Dissertations / Theses on the topic 'Soft clusters'

Two novel cyclodiphosphazane cluster frameworks with Cu₄X₄ clusters as tetrahedral nodes and ferrocenyl cyclodiphosphazanes [Fe(η5-C5H₄)₂(PNtBu)₂] as ditopic linkers have been synthesized. These frameworks having sodalite topology display a unique integration of porosity and redox activity and offer new opportunities for the synthesis of zeotype frameworks with soft phosphorus-based ligands.

Nanoparticles, simple yet groundbreaking objects have led to the discovery of invaluable information due to their physiological, chemical, and physical properties, have become a hot topic in various fields of study including but not limited to chemistry, biology, and physics. In the work presented here, demonstrations of various applications of chemical free nanoparticles are explored, from the determination of a non-invasive method for the study of the exposome via using soft-landing ion mobility (SLIM) deposited nanoparticles as a matrix-assisted laser desorption/ionization (MALDI-MS) matrix replacement, to the direct SLIM-exposure of nanoparticles onto living organisms. While there is plenty of published work in soft-landing at operating pressures of 1 Torr, the work presented here shows how this technology can be operated at the less common ambient pressure. The ease of construction of this instrument allows for various modifications to be performed for a wide array of applications, furthermore the flexibility in metallic sample, operating pressure, and deposition time only open doors to many other future applications. The work presented will also show that our ambient SLIM system is also able to be operated for toxicological studies, as the operation at ambient pressure opens the door to new applications where vacuum conditions are not desired.

The reaction of [BrRe(CO)₄]₂ with 2-(diphenylphosphino)pyridine (PN) and 6-(diphenylphosphino)-2-formylpyridine (PON) was investigated. The reactions were regiospecific and exclusively produced the phosphorus-coordinated products, BrRe(CO)₄(κᵖ-PN) and BrRe(CO)₄(κᴾ-PON). The kinetics for the chelate ring closure (κᴾ→ κᴾᴺ) in BrRe(CO)₄(κᴾ-PN) were confirmed to occur by dissociative CO loss. The reaction of [BrRe(CO)₄]₂ with 2-(diphenylphosphino)pyridine (PN) was modeled computationally by DFT calculations. The preferred reaction pathway for the substitution reaction was determined to occur by direct attack of the pnictogen donor on the dimer and formation of the κᴺ isomer as the kinetic substitution product occurs. The κᴺ kinetic product then rapidly isomerizes to the κᴾ thermodynamic product by way of a reversible ligand dissociation. Treatment of the tetrahedral cluster H₂Ru₃(CO)₃(μ₃-S) (1) with 2-(diphenylphosphino)thioanisole (PS) furnishes the cluster H₂Ru₃(CO)₇(κ²-PS)(μ₃-S) (2). Cluster 2, which exhibits a chelated thiophosphine ligand (κ²-PS), exists as a pair of diastereomers with Keq = 1.55 at 298 K that differ in their disposition of ligands at the Ru(CO)(κ²-PS) center. The PS ligand occupies the equatorial sites (Peq, Seq) in the kinetic isomer and axial and equatorial sites (Pax, Seq) in the thermodynamically favored species. The reversible first-order kinetics to equilibrium have been measured experimentally by NMR spectroscopy and HPLC over the temperature range 293-323 K. The substitution reaction involving 1 and the isomerization of the PS ligand in 2 were investigated by DFT calculations. The computational results support a phosphine-induced expansion of the cluster polyhedron that is triggered by the associative addition of the PS donor to 1. The observed isomerization of the PS ligand in 2 is best explained by a tripodal rotation of the CO and PS groups at the Ru(CO)(κ²-PS) center that is preceded by a regiospecific migration of one of the edge-bridging hydrides to the non-hydride-bridged Ru-Ru bond in 2. The chiral clusters 1,2-Ru₃(μ-H)₂(μ₃-S)(CO)₇(μ-1p1,2p2-POP) (A) and 1,2-Ru₃(μ-H)₂(μ₃-S)(CO)₇(μ-1p2,2p1-POP) (B) were formed were formed from reaction of Ru₃(μ-H)₂(μ₃-S)(CO)₉ with 1-diphenylphosphino-2-[2-(diphenylphosphino)ethoxy]benzene (POP). Chiral clusters A and B were fully characterized by IR and NMR spectroscopy. Additionally, the molecular structure of A was solved by X-ray crystallography. Chiral cluster A was resolved into its enantiomers by preparative HPLC with a chiral column. The enantiomers were characterized by electronic circular dichroism (ECD) spectroscopy and their absolute stereochemical configuration was determined by X-ray crystallography.

The arrangement of soft materials through solution processing techniques is a topic of profound importance for next generation solar cells; the resulting morphology has a major influence on construction, performance and lifetime. This thesis investigates the connections between the soft matter physics of colloidal systems and solid state dye sensitised (SSDS) and bulk heterojunction (BHJ) solar cells. A study of aqueous titanium dioxide nanoparticulate suspensions was carried out in order to observe how suspension structure can be controlled by altering the inter-colloid potential via pH-induced electrostatic charging. Measurements were performed at volume fractions between 0.025% and 8.2% with the solution pH set to 3.1, 3.5 or 4.5 before mixing. Suspensions with a volume fraction above 4% formed self-supporting gels regardless of the set pre-mix pH. These gels displayed shear thinning behaviour with a power law exponent of 0.8, a yield stress of 11(1) Pa and rheological response consistent with an aggregated fractal network. At lower volume fractions, suspensions exhibited consolidation interpreted as the collapse of a gel of fractal clusters with a fractal dimension of 2.36. The velocity of the suspension/supernatant interface exhibited delayed sedimentation behaviour, as well as further fractal-based power law scalings with volume fraction. Lower volume fraction suspensions were explored using dynamic light scattering. Limited aggregation of ‘stable’ suspensions was observed when compared to primary aggregate radii measured from electron microscopy images. To connect suspension structure and cell manufacture, the behaviour of more concentrated suspensions was observed during the drying of thin films, a process which forms an essential part of a SSDS solar cell. Lowering the pH of the suspension after mixing from 4 to 3 resulted in an ordering of observed crack domains. An increase in film delamination was also observed. Rates of mass loss during drying followed the expected three phase process, although there was an unexpected increase in rate during the initial phase (where rate is usually constant in time). Dynamic light scattering was found to be a useful but demanding technique for studying cluster formation in titanium dioxide suspensions. A non-linear fitting technique utilising the method of moments was thoroughly explored using computer simulated datasets. The algorithm reduced the systematic error in fitted parameters for moderately polydisperse (0:2 < < 0:4) datasets as compared to the commonly applied linear algorithm. The fitting algorithm was also robust to bad initial estimates of parameters. Finally, test solar cells have been built using blends of titanium dioxide and poly-3-hexylthiophene. Device performance was reduced with blend standing time after mixing but could be improved by remixing the blend before spin coating, implicating a reversible process (e.g. aggregation of titanium dioxide or crystallisation of P3HT) in the loss of performance. Addition of a titanium dioxide hole blocking layer before spin coating reduced cell performance. Combining the above studies and these device designs provides a future platform for continuation of this work in the context of real devices.

The main aim of this study is to examine the interrelationship of finance and government intervention in explaining the rise of co-productions in the international film industry in the time period between 1997 and 2004. Mainstream economic geography literature presents the film industry typically as a case study for embeddedness and agglomeration effects, with successful industry clusters drawing their strength from process knowledge, networks and local interaction. However, there is an increasing disparity in the literature between what mainstream theory suggests, and what empirical studies find with respect to the importance of cluster-external relations and dynamics. This, as I will argue, is particularly evident when looking at the picture of the whole film industry production system that emerges from the literature, which fails to include the alternative and complimentary pattern of co-productions. Co-productions are collaborations between film producers from at least two different countries, pooling their resources across distance to produce a feature film project. In the past fifteen years, the number of films made as co-productions has risen continuously in Europe, with co-productions accounting for more than 30 per cent of European film production activity. As a mode of production based on temporary, cross-border collaboration that is supported in its coordination by temporary clusters, such as trade fairs and industry events, the coproduction phenomenon poses a conundrum to economic geography literature and challenges its explanatory framework. As I will argue, in order to arrive at a satisfactory understanding of the phenomenon, it is necessary to look beyond social factors associated with locality, and to examine instead dynamics impacting on the industrial organization of the whole production system. I will argue that in the context of the pervasive demand uncertainty characterizing the film industry, the analytical focus should be on financial dynamics, as production activity and its organizational form are ultimately dependent on finance as an enabling force. Based on a description of the film financing process as the primary process in which the relationship between the economic categories of financial and production capital are played out, I propose that in order to explain the growth of co-productions empirically, it is necessary to examine changes in the film financing environments of the increasingly interrelated European and US film industries. As the State is the most important provider of financial capital in the European film industry through the provision of public aid, the focus will lie in particular on the consequences of a paradigm change in the rationale of State intervention in Europe moving away from funding film for cultural reason, to supporting the industry on economic grounds since the mid 1990s. As will be shown, the most important consequence of this paradigm change has been the introduction of tax incentives to encourage investment into film in a number of European and international countries within a short period of time. As will be demonstrated, this has led to the formation of significant, locally confined capital pools that can dis-embed production; and to the emergence of a distinct capital cycle in international film financing, which has strongly impacted on the productive system of the film industry. Finally, a dynamic explanation for the growth of co-productions in Europe in the time period between 1997 and 2004 will be provided. I will argue that co-productions have firstly grown in order to overcome a lack of finance, but have in the context of a capital cycle based on tax incentives from Germany and the UK, increasingly become driven by the opposite dynamic, namely an abundance of financial capital seeking profitable investment opportunities. The study will conclude with a discussion of policy implications, a summary of contributions to the literature and a brief overview of future research opportunities.

This thesis presents a shared memory parallel version of the hybrid classification algorithm IGSCR (iterative guided spectral class rejection), a novel data reduction technique that can be used in conjunction with PIGSCR (parallel IGSCR), a noise removal method based on the maximum noise fraction (MNF), and a continuous version of IGSCR (CIGSCR) that outputs soft classifications. All of the above are either classification algorithms or preprocessing algorithms necessary prior to the classification of high dimensional, noisy images. PIGSCR was developed to produce fast and portable code using Fortran 95, OpenMP, and the Hierarchical Data Format version 5 (HDF5) and accompanying data access library. The feature reduction method introduced in this thesis is based on the singular value decomposition (SVD). This feature reduction technique demonstrated that SVD-based feature reduction can lead to more accurate IGSCR classifications than PCA-based feature reduction. This thesis describes a new algorithm used to adaptively filter a remote sensing dataset based on signal-to-noise ratios (SNRs) once the maximum noise fraction (MNF) has been applied. The adaptive filtering scheme improves image quality as shown by estimated SNRs and classification accuracy improvements greater than 10%. The continuous iterative guided spectral class rejection (CIGSCR) classification method is based on the iterative guided spectral class rejection (IGSCR) classification method for remotely sensed data. Both CIGSCR and IGSCR use semisupervised clustering to locate clusters that are associated with classes in a classification scheme. This type of semisupervised classification method is particularly useful in remote sensing where datasets are large, training data are difficult to acquire, and clustering makes the identification of subclasses adequate for training purposes less difficult. Experimental results indicate that the soft classification output by CIGSCR is reasonably accurate (when compared to IGSCR), and the fundamental algorithmic changes in CIGSCR (from IGSCR) result in CIGSCR being less sensitive to input parameters that influence iterations.
Ph. D.

Recenti studi riguardanti gli Atomi di Rydberg hanno mostrato che, tali sistemi realizzano una peculiare interazione che sembra essere attiva soltanto entro una distanza finita. A partire da questa osservazione negli ultimi anni sono stati svolti studi teorici e sperimentali volti a descrivere le peculiari caratteristiche di tali sistemi. Un’analisi particolarmente interessante `e stata fatta focalizzando su sistemi unidimensionali implementati con bosoni hard-core e fermioni spinless, da cui è stato osservato che dal diagramma di fase si distinguono tre fasi: una prima che sembra soddisfare il paradigma dei liquidi di Luttinger, una seconda che ricalca una struttura critallina, ed una terza che si comporta come un liquido di Luttinger di cluster. Dove per cluster sono intesi particolari aggregati di particelle che vengono a formarsi in funzione del rapporto tra il raggio di interazione e la densita’ di particelle nel sistema. In questo lavoro è stato esteso lo studio a sistemi a due specie fermioniche, cercando di osservare il diagramma di fase solo in un particolare caso limite della teoria definito “limite ad una specie”. Da questa analisi si è dimostrato che in questo limite la struttura del diagramma di fase sembra ricalcare perfettamente quanto visto per i casi precedenti, in pieno accordo con le attese teoriche.

A día de hoy es prácticamente imposible encontrar una gran institución que no disponga de un parque de ordenadores considerable, debido al alto nivel de informatización de la sociedad actual. El enorme potencial que representan estos miles de ordenadores atrae poderosamente la atención en los ámbitos científicos e industriales, generando opciones viables para su aprovechamiento. Las universidades, instituciones que históricamente se han mantenido a la vanguardia en la investigación e innovación científica, representan un caso especialmente bien posicionado a la hora de generar tanto los recursos informáticos como la necesidad de su uso.
El poder de cómputo existente en los laboratorios y aulas de estudio universitarias, agrupaciones naturales de recursos informáticos, crea grandes oportunidades para la computación paralela, animándonos a buscar opciones viables para su aprovechamiento. Como resultado de este interés, nuestro grupo ha creado un entorno de planificación, enfocado hacia los clusters no dedicados. La constante y rápida evolución de los componentes, tanto a nivel de la arquitectura de la CPU como del sistema operativo, así como de las aplicaciones ejecutadas, hace que tengamos que adaptar nuestras propuestas. Nuestra propuesta consiste en crear una Máquina Virtual con doble funcionalidad, ejecutar la carga local de usuario y aprovechar los períodos de inactividad de nodos a efectos de poder usarlos para ejecutar carga paralela. Tanto el tipo de las aplicaciones como las características del hardware del escenario objetivo, y en el momento actual ambas han evolucionado. Nuevos tipos de aplicaciones paralelas con requerimientos periódicos de CPU son cada día más comunes en el mundo científico e industrial. Este tipo de aplicaciones pueden requerir un tiempo de retorno (turnaround) específico o una Calidad de Servicio (Quality of Service, QoS) determinada.
Para nuestro caso particular, reviste especial importancia el conocimiento que poseemos de los usuarios locales, debido a que nuestro entorno está diseñado para trabajar en clusters no dedicados. Un usuario local puede estar visualizando un vídeo almacenado en su ordenador, lo cual implica necesidades de CPU periódicas y un mayor uso de memoria. La aparición de nuevos tipos de aplicaciones, como vídeo bajo demanda ó realidad virtual, que se caracterizan por la necesidad de cumplir sus deadlines, presentando requerimientos periódicos de recursos. Este tipo de aplicaciones, donde la pérdida de deadlines no se considera un fallo severo, han sido denominadas en la literatura aplicaciones soft-real time (SRT) periódicas.
Esta interesante evolución de las necesidades de los usuarios no es el único digno de atención. El crecimiento en la capacidad de cómputo de los procesadores en los últimos años se ha visto frenado a causa de las barreras físicas del espacio y la velocidad de las señales, obligando a los fabricantes de procesadores a explorar otras vías de crecimiento. Desde hace ya algún tiempo el paralelismo de las aplicaciones se ha convertido en una de las grandes apuestas. Hoy en día los procesadores de dos núcleos son la mínima configuración que encontraremos en un ordenador, y se prevé que el número de núcleos continuará creciendo en los próximos años.
Los clusters no dedicados ofrecen un gran potencial de un uso, debido a que los recursos materiales ya están disponibles y el cálculo paralelo se realiza simultáneamente con el del usuario local. Imaginando el escenario actual en los clusters no dedicados, encontramos nuevas aplicaciones de escritorio y paralelas, así como plataformas hardware más potentes y complejas. En esta situación investigar el problema y realizar propuestas relacionadas con la planificación de los diferentes tipos de aplicaciones en clusters no dedicados, considerando las plataformas multicore, supone un nuevo reto a asumir por los investigadores y conforma el núcleo de este trabajo.
Today it is virtually impossible to find an institution that does not have a considerable amount of computers, due to the high level of computerization of society. The enormous potential of these large number of computers attract a lot of attention in science and industry, generating viable options for their use. The universities, institutions that historically have remained at the forefront of research and scientific innovation, represent a case particularly well positioned in generating both, computing resources and the need to use.
The computational power present in laboratories and university study rooms, natural groupings of information resources, creating great opportunities for parallel computing, encouraging us to seek viable options for their use. As a result of this interest, our group has created a parallel scheduling environment, focused on non-dedicated clusters. The constant and fast evolution of the components, both at the architecture of the CPU and the operating system and applications executed, forces us to adapt our proposals. Our proposal is to create a Virtual Machine with dual functionality, run the local load user and take advantage of downtime for the purposes of nodes it can be used to run parallel load. At present both, applications and hardware specifications of the target scenario, have evolved. New types of parallel applications with periodic CPU requirements are becoming more common in science and industry. Such applications may require a return time (turnaround) or a specific QoS (Quality of Service).
Since our framework is designed to work in non-dedicated clusters, having knowledge of the local users behavior is of particular importance. A local user may be viewing a video stored on your computer, which involves periodic CPU requirements and increased use of memory. The emerging new types of applications, such as video on demand or virtual reality are a fact. This new types of applications are characterized by the need to meet their deadlines, presenting periodic resource requirements. This type of application, where the loss of deadlines is not considered a severe failure, has been named in the literature uses soft-real time (SRT) at regular intervals.
This exciting evolution of user needs is not the only one worthy of attention. The growth in computing power of processors in recent years has been hampered because of the physical barriers of space and speed of the signals, forcing chip makers to explore other avenues of growth. For some time the parallelism of the applications has become one of the biggest bets. Today's dual-core processors are the minimum configuration of any computer, and it is expected that the number of nuclei continue to grow in the coming years.
The non-dedicated clusters offer great potential for use, because the computational resources are already available, and parallel computing is performed simultaneously with the local user. Figuring out the current scenario in the non-dedicated clusters, we find new desktop applications, parallel and more powerful and complex hardware platforms. In this situation, research lines related to the planning of the different types of applications in non dedicated clusters, considering multi-core platforms, is a new challenge to be assumed by researchers and constitute the core of this work.

The use of Insulated Gate Bipolar Transistors (IGBT) have enabled better switching performance than the Metal Oxide Semiconductor Field effect Transistor (MOSFET) in medium to high power applications due to their lower on-state power loss and higher current densities. The power ratings of IGBTs are slowly increasing and are envisaged to replace thyristors in medium power applications such as High Voltage Direct Current (HVDC) inverter systems and traction drive controls. Devices such as the MOS Controlled Thyristor (MCT) and Emitter Switched Thyristor (EST) were developed in an effort to further simplify drive requirements of thyristors by incorporating a voltage controlled MOS gate into the thyristor structure. However, the MCT is unable to achieve controlled current saturation which is a desirable characteristic of power switching devices while the EST has only limited control. The IGBT can achieve current saturation, however, due to the transistor based structure it exhibits a larger on-state voltage in high power applications compared with thyristor based devices. MOS Gated Thyristor (MGT) devices are a promising alternative to transistor based devices as they exhibit a lower forward voltage drop and improved current densities. This current research focuses on the Clustered Insulated Gate Bipolar Transistor (CIGBT) whilst being operated under soft-switching regimes. The CIGBT is a MOS gated thyristor device that exhibits a unique self-clamping feature that protects cathode cells from high anode voltages under all operating conditions. The self-clamping feature also enables current saturation at high gate biases and provides low switching losses. Its low on-state voltage and high voltage blocking capabilities make the CIGBT suitable as a contender to the IGBT in medium to high power switching applications. For the first time, the CIGBT has been operated under soft-switching regimes and transient over-voltages at turn-on have been witnessed which have been found to be associated with a number of factors. The internal dynamics of the CIGBT have been analysed using 2D numerical simulations and it has been shown that a major influence on the peak voltage is the P well spacing within the CIGBT structure. For example, Small adjacent P well spacings within the device results in an inability for the CIGBT to switch iv on correctly. Further to this, implant concentrations of the n well region during device fabrication can also affect the turn-on transients. Despite this, the CIGBT has been experimental analysed under soft-switching conditions and found to outperform the IGBT by 12% and 27% for on-state voltage drop and total energy losses respectively. Turn off current bumps have been seen whilst switching the device in zero voltage and zero current switching mode of operation and the internal dynamics have been analysed to show the influence upon the current at turn off. Preliminary results on the Trench CIGBT (TCIGBT) under soft switching conditions has also been analysed for the first time and was found to have a reduced peak over-voltage and better switching performance than the planer CIGBT. Through optimisation of the CIGBT structure and fabrication process, it is seen that the device will become a suitable replacement to IGBT in medium power application.

Heterogeneity in modern datacenters is on the rise, in hardware resource characteristics, in workload characteristics, and in dynamic characteristics (e.g., a memoryresident copy of input data). As a result, which machines are assigned to a given job can have a significant impact. For example, a job may run faster on the same machine as its input data or with a given hardware accelerator, while still being runnable on other machines, albeit less efficiently. Heterogeneity takes on more complex forms as sets of resources differ in the level of performance they deliver, even if they consist of identical individual units, such as with rack-level locality. We refer to this as combinatorial heterogeneity. Mixes of jobs with strict SLOs on completion time and increasingly available runtime estimates in production datacenters deepen the challenge of matching the right resources to the right workloads at the right time. In this dissertation, we hypothesize that it is possible and beneficial to simultaneously leverage all of this information in the form of declaratively specified spacetime soft constraints. To accomplish this, we first design and develop our principal building block—a novel Space-Time Request Language (STRL). It enables the expression of jobs’ preferences and flexibility in a general, extensible way by using a declarative, composable, intuitive algebraic expression structure. Second, building on the generality of STRL, we propose an equally general STRL Compiler that automatically compiles STRL expressions into Mixed Integer Linear Programming (MILP) problems that can be aggregated and solved to maximize the overall value of shared cluster resources. These theoretical contributions form the foundation for the system we architect, called TetriSched, that instantiates our conceptual contributions: (a) declarative soft constraints, (b) space-time soft constraints, (c) combinatorial constraints, (d) orderless global scheduling, and (e) in situ preemption. We also propose a set of mechanisms that extend the scope and the practicality of TetriSched’s deployment by analyzing and improving on its scalability, enabling and studying the efficacy of preemption, and featuring a set of runtime mis-estimation handling mechanisms to address runtime prediction inaccuracy. In collaboration with Microsoft, we adapt some of these ideas as we design and implement a heterogeneity-aware resource reservation system called Aramid with support for ordinal placement preferences targeting deployment in production clusters at Microsoft scale. A combination of simulation and real cluster experiments with synthetic and production-derived workloads, a range of workload intensities, degrees of burstiness, preference strengths, and input inaccuracies support our hypothesis that leveraging space-time soft constraints (a) significantly improves scheduling quality and (b) is possible to achieve in a practical deployment.

Superaggregates are clusters formed by diverse aggregation mechanisms at different scales. They can be found in fluidized nanoparticles and soot formation. An aggregate, with a single aggregation mechanism, can be described by the fractal dimension, df , which is the measure of the distribution and configuration of primary particles into the aggregates. Similarly, a su-peraggregate can be analyzed by the different fractal dimensions that are found at each scale. In a fractal structure aggregate, a self-similarity can be identified at different scales and it has a power law relation between the mass and aggregate size, which can be related to properties like density or light scattering. The fractal dimension, df , can be influenced by aggregation mechanism, particles concentration, temperature, residence time, among other variables. More-over, this parameter can help on the estimation of aggregates’ properties which can help on the design of new processes, analyze health issues and characterize new materials.A multi-dimensional soot aggregate was simulated with the following approach. The first aggregation stage was modeled with a Diffusion Limited cluster-cluster aggregation (DLCA) mechanism, where primary clusters with a fractal dimension, df1, close to 1.44 were obtained. Then, the second aggregation stage was specified by Ballistic Aggregation (BA) mechanism, where the primary clusters generated in the first stage were used to form a superaggregate. All the models were validated with reported data on different experiments and computer models. Using the Ballistic Aggregation (BA) model with primary particles as the building blocks, the fractal dimension, df2, was close to 2.0, which is the expected value reported by literature. However, a decrease on this parameter is appreciated using primary clusters, from a DLCA model, as the building blocks because there is a less compact distribution of primary particles in the superaggregate’s structure.On the second aggregation stage, the fractal dimension, df2, increases when the superaggre-gate size increases, showing an asymptotic behavior to 2.0, which will be developed at higher scales. Partial reorganization was implemented in the Ballistic Aggregation (BA) mechanism where two contact points between primary clusters were achieved for stabilization purposes. This implementation showed a faster increase on the fractal dimension, df2, than without par-tial reorganization. This behavior is the result of a more packed distribution of primary clusters in a short range scales, but it does not affect the scaling behavior of multi-dimensional fractal structures. Moreover, the same results were obtained with different scenarios where the building block sizes were in the range from 200 to 300 and 700 to 800 primary particles.The obtained results demonstrate the importance of fractal dimension, df , for aggregate characterization. This parameter is powerful, universal and accurate since the identification of the different aggregation stages in the superaggregate can increase the accuracy of the estimation of properties, which is crucial in physics and process modeling.

Background: In England, 70% of young people consume soft drinks on a regular basis which increases the likelihood of obesity and dental caries. Many young people often visit the dentist; however primary dental care has been an underutilised research setting for delivering and evaluating dietary interventions. Motivational Interviewing (MI) has been shown to be effective in behaviour change. Therefore, this study provides an opportunity to assess an obesity intervention to reduce soft drinks consumption among young people attending NHS dental practices. Aims: To conduct a feasibility trial of an MI intervention to reduce soft drinks consumption among young people attending primary dental care practices in North Central London (Camden, Islington, and Haringey). Methods: The initial developmental phase explored dental teams’ attitudes, behaviours and perceived barriers in delivering prevention, using a mixed methods approach (focus groups followed by a questionnaire survey). A purposive sample of 11 out of 22 dental practices was selected from the study area for the four focus groups (two with dentists and two with dental care professionals). A cross-sectional survey of all NHS dentists was also conducted using a self-complete questionnaire. Descriptive analysis of the questionnaire data was undertaken and differences by the age and sex of the sample were assessed. A cluster feasibility trial with embedded process evaluation was then undertaken in the second phase of the study. Participants: Ten randomly allocated NHS dental practices recruited 39 overweight or obese young people (11-16 years). Intervention: 3-4 MI sessions. The three main outcomes were: mean daily consumption of sugary soft drinks using a 24 hour dietary recall, body mass index (BMI) and waist circumference. Outcomes were measured at baseline and at 6 month follow-up. Process evaluation outcomes: Recruitment, retention, fidelity and acceptability of the intervention were assessed using mixed methods. MI Fidelity was assessed using the Motivational Interviewing Treatment Integrity Code (MITI). Thematic analysis was used to analyse the qualitative data and descriptive analysis for the quantitative data. The study was approved by the local Ethics Committee (10/H0722/2). Results: Of the 300 eligible respondents, 164 questionnaires were returned (response rate: 54.7%). In Phase I, there were broad similarities in the themes that emerged from the four focus groups and the questionnaire survey. The majority of respondents strongly agreed that NHS dentists had a role to play in smoking cessation counselling, diet and alcohol advice. For dentists, prevention was viewed to be part of their ethical obligation. A high proportion of respondents 95.7% delivered oral hygiene advice, 85.4% provided diet advice, and 76.7% offered smoking cessation advice always or frequently. Barriers to delivering prevention included organisational factors (lack of adequate remuneration, bureaucracy), patient-related factors (motivation and compliance), and clinician-related factors (lack of training and resources). There was significant variation in providing prevention by the age and sex of dentists with female and younger dentists more likely to provide certain aspects of preventive care. Phase II- 10 dental practices were recruited (5 control and 5 intervention). Out of 149 potential participants, 39 (26.2%) conformed to the eligibility criteria, 5 (12.8%) withdrew from the study, and 3 (7.6%) were lost to follow-up, resulting in a retention rate of 79.5%. Facilitators of study implementation included the use of a screening questionnaire, support from the research team and incentives for young people and dental teams. In terms of acceptability, participants reported that the study had a positive impact on their knowledge or behaviours. At follow-up, soft drinks consumption decreased. Dental teams valued being involved in research. They recognised the challenges in recruiting young people and employed varying mechanisms to overcome them. Intervention fidelity revealed the MI practitioners to be proficient in MI and there was little variation in proficiency among the practitioners. Conclusions: Dentists in the study area were actively involved and had a generally positive attitude towards delivering prevention in primary dental care. The intervention was acceptable and feasible among dental teams and young people. MI is a feasible tool for reducing sugary soft drinks consumption among overweight young people attending dental practices. Further research is required to assess its effectiveness in primary dental care.

This thesis addresses the electronic structure of vanadium and copper oxides using soft X-ray absorption (SXA) spectroscopy and resonant inelastic X-ray scattering (RIXS) at high brightness synchrotron radiation sources. In RIXS incident photons, tuned to the energy of specific absorption resonances, are inelastically scattered leaving behind a low energy valence excitation in the system studied. Effects of electron localization are reflected by the occurrence of low-energy excitations in form of dd- and charge-transfer excitations that are modelled by cluster calculations. Band-like states are dominating when the intermediate core excited state is delocalized.

RIXS at V 2p and O 1s resonances has been used to study the electronic structure of the monovalent vanadium oxides VO₂ and V₂O₃, and of the mixed valence compounds, NaV₂O₅ and V₆O₁₃. For NaV₂O₅ and V₆O₁₃ significant contributions from localized low-energy excitations reflect the partly localized character of their valence band electronic structure, whereas VO₂ and V₂O₃ appear mostly as band-like. Effects of carrier doping are addressed for the case of Mo doping into VO₂ and reveal a quasi-rigid band behavior. In the cases of VO₂ and V₆O₁₃ the temperature dependent metal-insulator transition could be monitored by following the spectral evolution of bands originating from V 3d and V 3d - O2p hybridized states. For Na₂V₃O₇ nanotubes it was possible to selectively probe states from the apical and the basal oxygen sites of VO₅ pyramids that constitute these nanotubes. Furthermore, the RIXS technique has been demonstrated to be highly valuable in characterizing the charge transfer processes that accompany lithium insertion into vanadium oxide battery cathodes. Finally, for insulating cuprates RIXS at O 1s, Cu 3p and Cu 3s resonances has been recorded at high-resolution for the detailed investigation of crystal field excitations.

>Magister Scientiae - MSc
Ambient air pollution is the biggest environmental threat to human health. According to the World Health Organisation (WHO), ambient air pollution kills millions of people worldwide every year. Airborne particulate matter (APM) affects more people than any other air pollutant and has been linked with various adverse health outcomes, especially fine fractions (commonly abbreviated to PM2.5). PM2.5 penetrates lung tissue to enter the cardiovascular system where it poses the greatest risk. Detailed ambient APM studies are rare in Africa. Such studies are needed to better understand the characteristics, origins and trends of particulate pollution. This study was conducted in Cape Town (the first of its kind for the area) as part of a bigger project on ambient PM2.5 and soot concentrations in South Africa. PM2.5 filter samples were collected at a fixed sampling site in the suburb of Kraaifontein from April 2017 to April 2018, yielding 121 days of data. PM2.5 mass concentration and absorption coefficient determinations were done using gravimetric analysis and smoke stain reflectometry (SSR). Mean PM2.5 concentration for the study period was 13.4 ± 8.1 μg.m-3 (range: 1.17-39.1 μg.m-3) that fell below the South African National Ambient Air Quality Standard (SA NAAQS) annual limit of 20 μg.m-3 but exceeded the WHO annual limit of 10 μg.m-3. Mean absorption coefficient for the same period was 1.38 ± 1.23 m-1.10-5 (range: 0.00-5.38 m-1.10-5) which did not exceeded any limits. Source-region analyses were performed using single, 24-hour backward trajectories and trajectory clusters derived from the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model. Six single trajectories were identified; the most frequent were trajectories Atlantic-Ocean (38.8 %) and Indian-Ocean (26.4 %). Cluster analyses yielded three to four clusters per season. Dominating clusters were Atlantic-Ocean (61.8 %) and Indian-Ocean (29.5 %) and Inland (8.50 %). Contributions by local sources (within 40 km of the sampling site) to PM mass in samples far exceeded those of distant sources through long-range transport (LRT).

Nous avons proposé des techniques de RMN à une et deux dimensions pour mesurer quantitativement les saturations et dynamiques de différents mélanges de saumure/huile brute/filtrat de boue confinés dans des roches pétrolières. La plupart de ces techniques peuvent être utilisées dans les puits. Les échantillons de roches ont été caractérisés par les techniques standards de la pétrologie. La résonance paramagnétique électronique (RPE) a mis en évidence les sources de relaxation composées d'ions métalliques paramagnétiques de vanadyl (VO2+) et de radicaux libres stables dans des huiles brutes avec et sans asphaltène. La micro-chromatographie en exclusion de tailles couplée à la détection par spectroscopie de masse en haute résolution a démontré le piégeage de ces ions métalliques dans des composés de type " métallo-porphyrine " (MP) inclus dans les nanoagrégats d'asphaltène. Ces deux dernières techniques ont montré qu'environ 2/3 des ions VO2+ sont inclus dans les parties polaires de l'asphaltène tandis que 1/3 sont libres en solution. L'utilisation jointe des chromatographies en phase gazeuse (GC) et par perméation de gel ou d'exclusion stérique (GPC) a permis d'étendre considérablement l'étendue spatiale des distributions de longueurs de chaînes des hydrocarbures. Les distributions des temps de relaxation transverses et longitudinaux ont la même allure que celle obtenues par GC et GPC. Ce qui prouve que la dynamique des hydrocarbures est fortement corrélée à la longueur des chaînes. La partie centrale de notre étude a consisté en la proposition de nouvelles séquences de RMN à deux dimensions (D-T2) corrélant les distributions des coefficients de diffusion D et les temps de relaxation transverses T2. Nous avons appliqué ces séquences (D-T2) pour suivre continument les saturations progressives de fluides pétroliers confinés dans des grès poreux en conditions monophasique, diphasique et triphasique à température et pression variables. On a pu ainsi suivre continument, pour la première fois, les saturations et les diffusions de ces différents fluides au cours d'expériences d'imbibition-drainage dans des grès pétroliers. Finalement, on a confirmé les comportements anormaux D∝√T2 pour les T2 courts (longues chaînes d'hydrocarbures) dans le cas des huiles brutes avec asphaltène. On a aussi observé un comportement D≈Cte pour les T2 longs (courtes chaînes d'hydrocarbures). Un scaling non-linéaire T2->√T2 a montré que c'est au niveau de la dynamique moléculaire (de rotation et de translation) des hydrocarbures de longues chaînes moléculaires que résident les principaux effets de l'asphaltène. Le comportement nouveau D≈Cte a pu être expliqué en termes d'une affinité dynamique transitoire entre les petits hydrocarbures et les nanoagrégats d'asphaltène. Ce qui est compatible avec les dépendances bilogarithmiques des variations de la vitesse de relaxation longitudinale moyenne avec la fréquence de Larmor que nous avons observées dans le cas des huiles brutes avec asphaltène.

Current CDMA networks do not support soft handoffs between adjacent cell-clusters serviced by different mobile switching centers (MSCs). Three novel network configurations are proposed here to overcome this limitation. To allow diversity combining of signals, configurations I and II employ cross links from boundary cells to MSCs serving opposite cell-clusters. On the other hand, configuration in direcdy connects the MSCs of adjacent cell-clusters together to accomplish this objective. A channel assignment scheme associated with configuration I is proposed to reduce the handoff blocking probability by lengthening the handoff transition time. Configuration II expands each cell-cluster by some layers of cells which overlap with adjacent cell-clusters, thus enabling the use of different inter-cluster handoff regions for different handoff directions, an innovation which prevents handoff oscillations which may occur in configuration I while also reducing the necessary number of inter-cluster handoffs. Configuration in accommodates the highest number of calls of the three configurations by allowing the handoff and new calls to fully share the channels linking each cell to the respective MSC. A secondary contribution of this thesis is the development-of a mobility model which enables dimensioning of trunk groups for each configuration. In configurations I and II, the optimal partitioning of trunk groups serving a boundary cell into direct and cross links is determined by mathematical analysis and computer simulation. For configuration IB, the optimal number of trunks required to interconnect adjacent MSCs is investigated by computer simulation. Both vehicle and pedestrian traffic is examined and the performance in all three configurations is compared. Each configuration has its own distinct features: scheme I, configuration I is the simplest; scheme II, configuration I produces the lowest handoff blocking probability by trading off the new call blocking probability; configuration II eliminates the handoff oscillations and reduces the number of inter-cluster handoffs; configuration IE provides the best trunking efficiency to accommodate the highest number of calls.

‘Soft condensed matter’ is a newly-emerged sub-discipline of physics concerned with the study of systems that are mechanically soft such as colloids, emulsions, surfactants, polymers, liquid crystals, granular media and various biomaterials including DNA and proteins. These materials display a broad range of interesting microstructures and phase behaviours and have a myriad of applications in the materials, food, paint and cosmetic industries as well as medical technologies. Soft condensed matter physics presents new opportunities and challenges for the development of new ideas and concepts in experimental and theoretical physics alike. Because the field overlaps with many different disciplines, the study of soft matter also offers promising developments to other fields of science including chemistry, chemical engineering, materials science, biology, and environmental science. The behaviour of these systems is dominated by one simple fact: they contain mesoscopic structures in the size range 10 nm to 1 µm that are held together by weak entropic forces. The elastic constants of these materials are 109 times smaller than the conventional atomic materials and hence are easily deformable by external stresses, electric or magnetic fields, or even by thermal fluctuations. We have studied two important classes of soft matter systems in this thesis -colloidal suspensions and surfactant systems. The thesis is divided into two main themes: (a) Effects of electric field on the colloidal suspensions, and (b) Effects of shear on surfactant solutions. Motions of colloidal particles under the influence of applied electric field were observed under a microscope and were studied using image analysis and particle tracking. We have also used tracking of thermal fluctuations of colloidal particles embedded in surfactant gels to study microrheology of surfactant solutions. Linear and non-linear rheology of aqueous solutions of cationic cetyltrimethyl ammonium bromide (CTAB) and anionic sodium-3-hydroxynapthalene-2-carboxylate (SHNC) were studied using bulk rheology in a commercial rheometer. Rheological studies of an anionic surfactant sodium dodecyl sulphate (SDS) in the presence of strongly binding counterion p-toluidine hydrochloride (PTHC) has also been done. Chapter 1 starts with a general introduction to soft condensed matter systems and then we proceed to describe two specific class of soft condensed materials which we have studied in this thesis -colloidal suspensions and surfactant/water systems. After describing different types of colloids, we discuss why colloids are suitable as model systems in condensed matter physics. This is followed by a discussion on the chemical structure, phase behaviour and self assembling properties of surfactant molecules in water. We then discuss the inter-macromolecular forces such as van der Waals interaction, the screened Coulomb repulsion, hydrogen bond, hydrophobic and hydration forces and steric repulsion which are the major players in the interaction in soft condensed matter systems. The systems that have been the subject of our experimental studies, viz. polystyrene colloidal suspensions, CTAB+SHNC, SDS+PTHC and CTAT have also been discussed in detail. Then we have given an overview of effects of electric field on the colloidal suspensions. Two types of geometries have been discussed: one in which the field is parallel to the plates and another when the field is perpendicular to the electrodes. Application of colloidal particles in diagnostic tests (Latex Agglutination Tests) has been discussed after this. Some methods used to enhance the sensitivity of LATs have also been reviewed. This is followed by a theoretical background of linear and non-linear rheology. We have also given an introduction to digital video microscopy, its advantages and discussed few quantities like pair correlation function, structure factor which can be extracted using digital video microscopy and particle tracking. The concluding part of this chapter describes the organization of this thesis. Chapter 2 discusses the experimental apparatus and techniques used in our studies. We describe our setup for applying the electric field to the colloidal particles and imaging and tracking their motion. We also discuss the image processing and analyzing methods for extracting the useful quantities from the digitized images. We have described the various components of the MCR-300 stress-controlled rheometer (Paar Physica, Germany) and the AR-1000N stress-controlled rheometer (T. A. Instruments, U. K.) followed by different experimental geometries that we have used for our experiments. Next we have described the various experiments that can be done using a commercial rheometer. Calculation of surface charge of colloidal particles using a conductivity meter has been demonstrated for our colloidal particle suspensions. We also describe the sample preparation methods employed in different experiments. In Chapter 3, we have discussed our study of clustering of colloidal particles under the influence of an ac electric field as a function of frequency. The field was applied in a direction perpendicular to the confining walls. Two regimes are observed, a low frequency regime where the clusters are isotropic with a local triangular order and a new high-frequency regime where the clusters are highly elongated (anisotropic) with no local order. The crossover from one regime to the other occurs at a critical frequency, fc. The formation of elongated clusters seen at high frequencies is explained in terms of rotation of particles due to a phase lag between the polarization of the electric double layer around a particle and the applied electric field that arises because of inhomogeneities of the conducting surface. We have also observed that the threshold field for the cluster formation, Eth, increases with frequency in both the regimes. We did these studies on two different sizes of particles and found that both Eth and fc were lower for the larger particles. Our model based on particle rotation was able to estimate the value of fc correctly for both the sizes of the particles. Chapter 4 describes a method employing an ac electric field applied perpendicular to the confining walls to increase the sensitivity of recognition of ligands by their corresponding receptors grafted on Brownian latex particles. Application of electric field assists the colloidal micro-particles grafted with receptors to come nearer due to electro-hydrodynamic drag. This increase in the local concentration of the latex particles results in improving the chances of ligand-receptor interaction leading to the aggregation of the latex particles. With this technique we have been able to increase the sensitivity of the ligand-receptor recognition by a factor as large as 50. We have demonstrated the utility of our method using streptavidin as the model receptor and biotinylated RNase A as the model ligand. We have also applied our technique to a commercially available kit for rheumatoid factor (RF) with successful results. The same method was also successfully applied for the detection of typhoid whose antibodies were purified and attached to polystyrene particles by our collaborators from DRDE Gwalior. In Chapter 5, we have studied the statics and dynamics of colloidal particles at different applied electric fields from zero to beyond the threshold field. We have taken a series of time-lapsed images and calculated out the pair-correlation function, mean squared displacement, structure factor, non-Gaussian parameter etc. We have studied both mono-dispersed colloidal system and binary colloidal system (mixture of two different sizes of particles). The aggregates formed in the two cases were analysed with the help of Voronoi polygons to quantify the microscopic structure. In mono-dispersed system, the aggregates formed were two-dimensional hexagonal crystals and we have used this system to study the freezing transition in 2-dimension. The properties of the system in the liquid and the crystalline state satisfy various criteria for the 2-d freezing transition. The first maximum of the structure factor at the voltage at which freezing occurs, is 5.5 as has been suggested for the 2-d freezing. This is reflected in the dynamics of the system also, where the ratio D/D0 falls below 10%, in accordance with the LPS (L¨owen, Palberg, Simon) criterion for freezing in 2-d colloidal systems [Phys. Rev. Lett. 70, 1557 (1993)]. However, in the binary colloidal system the clusters formed were not crystalline but more like 2-d dense liquids. A closer inspection of these clusters reveals that the motion of a smaller subset of particles is cooperative and follows string-like paths. The mean square displacement of such a system shows a plateau in the intermediate times which indicates the “caging” of particles by its neighbours. A peak in non-gaussian parameter indicates the presence of dynamical heterogeneities in the system. In Chapter 6, we have described the use of multiple particle tracking to study the microrheology of semidilute solutions of wormlike micelles and compared the results with those from macrorheology experiments done on the same samples. Two concentrations of CTAT (1.3% and 2%) were used. We observed that, in spite of the mesh size being much smaller than the size of the probe particles, the viscoelastic response function calculated using the one-point microrheology does not match with that measured from macrorheology. This can be attributed to the fact that there is another important length scale in the system, the mean micellar length, and it is comparable to the probe particle size. Two-point microrheology was successful in verifying the macrorheology results for CTAT 1.3% but it fails to do so for CTAT 2%. We attribute this to the fact that in a higher viscosity sample (2%), the hydrodynamic force propagate to a lesser distance, thereby limiting the measurable correlation between the particles and precluding the success of two-point microrheology. Chapter 7 describes a rheological study of aqueous solutions of varying concentration of cationic cetyltrimethyl ammonium bromide (CTAB) and anionic sodium-3-hydroxynapthalene-2-carboxylate (SHNC) kept at a fixed molar concentration ratio [CTAB]/[SHNC] = 2. At this molar ratio, the surfactants self-assemble into wormlike micelles which get entangled above the overlap concentration to form viscoelastic gel. The range of the total surfactant concentration φ varies from 1.17% to 5.16% by weight. We found that, plateau modulus, G0, shows a power law dependence on the surfactant concentration, φ, with an exponent 3, which is higher than the expected value of 2.25 observed for the one-component wormlike micelles. Zero shear viscosity, η0, and relaxation time, τR show a maximum at the surfactant concentration, φmax = 1.9% in contrast to a monotonic increase with φ. We propose that this non-monotonic behaviour is due to the unusual dependence of the average micellar length L ¯on φ, showing a maximum in average micellar length L at φmax. This argument provides a strong support to the model of micellar growth in the presence of electrostatic interactions developed by Mackintosh et. al [Europhys. Lett. 12, 697 (1990)]. The presence of electrostatic interactions also appears in the behaviour of the plateau modulus G0 that exhibits a larger φ dependence than in highly screened micelles. In the non-linear flow experiments, a minimum observed in critical shear rate (the shear rate at which shear thinning starts), ˙γc, at φmax strengthens our arguments. In Chapter 8, we describe the phase behaviour and rheology of SDS+PTHC (sodium dodecyl sulphate + p-toluidine hydrochloride) micellar solutions at different molar ratios α=[PTHC]/[SDS]) of the two components. At low values of α, polarizing microscopy observations reveal a transition from an isotropic to a nematic phase of disk-like micelles, whereas a transition to a lamellar phase occurs at higher α values > 0.5, on increasing the surfactant content. Linear rheology of the isotropic micellar solution reveal a viscous behaviour over a large range of surfactant concentrations. Surprisingly, this also extends to the nematic phase of disk-like micelles observed at α =0.2 and φ =0.35. These systems also exhibit a viscoelastic behaviour over a narrow range of surfactant concentration as reported in earlier studies. The extent of the viscoelastic region of the isotropic micellar solution also decreases with increase in α. Frequency sweep curves in this region, scaled on to a master curve is reminiscent of dilute suspensions of hard spheres or rigid Brownian rods. Consistent with the results from oscillatory shear measurements, the f;ow behaviour examined under steady shear is Newtonian over a large range of surfactant content in the isotropic micellar solution. An interesting result in these studies is the non-monotonic behaviour of the viscosity with increase in surfactant concentration. It is likely that the sharp rise in viscosity arises from a jamming effect of the rigid rods. Dynamic light scattering studies suggest that the drop in viscosity is due to the decrease in the length of the micellar aggregates. This is followed by a change in the morphology of the micelles from rods to disks as indicated by the transition to a nematic phase of disk-like micelles or a lamellar phase. A change in the morphology of micellar aggregates with increase in α is expected in mixed surfactant systems with strongly binding counterions. However, the surprising result is the change in morphology of the micellar aggregates with surfactant content. Such a behaviour is seen in mixed surfactant systems for the first time. The thesis concludes with a summary of our main results and a brief discussion of the scope of future work in Chapter 9.

Over the past several years, there has been an explosion in the amount of medical data generated and subsequently collected in medical domain. Data mining techniques have been used extensively in mining the medical data. Obtaining high quality data mining results is very challenging because of the inconsistency of the results of different data mining algorithms and noise in the medical data.

This thesis presents a novel hybrid data mining approach for knowledge extraction and classification in medical databases. The proposed approach is formulated to cluster extracted features from medical databases into soft clusters using unsupervised learning strategies and fuse the decisions using serial and parallel data fusion techniques. The idea is to observe associations in the features and fuse the decisions made by learning algorithms to find the strong clusters which can make impact on overall classification accuracy. The novel techniques such as serial cascaded data fusion, parallel majority-voting based neural data fusion and parallel neural network based data fusion are proposed that allow integration of various clustering algorithms for hybrid data mining approach.

The proposed approach has been implemented and evaluated on the benchmark databases such as Digital Database for Screening Mammograms, Wisconsin Breast Cancer, Pima Indian Diabetics and ECG Heart Arrhythmia.

A comparative performance analysis of the proposed hybrid data mining approach with other existing approaches for knowledge extraction and classification is presented. The experimental results demonstrate the effectiveness of the proposed approach in terms of improved classification accuracy on benchmark medical databases.

碩士
開南大學
財務金融學系
97
This study utilized Back-Propagation Neural Network (BPNN)、Support Vector Machine(SVM) and Genetic Algorithm Fuzzy Decision Tree (GAFDT) as three primary prediction models to forecast the stock trading signals. And the data clustered or not will be compared in both models. The object of this research was the Weighted Price Index of the Taiwan Stock Exchange. In 25 technical indices, we first used stepwise regression analysis to sieve out relative important factors, and k-means cluster analysis was adopted for data clustering. Then five-full alternation was applied in the experimental design process. Model from one to three are utilized to predict by BPNN, SVM and GAFDT, respectively. GAFDT began with transferring continuous data to discrete data by fuzzy sets so as to increase the comprehensibility of decision tree, and then utilized Genetic Algorithms to optimize the parameters like fuzzy term numbers. Finally, we compare the different intervals between Weighted Price Index of the Taiwan Stock Exchange and S&P 500 by GAFDT. Empirical findings revealed that the highest accurate rate was GAFDT (82.93%), followed by SVM (76.12%) and BPNN (75.53%). The more clusters GAFDT and SVM swarm off, the higher the accurate rate is. The accurate rate of falling of stock was higher than correction and rising in Weighted Price Index of the Taiwan Stock Exchange. The accurate rate of rising of stock is higher than correction and falling in S&P 500. Consequently, the prediction will be affected by training and testing in different stock markets and tendencies.

碩士
國立臺灣大學
材料科學與工程學研究所
105
Over the past few years, gas cluster ion beams (GCIB) has shown great capability of dealing with bio-materials and soft materials owing to its high sputter yield and low damage accumulation that preserved the molecular structures during depth profiling, therefore molecular ion of high mass can be obtained in subsequent analysis. However, although GCIB has lower damage accumulation comparing with C60+ and monoatomic ions, the inevitable alteration in chemical structure can still change the property of remaining surface and gradually affect the accuracy of depth profile. As a result, artifacts can still be observed in the resulting depth profiles. In order to further improve the depth profile of soft materials, low energy O2+ can be used to cosputter the surface to enhance the ionization yield and mask the damage accumulation. While the energy per atom (E/n) is known to be another important factor to the sputter process and previous works concluded higher E/n is beneficial, how the change in E/n affects GCIB-O2+ cosputter depth profile is not clear yet. In this study, bulk (Polyethylene Terephthalate, PET) was chosen as the modeling material and 10 - 20 kV Ar1000-4000 (E/n = 2.5 - 20) with or without O2+ beam was used to cosputter the surface. Spectra at different depths were obtained by a time-of-flight secondary ion mass spectrometer (ToF-SIMS) with pulsed Bi32+ as primary ion to construct the depth profile. After sputtering, the craters were measured by alpha step and atomic force microscope (AFM) with quantitative imaging mode. The result shows that with higher E/n, the resulting surface is more rough and the surface Young’s modulus became smaller compared with the pristine surface. This result suggested that depolymerization took place. Also, relative intensity (I) was comparatively weak in the depth profile, which indicated more damage accumulated. Furthermore, at E/n = 20, the sputter rate and secondary ion intensity decreased rapidly. Because the Young’s modulus increased significantly with low roughness, the result suggested that the system reached gel point rapidly. In other words, radical induced cross-link dominated the damage process. With the auxiliary O2+ as cosputter ion, it helps to break the ion-beam induced morphology and enhance the ionization yield that masked damage. As a result, steady-state can be obtained and depth profiles with less artifact were observed.