Nickel und Nickel (II) -oxid werden in großem Umfang in fortgeschrittenen elektronischen Geräten verwendet. In der Mikroelektronik-Industrie wird Nickel verwendet werden, um Nickelsilizid bilden. Die Nickelmono Silizid (NiSi) wurde als ausgezeichnetes Material für Source-Drain-Kontaktanwendungen unter 45 nm-CMOS-Technologie entwickelt. Im Vergleich zu anderen Siliziden für die Kontaktanwendungen verwendet wird NiSi wegen seines niedrigen spezifischen Widerstand, niedrigen Kontaktwiderstand, relativ niedrigen Bildungstemperatur und niedrigem Siliziumverbrauchs bevorzugt. Nickel in Nickelbasis-Akkus und ferromagnetischen Direktzugriffsspeicher (RAMs) verwendet. Nickel (II) oxid wird als Transistor-Gate-Oxid und Oxid in resistive RAM genutzt wird. Atomic Layer Deposition (ALD) ist eine spezielle Art der Chemical Vapor Deposition (CVD), das verwendet wird, um sehr glatte sowie homogene Dünnfilme mit hervorragenden Treue auch bei hohen Seitenverhältnissen abzuscheiden. Es basiert auf selbstabschließenden sequentielle Gas-Feststoff-Reaktionen, die eine präzise Steuerung der Filmdicke auf wenige Angström lassen sich auf der Basis. Zur Herstellung der heutigen 3D-elektronische Geräte, sind Technologien wie ALD erforderlich. Trotz der Vielzahl von praktischen Anwendungen von Nickel und Nickel (II) -oxid, sind einige Nickelvorstufen zur thermischen basierend ALD erhältlich. Darüber hinaus haben diese Vorstufen bei schlechten Filmeigenschaften führte und die Prozesseigenschaften wurden ebenfalls begrenzt. Daher in dieser Masterarbeit mussten die Eigenschaften verschiedener neuartiger Nickelvorstufen zu bewerten. Alle neuen Vorstufen heteroleptische (verschiedene Arten von Liganden) und Komplexe wurden vom Hersteller speziell zur thermischen basierend ALD aus reinem Nickel mit H 2 als ein Co-Reaktionsmittel gestaltet. Um die neuartige Vorläufer zu untersuchen, wurde eine neue Methode entwickelt, um kleine Mengen in einer sehr zeitsparend (bis zu 2 g) von Ausgangsstoffen zu testen. Diese Methodologie beinhaltet: TGA / DTA-Kurve analysiert der Vorstufen, thermische Stabilitätstests in dem die Vorläufer (<0,1 g) wurden bei erhöhter Temperatur in einer abgedichteten Umgebung für mehrere Stunden wurde die Abscheidung Experimenten und Film Charakterisierungen erhitzt. Die Abscheidungen wurden mit Hilfe der in situ Quarzmikrowaage überwacht, während die anwendungsbezogenen Filmeigenschaften, wie chemische Zusammensetzung, physikalische Phase, Dicke, Dichte, Härte und Schichtwiderstand wurden mit Hilfe von ex situ Messverfahren untersucht. Vor der Evaluierung neuartiger Nickelvorstufen ein Benchmark ALD-Prozess war vom Referenznickelvorläufer (Ni (AMD)) und Luft als Reaktionspartner entwickelt. Das Hauptziel der Entwicklung und Optimierung von solchen Benchmark-ALD-Prozess war es, Standard-Prozessparameter wie zweite Reaktionspartner Belichtungszeiten, Argonspülung Zeiten, gesamtprozessdruck, beginnend Abscheidungstemperatur und Gasströme zu extrahieren. Diese Standard-Prozessparameter mussten verwendet, um die Prozessentwicklung Aufgabe (das spart Vorläufer Verbrauch) zu verkürzen und die Sublimationstemperatur Optimierung für jede neuartige Vorstufe werden. Die ALD Verhalten wurde in Bezug auf die Wachstumsrate durch Variation des Nickelvorläuferbelichtungszeit, Vorläufer Temperatur und Niederschlagstemperatur überprüft.:Lists of Abbreviations and Symbols VIII Lists of Figures and Tables XIV 1 Introduction 1 I Theoretical Part 3 2 Nickel and Nickel Oxides 4 2.1 Introduction and Existence 5 2.2 Material properties of Nickel and Nickel Oxide 5 2.3 Application in electronic industry 5 3 Atomic Layer Deposition 7 3.1 History 8 3.2 Definition 8 3.3 Features of thermal-ALD 8 3.3.1 ALD growth mechanism – an ideal view 8 3.3.2 ALD growth behaviour 10 3.3.3 Growth mode 11 3.3.4 ALD temperature window 11 3.4 Benefits and limitations 12 3.5 Precursor properties for thermal-ALD 13 3.6 ALD & CVD of Nickel – A literature survey 13 4 Metrology 17 4.1 Thermal analysis of precursors 18 4.2 Film and growth characterization 21 4.2.1 Quartz Crystal Microbalance 21 4.2.2 Spectroscopic Ellipsometry 24 4.2.3 X-Ray Photoelectron Spectroscopy 28 4.2.4 Scanning Electron Microscopy 29 4.2.5 X-Ray Reflectometry and X-Ray Diffraction 29 4.2.6 Four Point Probe Technique 20 5 Rapid Thermal Processing 32 5.1 Introduction 33 5.2 Basics of RTP 33 5.3 Nickel Silicides-A literature survey 33 II Experimental Part 36 6 Methodologies 37 6.1 Experimental setup 38 6.2 ALD process 41 6.2.1 ALD process types and substrate setups 41 6.2.2 Process parameters 41 6.3 Experimental procedure 42 6.3.1 Tool preparation 42 6.3.2 Thermal analysis and ALD experiments from nickel precursors 43 6.3.3 Data acquisition and evaluation 44 6.3.4 Characterization of film properties 46 7 Results and discussion 48 7.1 Introduction 49 7.2 QCM verification with Aluminum Oxide ALD process 49 7.3 ALD process from the reference precursor 50 7.3.1 Introduction 50 7.3.2 TG analysis for Ni(amd) precursor 51 7.3.3 Thermal stability test for Ni(amd) 51 7.3.4 ALD process optimization 52 7.3.5 Film properties 54 7.4 Evaluating the novel Nickel precursors 55 7.4.1 Screening tests for precursor P1 55 7.4.2 Screening tests for precursor P2 62 7.4.3 Screening tests for precursor P3 66 7.4.4 Screening tests for precursor P4 70 7.4.5 Screening tests for precursor P5 72 7.5 Comparison of all nickel precursors used in this work 74 8 Conclusions and outlook 77 References 83 III Appendix 101 A Deposition temperature control & Ellipsometry model 102 B Gas flow plan 105 / Nickel and nickel(II) oxide are widely used in advanced electronic devices . In microelectronic industry, nickel is used to form nickel silicide. The nickel mono-silicide (NiSi) has emerged as an excellent material of choice for source-drain contact applications below 45 nm node CMOS technology. As compared to other silicides used for the contact applications, NiSi is preferred because of its low resistivity, low contact resistance, relatively low formation temperature and low silicon consumption. Nickel is used in nickel-based rechargeable batteries and ferromagnetic random access memories (RAMs). Nickel(II) oxide is utilized as transistor gate-oxide and oxide in resistive RAMs. Atomic Layer Deposition (ALD) is a special type of Chemical Vapor Deposition (CVD) technique, that is used to deposit very smooth as well as homogeneous thin films with excellent conformality even at high aspect ratios. It is based on self-terminating sequential gas-solid reactions that allow a precise control of film thickness down to few Angstroms. In order to fabricate todays 3D electronic devices, technologies like ALD are required. In spite of huge number of practical applications of nickel and nickel(II) oxide, a few nickel precursors are available for thermal based ALD. Moreover, these precursors have resulted in poor film qualities and the process properties were also limited. Therefore in this master thesis, the properties of various novel nickel precursors had to be evaluated. All novel precursors are heteroleptic (different types of ligands) complexes and were specially designed by the manufacturer for thermal based ALD of pure nickel with H 2 as a co-reactant. In order to evaluate the novel precursors, a new methodology was designed to test small amounts (down to 2 g) of precursors in a very time efficient way. This methodology includes: TGA/DTA curve analyses of the precursors, thermal stability tests in which the precursors (< 0.1 g) were heated at elevated temperatures in a sealed environment for several hours, deposition experiments, and film characterizations. The depositions were monitored with the help of in situ quartz crystal microbalance, while application related film properties like chemical composition, physical phase, thickness, density, roughness and sheet resistance were investigated with the help of ex situ measurement techniques. Prior to the evaluation of novel nickel precursors, a benchmark ALD process was developed from the reference nickel precursor (Ni(amd)) and air as a co-reactant. The main goal of developing and optimizing such benchmark ALD process was to extract standard process parameters like second-reactant exposure times, Argon purge times, total process pressure, starting deposition temperature and gas flows. These standard process parameters had to be utilized to shorten the process development task (thus saving precursor consumption) and optimize the sublimation temperature for each novel precursor. The ALD behaviour was checked in terms of growth rate by varying the nickel precursor exposure time, precursor temperature and deposition temperature.:Lists of Abbreviations and Symbols VIII Lists of Figures and Tables XIV 1 Introduction 1 I Theoretical Part 3 2 Nickel and Nickel Oxides 4 2.1 Introduction and Existence 5 2.2 Material properties of Nickel and Nickel Oxide 5 2.3 Application in electronic industry 5 3 Atomic Layer Deposition 7 3.1 History 8 3.2 Definition 8 3.3 Features of thermal-ALD 8 3.3.1 ALD growth mechanism – an ideal view 8 3.3.2 ALD growth behaviour 10 3.3.3 Growth mode 11 3.3.4 ALD temperature window 11 3.4 Benefits and limitations 12 3.5 Precursor properties for thermal-ALD 13 3.6 ALD & CVD of Nickel – A literature survey 13 4 Metrology 17 4.1 Thermal analysis of precursors 18 4.2 Film and growth characterization 21 4.2.1 Quartz Crystal Microbalance 21 4.2.2 Spectroscopic Ellipsometry 24 4.2.3 X-Ray Photoelectron Spectroscopy 28 4.2.4 Scanning Electron Microscopy 29 4.2.5 X-Ray Reflectometry and X-Ray Diffraction 29 4.2.6 Four Point Probe Technique 20 5 Rapid Thermal Processing 32 5.1 Introduction 33 5.2 Basics of RTP 33 5.3 Nickel Silicides-A literature survey 33 II Experimental Part 36 6 Methodologies 37 6.1 Experimental setup 38 6.2 ALD process 41 6.2.1 ALD process types and substrate setups 41 6.2.2 Process parameters 41 6.3 Experimental procedure 42 6.3.1 Tool preparation 42 6.3.2 Thermal analysis and ALD experiments from nickel precursors 43 6.3.3 Data acquisition and evaluation 44 6.3.4 Characterization of film properties 46 7 Results and discussion 48 7.1 Introduction 49 7.2 QCM verification with Aluminum Oxide ALD process 49 7.3 ALD process from the reference precursor 50 7.3.1 Introduction 50 7.3.2 TG analysis for Ni(amd) precursor 51 7.3.3 Thermal stability test for Ni(amd) 51 7.3.4 ALD process optimization 52 7.3.5 Film properties 54 7.4 Evaluating the novel Nickel precursors 55 7.4.1 Screening tests for precursor P1 55 7.4.2 Screening tests for precursor P2 62 7.4.3 Screening tests for precursor P3 66 7.4.4 Screening tests for precursor P4 70 7.4.5 Screening tests for precursor P5 72 7.5 Comparison of all nickel precursors used in this work 74 8 Conclusions and outlook 77 References 83 III Appendix 101 A Deposition temperature control & Ellipsometry model 102 B Gas flow plan 105

Two computational fluid dynamics models are developed to predict the performance of a solid oxide fuel cell stack, a detailed and a simplified model. In the detailed model, the three dimensional momentum, heat, and species transport equations are coupled with electrochemistry. In the simplified model, the diffusion terms in the transport equations are selectively replaced by rate terms within the core region of the stack. This allows much coarser meshes to be employed at a fraction of the computational cost. Following the mathematical description of the problem, results for single-cell and multi-cell stacks are presented. Comparisons of local current density, temperature, and cell voltage indicate that good agreement is obtained between the detailed and simplified models, verifying the latter as a practical option in stack design. Then, the simplified model is used to determine the effects of utilization on the electrochemical performance and temperature distributions of a 10 cell stack. The results are presented in terms of fluid flow, pressure, species mass fraction, temperature, voltage and current density distributions. The effects of species and flow distributions on electrochemical performance and temperature are then analyzed for a 100 cell stack. The discussion highlights the importance of manifold design on performance and thermal management of large stacks. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2013-09-30 15:55:18.627

Structural colours arise from the interaction of visible light with nano-structured materials. The occurrence of such structures in nature has been known for over a century, but it is only in the last few decades that the study of natural photonic structures has fully matured due to the advances in imagining techniques and computational modelling. Even though a plethora of different colour-producing architectures in a variety of species has been investigated, a few significant questions are still open: how do these structures develop in living organisms? Does disorder play a functional role in biological photonics? If so, is it possible to say that the optical response of natural disordered photonics has been optimised under evolutionary pressure? And, finally, can we exploit the well-adapted photonic design principles that we observe in Nature to fabricate functional materials with optimised scattering response? In my thesis I try to answer the questions above: I microscopically investigate $\textit{in vivo}$ the growth of a cuticular multilayer, one of the most common colour-producing strategies in nature, in the green beetles $\textit{Gastrophysa viridula}$ showing how the interplay between different materials varies during the various life stages of the beetles; I further investigate two types of disordered photonic structures and their biological role, the random array of spherical air inclusions in the eggshells of the honeyguide $\textit{Prodotiscus regulus}$, a species under unique evolutionary pressure to produce blue eggs, and the anisotropic chitinous network of fibres in the white beetle $\textit{Cyphochilus}$, the whitest low-refractive index material; finally, inspired by these natural designs, I fabricate and study light transport in biocompatible highly-scattering materials.

The government has indicated that innovation fosters economic growth and is essential to maintaining a competitive position in international markets. Patents are the preferred mechanism by which the Australian Government and other governments encourage their nationals to protect their innovations. The question of the entitlement was raised in several cases in the Federal Court of Australia where there has been a failure to name all of the inventors on a patent grant (non-joinder) or where persons were mis-named as inventors, who were not and consequently have no interest in a grant (rnis-joinder). In both cases, parties who were not themselves daiming an entitlement to the invention, brought objections based on a number of grounds, including entitlement. The results have been the revocation of the patent in the case on the non-joinder of an inventor and in the case of mis-joinder, the preliminary view of a judge of the Federal Court has been, that the patent would be invalid through lack of entitlement. The result is that competitors are permitted to 'exploit' the invention, as the subject matter is not protected by a patent. The implications are far reaching, For example, where a research team in collaboration with another develops an invention but omits the inventive contribution of even one member of one team or includes a person who has not made an inventive contribution in the patent grant, the patent will be invalid. In these circumstances, the author considers that the result produces a disincentive to innovate. Consideration of this area in other jurisdictions reveals that the U.S. and the U.K. have recognised this as an unsatisfactory state of affairs. As a result, Congress in the U.S. made provision in their Patent Code in the early 1950's, that in the case of error or mistake giving rise to a non-joinder or mis-joinder of inventors, the patent would not be invalid but could be rectified by the Director of Patents and Trade Marks (the 'Director'). In the U.K., the Comptroller has powers to deal with a wide variety of cases involving entitlement to ownership of a patent. The situations include but are not limited to cases where some but not all of the persons entitled to the grant have been granted the patent, i.e. non-joinder, or where a person entitled to be granted a patent, has been granted a patent together with a person who is not entitled, i.e. mis-joinder. The thesis will focus on the non-joinder and mis-joinder of inventors, but the U.K. provision addresses a wider field of parties entitled, whether entitled as inventors or on some other basis. In addition, the U.K. and Germany have made provision restricting the persons who are able to challenge a patent on entitlement grounds. This is restricted to those persons having an interest in the patent, rather than open to any person, as is the case in Australia. The Australian decisions have been determined on historic cases dating back to the 17th century. It is timely to consider amendments which will overcome revocation of patents under Australian law, for what is essentially a matter between the persons interested. These amendments will accordingly encourage innovation, particularly in an environment where intellectual property has taken on greater importance and where the identification of the inventor has become more complex as collaborations in research become more common.

L'intégration de diélectriques High-k dans les empilements de grille des dispositifs MOS a fait naître de nouvelles interrogations concernant la fiabilité des futurs nœuds technologiques. La miniaturisation constante des dispositifs conduisant à l'amincissement des épaisseurs d'oxyde de grille, leur caractérisation électrique est rendue de plus en plus complexe à l'échelle du dispositif. Pour palier à ce problème, l'utilisation d'un microscope à force atomique en mode conducteur sous ultravide permet grâce à la faible surface de contact entre la pointe et l'échantillon de réduire suffisamment le courant tunnel pour pouvoir étudier la dégradation et le claquage diélectrique d'oxyde ultra fin. La comparaison systématique des résultats de fiabilité de l'empilement High-k du nœud 28nm et de la couche interfaciale seule ayant subi les mêmes étapes de développement que celles présentes dans l'empilement, obtenus par C-AFM sous ultra vide, ont permis de montrer expérimentalement que la probabilité de claquage des oxydes de grille High-k est gouvernée par la fiabilité propre des couches qui la composent, et de déduire une loi d'extrapolation de la durée de vie en tension et en surface ce qui permet de prédire la statistique de défaillance du dispositif. Les impacts d'un pré-stress en tension de l'ordre de la milliseconde sur les distributions de claquage des oxydes de grille simples et bicouches ont été rapportés. Ces résultats sont expliqués dans ce manuscrit par le déclenchement lors de l'application du stress, d'une dégradation au sein de l'oxyde, prenant naissance dans la couche interfaciale des oxydes High-k et conduisant à une réduction locale de l'épaisseur de diélectrique. Des phénomènes de résistance différentielle négative au moment de la rupture diélectrique ont été étudiés et modélisés pour différentes épaisseurs d'oxyde, par une croissance filamentaire de la dégradation. Il a été possible de donner une expression analytique reliant le temps caractéristique de croissance filamentaire et le temps moyen de claquage observé sur les distributions statistiques. Enfin, les mesures C-AFM de ce travail ont été étendues au cas des structures MIM utilisées pour le développement des futurs mémoires résistives OxRAM. Dans ce cas un effet d'auto-guérison à l'échelle nanométrique a été mis en évidence. / Integration of High-k dielectrics in gate oxides of MOS raised new issues concerning the reliability of futur technology nodes. The constant miniaturisation of devices leads to thinner gate oxides, making their electrical caracterisation more complex at the device scale. To solve this problem, an atomic force microscope in conductive mode under ultra high vacuum can be used thanks to the readuce contact area between the tip and the sample which allow a drastic decrease of the tunneling current and thus the study of the degradation and the dielectric breakdown of ultra-thin oxides. The systematic comparaison of the TDDB distributions obtained on the High-k gate oxide of the 28nm technology node on one side and obtained on the Interfacial layer alone revealed that the failure probability of High-k oxides is governed by the failure probability of each layer present in the stack. This allow to give an extrapolation law of the High-k gate oxide lifetime as a function of the applied voltage and the electrode area and to predict the failure statistic of the 28nm tehcnology node. The impact of voltage pre-stress with a microseconde range of duration on the TDDB and VBD distributions of both single layer and High-k gate oxides is given is the manuscript. The results are then interpreted by an invasive degradation nucleating from an interface during a stress and leading to a local thinned oxide. Pre-breakdown negative differential resistance have been studied and modeled for several oxide thickness, using a growing mecanism of the elctrical degradation. An analytic expression linking the growth caracteristic time of the filament and the mean time to breakdown observed on the statistical distributions has then been given. Finally, C-AFM measurements developped in this work has been extended to MIM structures used for oxide resistive random access memories (OxRAM). A self healing has been observed at the nanometric scale for these samples.

Thesis submitted in fulfilment of the requirements for the degree Master of Technology: Electrical Engineering in the Faculty of Engineering at the Cape Peninsula University of Technology, 2012 / The introduction of the IEC 61850 standard in substations for communication networks and systems by the International Electrotechnical Commission (IEC) in 2003 provided the possibility for communication between devices of different manufacturers. However, the advent of this standard also brought about many challenges associated with it. The challenges introduced by this fairly recent standard of communications in Substation Automation Systems (SAS), and the need for the development of cost effective IEC 61850- compliant devices, motivated the decision of the Centre for Substation and Energy Management Systems within the Electrical Engineering Department of the Cape Peninsula University of Technology to focus on the implementation of the IEC 61850 standard using an embedded hardware platform. The development of an IEC 61850 embedded application requires substantial knowledge in multiple domains such as data networking, software modelling and development of Intelligent Electronic Devices (IEDs), protection of the electrical system, system simulation and testing methods, etc. Currently knowledge about the implementation of the IEC 61850 standard usually resides with vendors and is not in the public domain. The IEC 61850 standard allows for two groups of communication services between entities within the substation automation system. One group utilizes a client-server model accommodating services such as Reporting and Remote Switching. The second group utilizes a peer-to-peer model for Generic Substation Event (GSE) services associated with time-critical activities such as fast and reliable communication between Intelligent Electronic Devices (IEDs) used for protection of the power network. The messages associated with the GSE services are the Generic Object Oriented Substation Event (GOOSE) messages. The use of GOOSE messages for protection of the electrical system is very important in modern substations. Detailed knowledge of the structure of these messages is important in instances requiring fault diagnosis to determine the cause of mal– operation or to address interoperability concerns or when developing custom IEC 61850- compliant devices with limited functionality. A practical protection application (overcurrent) case study is presented where GOOSE messages are exchanged between a commercial IED and an IEC 61850-compliant controller based on an embedded platform. The basic data model and software development of an actuator node for a circuit breaker is proposed using an IEC 61850 communication stack on an embedded platform. The performance of the GOOSE messages is confirmed to be as per the functional behaviour specified, and per the IEC 68150 standard in terms of the temporal behaviour required. This thesis document tables the methods, software programs, hardware interfacing and system integration techniques that allow for the development and implementation of a low cost IEC 61850-compliant controller unit on an embedded systems platform for the substation automation system. The overcurrent case study distributed between a commercial IED (SIEMENS Siprotec device) and the actuator application developed on an embedded platform for this project (DK60 board) is in compliance with the IEC 61850 standard and utilizing GOOSE messaging is successfully completed both in terms of functional and temporal behaviour. This novel research work contributes not only to the academic community, but to the international Power Systems community as a whole. Keywords: IEC 61850 standard, IEDs, GOOSE message, software modelling, software development, substation automation systems, communication stack, embedded systems, actuator.

L’analyse spatiale de données géographies connaît un regain d’intérêt dans la communauté des bases de données relationnelles. Plus spécifiquement, les opérations et les fonctions spatiales utilisées comme base de l’analyse spatiale sont implémentées par les grands noms des systèmes de gestion de bases de données relationnelles limitant ainsi l’hétérogénéité structurelle des systèmes. En outre, la littérature est abondante en publications dans le domaine des ontologies spatiales afin de limiter l’hétérogénéité sémantique des sources de données tout en améliorant l’interopérabilité de ces données. Bien que l’interopérabilité des données soit l’un des objectifs du Web Sémantique, tout le potentiel de ces outils et de ces techniques basés sur la connaissance n’a pas été révélé. Avec l’influence sans cesse croissante du Web Sémantique à travers ces outils et applications en gestion de la connaissance et système intelligent, les applications utilisant des données géospatiales suivent ce phénomène en bénéficiant de son influence. Cette thèse se focalise sur l’utilisation de la connaissance métier afin de gérer des données spatiales à l’aide des technologies du Web sémantique. L’activité de recherche menée dans le cadre de cette thèse est réalisée sur des données provenant du domaine de l’archéologie industrielle. Cet environnement se caractérise par son hétérogénéité et sa grande quantité de données offrant ainsi un cadre idéal pour la réalisation d’un outil de gestion de connaissance. Cet outil basé sur les technologies du Web Sémantique a été prototypé sous le nom d’ArchaeoKM suivant le principe des 4 K, Knowledge Acquisition, Knowledge Management, Knowledge Visualization and Knowledge Analysis. Ce même principe est mis en œuvre pour les données spatiales. Une ontologie de haut niveau a été développée pour servir de cadre applicatif à la gestion des données spatiales permettant d’ajuster une ontologie de domaines sans composante spatiale. Le processus de gestion de la connaissance commence avec l’acquisition de la signature spatiale des objets identifiés. Cette signature est stockée dans un système de gestion de bases de données spatiales et est référencée par l’objet correspondant dans la base de connaissance. La connaissance spatiale de ces objets est générée à l’aide des fonctions et des opérations spatiales au niveau de la base de données spatiale et l’enrichissement de la base de connaissance est réalisé avec le résultat de ces opérations et fonctions. L’inférence de nouvelle connaissance sur la base des données existante est réalisée à l’aide de SWRL (Semantic Web Rule Language). De plus, ce langage a été étendu à l’aide de nouveaux built-ins spatiaux afin de prendre en sidération la dimension spatiale des données. De même, cette dimension spatiale a été apportée au langage SPARQL afin de réaliser des requêtes spatiales sur la base de connaissances.En effet, l’objectif principal de cette thèse est d’initier le premier pas vers l’intégration des composantes spatiales avec les technologies du Web Sémantique. Le processus d’intégration est premier plan pour les deux technologies. D’un point de vue Web Sémantique, l’intégration de données non communes dans ce cadre applicatif ouvre la porte à l’intégration de données beaucoup plus large. D’un point de vue des systèmes d’information géographique, l’inclusion de la connaissance permet une gestion métier des données rendant l’analyse plus proche de l’interprétation humaine. / Spatial technology has gained momentum under database systems. More specifically, the spatial operations and spatial functions are used to carry out spatial analysis which can be executed through these database systems. In addition, there has been significant amount of research in the field of the geospatial ontology domain in order to achieve the semantic interoperability between different data sources. Although, data interoperability is one of the main objectives of the Semantic Web technologies, the potentiality of the underlying knowledge tools and techniques have not been completely identified. With the growing influence of the Semantic Web technologies towards the application based on knowledge management and intelligent systems, the geospatial application benefits from this influence. This thesis emphasizes on the use of knowledge to manage spatial data within spatial information systems through the Semantic Web framework. This research activity is carried out with the backdrop of the case study of the industrial archaeology. It sets up an ideal environment for the application of knowledge to manage the huge and heterogeneous dataset. The use of knowledge to manage the diversity of information was well executed through the application prototype named ArchaeoKM which is based on the Semantic Web. The ArchaeoKM framework follows the 4Ks processing steps: Knowledge Acquisition, Knowledge Management, Knowledge Visualization and Knowledge Analysis. The same processing principle of 4Ks was implemented during the spatial knowledge processing. A top level ontology was developed in order to serve as the background representation of the case study in order to adjust the spatial components. Keeping the custom, the spatial knowledge processing begins with acquiring spatial signatures of the identified objects. The spatial signatures are stored within the spatial database system with proper mapping to the objects in the knowledge base. The spatial knowledge of these objects is managed through executing the spatial functions at the database level and enriching the knowledge base with the results. This spatially enriched knowledge base is used again to analyze the spatial knowledge. This research thesis benefits from Semantic Web Rule Language in order to infer knowledge. In addition, the spatial built-ins proposed during the course add up spatial dimension to the SWRL for spatial inferences. Similarly, a spatial extension of the query language SPARQL is proposed in order to query spatial knowledge from the knowledge base. Actually, this research thesis provides the initial steps in integrating spatial components within the Semantic Web framework. This integration process is important for both technologies. Regarding the Semantic Web, the integration of non-typical semantic information within this framework opens up doors to other data pattern making the transformation of technologies easier. Likewise, geospatial technologies and GIS systems benefits through the inclusion of knowledge in the analysis process making the analysis much closer and efficient to human interpretation.

A prototype 3-dimensional (3D) anode, based on multiwall carbon nanotubes (MWCNTs), for Li-ion batteries (LIBs), with potential use in Electric Vehicles (EVs) was investigated. The unique 3D design of the anode allowed much higher areal mass density of MWCNTs as active materials, resulting in more amount of Li+ ion intake, compared to that of a conventional 2D counterpart. Furthermore, 3D amorphous Si/MWCNTs hybrid structure offered enhancement in electrochemical response (specific capacity 549 mAhg-1). Also, an anode stack was fabricated to further increase the areal or volumetric mass density of MWCNTs. An areal mass density of the anode stack 34.9 mg/cm2 was attained, which is 1,342% higher than the value for a single layer 2.6 mg/cm2. Furthermore, the binder-assisted and hot-pressed anode stack yielded the average reversible, stable gravimetric and volumetric specific capacities of 213 mAhg-1 and 265 mAh/cm3, respectively (at 0.5C). Moreover, a large-scale patterned novel flexible 3D MWCNTs-graphene-polyethylene terephthalate (PET) anode structure was prepared. It generated a reversible specific capacity of 153 mAhg-1 at 0.17C and cycling stability of 130 mAhg-1 up to 50 cycles at 1.7C.

The work in this thesis is separated into two parts. The first part deals with the load planning and sequencing problem for double-stack intermodal railcars, an operational problem found at many rail container terminals. In this problem, containers must be assigned to a platform on which the container will be loaded, and the loading order must be determined. These decisions are made with the objective of minimizing the costs associated with handling the containers, as well as minimizing the cost of containers left behind. The deterministic version of the problem can be cast as a shortest path problem on an ordered graph. This problem is challenging to solve because of the large size of the graph. We propose a two-stage heuristic based on the Iterative Deepening A* algorithm to compute solutions to the load planning and sequencing problem within a five-minute time budget. Next, we also illustrate how a Deep Q-learning algorithm can be used to heuristically solve the same problem.The second part of this thesis considers sequential models in deep learning. A recent strategy to circumvent the exploding and vanishing gradient problem in recurrent neural networks (RNNs) is to enforce recurrent weight matrices to be orthogonal or unitary. While this ensures stable dynamics during training, it comes at the cost of reduced expressivity due to the limited variety of orthogonal transformations. We propose a parameterization of RNNs, based on the Schur decomposition, that mitigates the exploding and vanishing gradient problem, while allowing for non-orthogonal recurrent weight matrices in the model. / Le travail de cette thèse est divisé en deux parties. La première partie traite du problème de planification et de séquencement des chargements de conteneurs sur des wagons, un problème opérationnel rencontré dans de nombreux terminaux ferroviaires intermodaux. Dans ce problème, les conteneurs doivent être affectés à une plate-forme sur laquelle un ou deux conteneurs seront chargés et l'ordre de chargement doit être déterminé. Ces décisions sont prises dans le but de minimiser les coûts associés à la manutention des conteneurs, ainsi que de minimiser le coût des conteneurs non chargés. La version déterministe du problème peut être formulé comme un problème de plus court chemin sur un graphe ordonné. Ce problème est difficile à résoudre en raison de la grande taille du graphe. Nous proposons une heuristique en deux étapes basée sur l'algorithme Iterative Deepening A* pour calculer des solutions au problème de planification et de séquencement de la charge dans un budget de cinq minutes. Ensuite, nous illustrons également comment un algorithme d'apprentissage Deep Q peut être utilisé pour résoudre heuristiquement le même problème. La deuxième partie de cette thèse examine les modèles séquentiels en apprentissage profond. Une stratégie récente pour contourner le problème de gradient qui explose et disparaît dans les réseaux de neurones récurrents (RNN) consiste à imposer des matrices de poids récurrentes orthogonales ou unitaires. Bien que cela assure une dynamique stable pendant l'entraînement, cela se fait au prix d'une expressivité réduite en raison de la variété limitée des transformations orthogonales. Nous proposons une paramétrisation des RNN, basée sur la décomposition de Schur, qui atténue les problèmes de gradient, tout en permettant des matrices de poids récurrentes non orthogonales dans le modèle.

Seismic velocities are an important prerequisite for seismic processing as a method for hydrocarbons accumulations detection. Seismic velocities are often displayed for mutual comparing, improvement checking, they are filtrated and recalculated for its different characteristic determination. This work deals with basic seismic propagation laws, the meaning of velocities in different stages of seismic processing and this theoretical background is followed by a proposition of method for calculating and displaying of stack velocities.