the reactivation of telomerase to maintain telomere length is necessary for the unlimited growth potential of most human cancer cells. The aim of this work was the development of a new, simple and selective assay for the detection of human telomerase. The targets of the new assay were the two core subunits of the enzyme : hTERT and hTR. The test comprises two principal steps : 1. Immobilization of the telomerase via the hTERT subunit on a solid phase, coated with heparin or phosphorothioate-modified (PS) oligonucleotides. In this step telomerase is bound together with the heparin-binding proteins of the analysed sample to the surface. 2. Specific detection of the hTR. For the detection of the hTR an oligonucleotide ligation assay (OLA) or a reverse transcriptase PCR (RT-PCR) was used. In the optimized final version of the assay a PS-coated solid phase was used for the immobilization of the enzyme. Reverse transcriptase PCR was applied for detection of the hTR. 75 tumor cells were successfully detected with the new assay.

G-quadruplexes, characterized by stacked G-tetrad rings held together by Hoogsteen hydrogen bonds, have been visualized in human cells and implicated in transcriptional and translational control, telomere maintenance and disease. RHA Helicase associated with AU-rich element (RHAU), a DEAH-box helicase, is a major G-quadruplex resolvase in human cell lysates. It binds G-quadruplexes through the RHAU specific motif in its N-terminus. In order to investigate the recognition of G-quadruplexes by helicases, the binding between the N-terminal construct of RHAU, RHAU53-105, and the DNA analog of the quadruplex formed by the 5’ terminus of human telomerase RNA component, hTR1-20, was investigated in a comprehensive biophysical approach followed by crystallization screening. RHAU53-105, hTR1-20 DNA and their complexes were analysed by gel electrophoresis, UV-visible spectroscopy, spectropolarimetry, dynamic light scattering and small angle X-ray scattering (SAXS). The findings reveal that hTR1-20 DNA, separated in two conformations by size exclusion chromatography in the presence of potassium cations, assumes a disk-like parallel G-quadruplex secondary structure in solution. Far-UV circular dichroism spectra and SAXS demonstrate that RHAU53-105 assumes an extended (Dmax = 7.8 nm , rG = 2.1 (±0.2) nm) and ordered conformation in solution. The analysis confirms the binding between RHAU53-105 and each conformation of the hTR1-20 DNA quadruplex. Circular dichroism spectra indicate the retention of quadruplex secondary structure in both RHAU53-105•hTR1-20 DNAc1 and RHAU53-105•hTR1-20 DNAc2 complexes. This analysis provides some insight into the interaction between G-quadruplexes and the N-terminal domain of RHAU and identifies 0.2 M sodium formate, 20 % (w/v) polyethylene glycol 3350 and 1.5 M sodium chloride, 10 % (v/v) ethanol as preliminary conditions for crystallization of the complex of RHAU53-105 and hTR1-20 DNAc2. / October 2014

An introduction to titanium minerals is provided along with a brief history of high tension roll (HTR) separation and its applications. Current HTR theory is critically reviewed and deficiencies in previous modelling work are identified. These problems are recognised as the likely reason why past models required empirical manipulation to reproduce observed separation behaviour. An updated theory of charge decay in a particle bed is developed for HTR applications. Particle bed interactions and their effects on electrostatic forces and discharge rates are examined, leading to a reviewed analysis of force balance. The theoretical contributions are incorporated into a new HTR model which takes a mechanistic approach to describing machine operation, as opposed to previous empirical-based work in the area. A simplified electrostatic field model is also developed. Atmospheric effects on separation behaviour are ignored in the modelling. Experimental separations are used to validate new model performance, where comparisons are specifically designed to test the theoretical developments on charge decay and force balance in a particle bed. The new model is also compared to the Dance (1992) HTR model, embedded in JKSimSand, the most recent HTR model with an empirical basis. An error analysis shows the new model performs as well as, or better, than JKSimSand in most respects. The fact that this is achieved with a mechanistic approach validates the new developments in particle bed charge decay and force theory, and sets the foundation for mechanistic HTR simulation.

In last years Hidden Markov Models (HMMs) have received significant attention in the task off-line handwritten text recognition (HTR). As in automatic speech recognition (ASR), HMMs are used to model the probability of an observation sequence, given its corresponding text transcription. However, in contrast to what happens in ASR, in HTR there is no standard set of local features being used by most of the proposed systems. In this thesis we propose the use of raw binary pixels as features, in conjunction with models that deal more directly with the binary data. In particular, we propose the use of Bernoulli HMMs (BHMMs), that is, conventional HMMs in which Gaussian (mixture) distributions have been replaced by Bernoulli (mixture) probability functions. The objective is twofold: on the one hand, this allows us to better modeling the binary nature of text images (foreground/background) using BHMMs. On the other hand, this guarantees that no discriminative information is filtered out during feature extraction (most HTR available datasets can be easily binarized without a relevant loss of information). In this thesis, all the HMM theory required to develop a HMM based HTR toolkit is reviewed and adapted to the case of BHMMs. Specifically, we begin by defining a simple classifier based on BHMMs with Bernoulli probability functions at the states, and we end with an embedded Bernoulli mixture HMM recognizer for continuous HTR. Regarding the binary features, we propose a simple binary feature extraction process without significant loss of information. All input images are scaled and binarized, in order to easily reinterpret them as sequences of binary feature vectors. Two extensions are proposed to this basic feature extraction method: the use of a sliding window in order to better capture the context, and a repositioning method in order to better deal with vertical distortions. Competitive results were obtained when BHMMs and proposed methods were applied to well-known HTR databases. In particular, we ranked first at the Arabic Handwriting Recognition Competition organized during the 12th International Conference on Frontiers in Handwriting Recognition (ICFHR 2010), and at the Arabic Recognition Competition: Multi-font Multi-size Digitally Represented Text organized during the 11th International Conference on Document Analysis and Recognition (ICDAR 2011). In the last part of this thesis we propose a method for training BHMM classifiers using In last years Hidden Markov Models (HMMs) have received significant attention in the task off-line handwritten text recognition (HTR). As in automatic speech recognition (ASR), HMMs are used to model the probability of an observation sequence, given its corresponding text transcription. However, in contrast to what happens in ASR, in HTR there is no standard set of local features being used by most of the proposed systems. In this thesis we propose the use of raw binary pixels as features, in conjunction with models that deal more directly with the binary data. In particular, we propose the use of Bernoulli HMMs (BHMMs), that is, conventional HMMs in which Gaussian (mixture) distributions have been replaced by Bernoulli (mixture) probability functions. The objective is twofold: on the one hand, this allows us to better modeling the binary nature of text images (foreground/background) using BHMMs. On the other hand, this guarantees that no discriminative information is filtered out during feature extraction (most HTR available datasets can be easily binarized without a relevant loss of information). In this thesis, all the HMM theory required to develop a HMM based HTR toolkit is reviewed and adapted to the case of BHMMs. Specifically, we begin by defining a simple classifier based on BHMMs with Bernoulli probability functions at the states, and we end with an embedded Bernoulli mixture HMM recognizer for continuous HTR. Regarding the binary features, we propose a simple binary feature extraction process without significant loss of information. All input images are scaled and binarized, in order to easily reinterpret them as sequences of binary feature vectors. Two extensions are proposed to this basic feature extraction method: the use of a sliding window in order to better capture the context, and a repositioning method in order to better deal with vertical distortions. Competitive results were obtained when BHMMs and proposed methods were applied to well-known HTR databases. In particular, we ranked first at the Arabic Handwriting Recognition Competition organized during the 12th International Conference on Frontiers in Handwriting Recognition (ICFHR 2010), and at the Arabic Recognition Competition: Multi-font Multi-size Digitally Represented Text organized during the 11th International Conference on Document Analysis and Recognition (ICDAR 2011). In the last part of this thesis we propose a method for training BHMM classifiers using In last years Hidden Markov Models (HMMs) have received significant attention in the task off-line handwritten text recognition (HTR). As in automatic speech recognition (ASR), HMMs are used to model the probability of an observation sequence, given its corresponding text transcription. However, in contrast to what happens in ASR, in HTR there is no standard set of local features being used by most of the proposed systems. In this thesis we propose the use of raw binary pixels as features, in conjunction with models that deal more directly with the binary data. In particular, we propose the use of Bernoulli HMMs (BHMMs), that is, conventional HMMs in which Gaussian (mixture) distributions have been replaced by Bernoulli (mixture) probability functions. The objective is twofold: on the one hand, this allows us to better modeling the binary nature of text images (foreground/background) using BHMMs. On the other hand, this guarantees that no discriminative information is filtered out during feature extraction (most HTR available datasets can be easily binarized without a relevant loss of information). In this thesis, all the HMM theory required to develop a HMM based HTR toolkit is reviewed and adapted to the case of BHMMs. Specifically, we begin by defining a simple classifier based on BHMMs with Bernoulli probability functions at the states, and we end with an embedded Bernoulli mixture HMM recognizer for continuous HTR. Regarding the binary features, we propose a simple binary feature extraction process without significant loss of information. All input images are scaled and binarized, in order to easily reinterpret them as sequences of binary feature vectors. Two extensions are proposed to this basic feature extraction method: the use of a sliding window in order to better capture the context, and a repositioning method in order to better deal with vertical distortions. Competitive results were obtained when BHMMs and proposed methods were applied to well-known HTR databases. In particular, we ranked first at the Arabic Handwriting Recognition Competition organized during the 12th International Conference on Frontiers in Handwriting Recognition (ICFHR 2010), and at the Arabic Recognition Competition: Multi-font Multi-size Digitally Represented Text organized during the 11th International Conference on Document Analysis and Recognition (ICDAR 2011). In the last part of this thesis we propose a method for training BHMM classifiers using discriminative training criteria, instead of the conventionalMaximum Likelihood Estimation (MLE). Specifically, we propose a log-linear classifier for binary data based on the BHMM classifier. Parameter estimation of this model can be carried out using discriminative training criteria for log-linear models. In particular, we show the formulae for several MMI based criteria. Finally, we prove the equivalence between both classifiers, hence, discriminative training of a BHMM classifier can be carried out by obtaining its equivalent log-linear classifier. Reported results show that discriminative BHMMs clearly outperform conventional generative BHMMs. / Giménez Pastor, A. (2014). Bernoulli HMMs for Handwritten Text Recognition [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/37978

HOLCIM has various cement production plants across India. These plants struggle to produce the projected amount of cement due to electricity shortages. Although coal is abundant in India, the production thereof is in short supply. It is proposed that a thorium HTR (100 MWt) combined with a PCU (Rankine cycle) be constructed to supply a cement production plant with the required energy. The Portland cement production process is investigated and it is found that process heat integration is not feasible. The problem is that for the feasibility of this IPP to be assessed, a Rankine cycle needs to be adapted and optimised to suit the limitations and requirements of a 100 MWt thorium HTR. Advantages of the small thorium HTR (100 MWt) include: on-site construction; a naturally safe design and low energy production costs. The reactor delivers high temperature helium (750°C) at a mass flow of 38.55 kg/s. Helium re-en ters the reactor core at 250°C. Since the location of the cement production plant is unknown, both wet and dry cooling tower options are investigated. An overall average ambient temperature of India is used as input for the cooling tower calculations. EES software is used to construct a simulation model with the capability of optimising the Rankine cycle for maximum efficiency while accommodating various out of the norm input parameters. Various limitations are enforced by the simulation model. Various cycle configurations are optimised (EES) and weighed against each other. The accuracy of the EES simulation model is verified using FlowNex while the optimised cycle results are verified using Excel’s X-Steam macro. It is recommended that a wet cooling tower is implemented if possible. The 85% effective heat exchanger delivers the techno-economically optimum Rankine cycle configuration. For this combination of cooling tower and heat exchanger, it is recommended that the cycle configuration consists of one de-aerator and two closed feed heaters (one specified). After the Rankine cycle (PCU) has been designed and optimised, it is evident that the small thorium HTR (100 MWt) can supply the HOLCIM plant with the required energy. The optimum cycle configuration, as recommended, operates with a cycle efficiency of 42.4% while producing 39.867 MWe. A minimum of 10 MWe can be sold to the Indian distribution network at all times, thus generating revenue. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2014

HOLCIM has various cement production plants across India. These plants struggle to produce the projected amount of cement due to electricity shortages. Although coal is abundant in India, the production thereof is in short supply. It is proposed that a thorium HTR (100 MWt) combined with a PCU (Rankine cycle) be constructed to supply a cement production plant with the required energy. The Portland cement production process is investigated and it is found that process heat integration is not feasible. The problem is that for the feasibility of this IPP to be assessed, a Rankine cycle needs to be adapted and optimised to suit the limitations and requirements of a 100 MWt thorium HTR. Advantages of the small thorium HTR (100 MWt) include: on-site construction; a naturally safe design and low energy production costs. The reactor delivers high temperature helium (750°C) at a mass flow of 38.55 kg/s. Helium re-en ters the reactor core at 250°C. Since the location of the cement production plant is unknown, both wet and dry cooling tower options are investigated. An overall average ambient temperature of India is used as input for the cooling tower calculations. EES software is used to construct a simulation model with the capability of optimising the Rankine cycle for maximum efficiency while accommodating various out of the norm input parameters. Various limitations are enforced by the simulation model. Various cycle configurations are optimised (EES) and weighed against each other. The accuracy of the EES simulation model is verified using FlowNex while the optimised cycle results are verified using Excel’s X-Steam macro. It is recommended that a wet cooling tower is implemented if possible. The 85% effective heat exchanger delivers the techno-economically optimum Rankine cycle configuration. For this combination of cooling tower and heat exchanger, it is recommended that the cycle configuration consists of one de-aerator and two closed feed heaters (one specified). After the Rankine cycle (PCU) has been designed and optimised, it is evident that the small thorium HTR (100 MWt) can supply the HOLCIM plant with the required energy. The optimum cycle configuration, as recommended, operates with a cycle efficiency of 42.4% while producing 39.867 MWe. A minimum of 10 MWe can be sold to the Indian distribution network at all times, thus generating revenue. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2014