Please read the abstract in the section 00front of this document / Thesis (PhD)--University of Pretoria, 2008. / Materials Science and Metallurgical Engineering / PhD / Unrestricted

Les complexes organométalliques de métaux alcalino-terreux lourds (Ca, Sr, Ba), ont émergé ces deux dernières décennies en tant que précatalyseurs efficaces, biocompatibles et disponibles à bas coûts pour des réactions d'hydroélémention d'oléfine ou de polymérisation de cycloesters. Cette thèse décrit la synthèse de complexes de métaux du groupe 2 (Ca, Sr, Ba) et leur utilisation en catalyse de couplage déshydrogénant N-H/H-Si. Le précatalyseur Ba[CH(SiMe3)2]2(THF)3 s'est montré le plus efficace pour le couplage d'amines avec des hydrosilanes, présentant une activité catalytique parmi les plus élevées à ce jour (TOF jusqu' 3600 h-1 ; TON jusqu'à 396). La combinaison de résultats théoriques (calculs DFT en collaboration avec le Dr. Sven Tobisch) et expérimentaux ont permis de comprendre les mécanismes opératoires de ces réactions. Le développement de procédés de synthèse pour la production d'oligo- ou polysilazanes a été mené par l'intermédiaire de ces couplages déshydrogénants. Des structures macromoléculaires linéaires ou cycliques de type polycarbosilazane, présentant de nombreuses applications en chimie des matériaux, ont notamment été synthétisées par polycondensation catalysée par Ba[CH(SiMe3)2]2(THF)3. Dans un dernier temps, la synthèse de complexes innovants de métaux des groupes 12 (Zn, Cd, Hg) et 14 (Pb) a été effectuée en collaboration avec l'équipe du Pr. Silvestru de l'université de Babeş-Bolyai (Cluj-Napoca, Roumanie). / In the past two decades, heavier alkaline-earth (Ca, Sr, Ba) complexes have emerged as highly efficient and biocompatible precatalyst for hydroelementation or polymerisation reactions. This PhD thesis describes the synthesis and characterisation of heavier alkaline-earth complexes and their applications as precatalyst for cross-dehydrocoupling of hydrosilanes and amines. The homoleptic precatalyst Ba[CH(SiMe₃)₂]₂(THF)3 displayed high catalytic activity (TOF up to 3600 h⁻¹ ; TON up to 396), with excellent chemoselectivity in reaction of (di)amines with (di)hydrosilanes. Combination of experimental and DFT investigations (collaboration with Dr. Sven Tobisch) revealed the reactions proceed by nucleophilic attack of a metal amide at the incoming silane and subsequent turnover-limiting hydrogen transfer to the metal center. Development of synthesis of oligo- and polysilazanes was performed using our best barium precatalysts. Cyclic or linear polycarbosilazanes (Mw up to 10 000 g.mol⁻¹) were also produced by fast and controlled barium-catalyzed N-H/H-Si polycondensation. In a collaboration with Prof. Silvestru from Babeş-Bolyai University (Cluj-Napoca, Roumanie), synthesis of interesting zinc, cadmium, mercury and lead complexes were achieved in complement of this work.

L’avènement de l’ère digitale a changé la façon dont les individus communiquent à travers le monde, et a amené de nouvelles problématiques en terme de vie privée. La notion d’anonymat la plus répandue pour les communications sur Internet consiste à empêcher tout acteur du réseau de connaître à la fois l’expéditeur d’un message et son destinataire. Bien que ce niveau de protection soit adéquat pour l’utilisateur d’Internet moyen, il est insuffisant lorsqu’un individu peut être condamné pour le simple envoi de documents à une tierce partie. C’est le cas en particulier des lanceurs d’alerte, prenant des risques personnels pour informer le public de pratiques illégales ou antidémocratiques menées par de grandes organisations. Dans cette thèse, nous envisageons un niveau d’anonymat plus fort, où l’objectif est de dissimuler le fait même qu’un utilisateur envoie ou reçoive des données. Pour cela, nous délaissons l’architecture client-serveur couramment utilisée dans les réseaux anonymes, en faveur d’une architecture entièrement distribuée et homogène, où chaque utilisateur remplit également le rôle de serveur relai, lui permettant de dissimuler son propre trafic dans celui qu’il relai pour les autres. Dans cette optique, nous proposons un nouveau protocole pour les communications pairs à pairs sur Internet. À l’aide de récents outils de preuves cryptographiques, nous prouvons que ce protocole réalise les propriétés d’anonymat désirées. De plus, nous montrons par une étude pratique que, bien que le protocole induise une grande latence dans les communications, il assure un fort anonymat, même pour des réseaux de petite taille. / With the development of online communications in the past decades, new privacy concerns have emerged. A lot of research effort have been focusing on concealing relationships in Internet communications. However, most works do not prevent particular network actors from learning the original sender or the intended receiver of a communication. While this level of privacy is satisfactory for the common citizen, it is insufficient in contexts where individuals can be convicted for the mere sending of documents to a third party. This is the case for so-called whistle-blowers, who take personal risks to alert the public of anti-democratic or illegal actions performed by large organisations. In this thesis, we consider a stronger notion of anonymity for peer-to-peer communications on the Internet, and aim at concealing the very fact that users take part in communications. To this end, we deviate from the traditional client-server architecture endorsed by most existing anonymous networks, in favor of a homogeneous, fully distributed architecture in which every user also acts as a relay server, allowing it to conceal its own traffic in the traffic it relays for others. In this setting, we design an Internet overlay inspired from previous works, that also proposes new privacy-enhancing mechanisms, such as the use of relationship pseudonyms for managing identities. We formally prove with state-of-the-art cryptographic proof frameworks that this protocol achieves our privacy goals. Furthermore, a practical study of the protocol shows that it introduces high latency in the delivery of messages, but ensures a high anonymity level even for networks of small size.

Atomically thin, semiconducting transition metal dichalogenides (TMDs), a special class of layered semiconductors, that can be shaped as a perfect two dimensional material, have garnered a lot of attention owing to their fascinating electronic properties which are achievable at the extreme nanoscale. In contrast to graphene, the most celebrated two-dimensional (2D) material thus far; TMDs exhibit a direct band gap in the monolayer regime. The presence of a non-zero bandgap along with the broken inversion symmetry in the monolayer limit brands semiconducting TMDs as the perfect candidate for future optoelectronic and valleytronics-based device application. These remarkable discoveries demand exploration of different materials that possess similar properties alike TMDs. Recently, III-VI layered semiconducting materials (example: InSe, GaSe etc.) have also emerged as potential materials for optical device based applications as, similar to TMDs, they can be shaped into a perfect two-dimensional form as well as possess a sizable band gap in their nano-regime. The perfect 2D character in layered materials cause enhancement of strong Coulomb interaction. As a result, excitons, a coulomb bound quasiparticle made of electron-hole pair, dominate the optical properties near the bandgap. The basis of development for future optoelectronic-based devices requires accurate characterization of the essential properties of excitons. Two fundamental parameters that characterize the quantum dynamics of excitons are: a) the dephasing rate, 𝛾, which represents the coherence loss due to the interaction of the excitons with their environment (for example- phonons, impurities, other excitons, etc.) and b) excited state population decay rate arising from radiative and non-radiative relaxation processes. The dephasing rate is representative of the time scale over which excitons can be coherently manipulated, therefore accurately probing the source of exciton decoherence is crucial for understanding the basic unexplored science as well as creating technological developments. The dephasing dynamics in semiconductors typically occur in the picosecond to femtosecond timescale, thus the use of ultrafast laser spectroscopy is a potential route to probe such excitonic responses. The focus of this dissertation is two-fold: firstly, to develop the necessary instrumentation to accurately probe the aforementioned parameters and secondly, to explore the quantum dynamics and the underlying many-body interactions in different layered semiconducting materials. A custom-built multidimensional optical non-linear spectrometer was developed in order to perform two-dimensional spectroscopic (2DFT) measurements. The advantages of this technique are multifaceted compared to regular one-dimensional and non-linear incoherent techniques. 2DFT technique is based on an enhanced version of Four wave mixing experiments. This powerful tool is capable of identifying the resonant coupling, probing the coherent pathways, unambiguously extracting the homogeneous linewidth in the presence of inhomogeneity and decomposing a complex spectra into real and imaginary parts. It is not possible to uncover such crucial features by employing one dimensional non-linear technique. Monolayers as well as bulk TMDs and group III-VI bulk layered materials are explored in this dissertation. The exciton quantum dynamics is explored with three pulse four-wave mixing whereas the phase sensitive measurements are obtained by employing two-dimensional Fourier transform spectroscopy. Temperature and excitation density dependent 2DFT experiments unfold the information associated with the many-body interactions in the layered semiconducting samples.

The representer theorem from the reproducing kernel Hilbert space theory is the origin of many kernel-based machine learning and signal modelling techniques that are popular today. Most kernel functions used in practical applications behave in a homogeneous manner across the domain of the signal of interest, and they are called stationary kernels. One open problem in the literature is the specification of a non-stationary kernel that is computationally tractable. Some recent works solve large-scale optimization problems to obtain such kernels, and they often suffer from non-identifiability issues in their optimization problem formulation. Many practical problems can benefit from using application-specific prior knowledge on the signal of interest. For example, if one can adequately encode the prior assumption that edge contours are smooth, one does not need to learn a finite-dimensional dictionary from a database of sampled image patches that each contains a circular object in order to up-convert images that contain circular edges. In the first portion of this thesis, we present a novel method for constructing non-stationary kernels that incorporates prior knowledge. A theorem is presented that ensures the result of this construction yields a symmetric and positive-definite kernel function. This construction does not require one to solve any non-identifiable optimization problems. It does require one to manually design some portions of the kernel while deferring the specification of the remaining portions to when an observation of the signal is available. In this sense, the resultant kernel is adaptive to the data observed. We give two examples of this construction technique via the grayscale image up-conversion task where we chose to incorporate the prior assumption that edge contours are smooth. Both examples use a novel local analysis algorithm that summarizes the p-most dominant directions for a given grayscale image patch. The non-stationary properties of these two types of kernels are empirically demonstrated on the Kodak image database that is popular within the image processing research community. Tensors and tensor decomposition methods are gaining popularity in the signal processing and machine learning literature, and most of the recently proposed tensor decomposition methods are based on the tensor power and alternating least-squares algorithms, which were both originally devised over a decade ago. The algebraic approach for the canonical polyadic (CP) symmetric tensor decomposition problem is an exception. This approach exploits the bijective relationship between symmetric tensors and homogeneous polynomials. The solution of a CP symmetric tensor decomposition problem is a set of p rank-one tensors, where p is fixed. In this thesis, we refer to such a set of tensors as a rank-one decomposition with cardinality p. Existing works show that the CP symmetric tensor decomposition problem is non-unique in the general case, so there is no bijective mapping between a rank-one decomposition and a symmetric tensor. However, a proposition in this thesis shows that a particular space of rank-one decompositions, SE, is isomorphic to a space of moment matrices that are called quasi-Hankel matrices in the literature. Optimization over Riemannian manifolds is an area of optimization literature that is also gaining popularity within the signal processing and machine learning community. Under some settings, one can formulate optimization problems over differentiable manifolds where each point is an equivalence class. Such manifolds are called quotient manifolds. This type of formulation can reduce or eliminate some of the sources of non-identifiability issues for certain optimization problems. An example is the learning of a basis for a subspace by formulating the solution space as a type of quotient manifold called the Grassmann manifold, while the conventional formulation is to optimize over a space of full column rank matrices. The second portion of this thesis is about the development of a general-purpose numerical optimization framework over SE. A general-purpose numerical optimizer can solve different approximations or regularized versions of the CP decomposition problem, and they can be applied to tensor-related applications that do not use a tensor decomposition formulation. The proposed optimizer uses many concepts from the Riemannian optimization literature. We present a novel formulation of SE as an embedded differentiable submanifold of the space of real-valued matrices with full column rank, and as a quotient manifold. Riemannian manifold structures and tangent space projectors are derived as well. The CP symmetric tensor decomposition problem is used to empirically demonstrate that the proposed scheme is indeed a numerical optimization framework over SE. Future investigations will concentrate on extending the proposed optimization framework to handle decompositions that correspond to non-symmetric tensors.

Orientadores: Caio Jose Colleti Negreiros, Nir Cohen / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-14T14:44:13Z (GMT). No. of bitstreams: 1 Silva_NeitonPereirada_D.pdf: 4231710 bytes, checksum: af4dc57e0a7215547662f87d1744bb27 (MD5) Previous issue date: 2009 / Resumo: Neste trabalho encontramos todas as métricas de Einstein invariantes em quatro famílias de variedades bandeira do tipo B1 e C1. Os nossos resultados são consistentes com a conjectura de Wang e Ziller sobre a finitude das métricas de Einstein. O nosso método para resolver as equações de Einstein e baseado nas simetrias do sistema algébrico. Obtemos os sistemas algébricos de Einstein para variedades bandeira generalizadas do tipo B1 C1e G2. Estes sistemas são as condições necessárias e suficientes para métricas invariantes nessas variedades serem Einstein. Os sistemas algébricos que obtivemos generalizam as equações de Einstein obtidas por Sakane nos casos maximais. As equações nos casos Al e Dl foram obtidas por Arvanitoyeorgos. Calculamos o conjunto das trazes para as variedades bandeira generalizadas dos grupos de Lie clássicos. Assim estendemos à essas variedades certos resultados sobre estruturas Hermitianas invariantes obtidos por San Martin, Cohen e Negreiros. / Abstract: In this work we and all the invariant Einstein metrics on four families of ag manifolds of type Bl and Cl. Our results are consistent with the finiteness conjecture of Einstein metrics proposed by Wang and Ziller. Our approach for solving the Einstein equations is based on the symmetries of the algebraic system. We obtain the Einstein algebraic systems for the generalized ag manifolds of type Bl, Cl and G2. These systems are necessary and sufficient conditions for invariant metrics on these manifolds to be Einstein. The algebraic systems that we obtained generalize the Einstein equations obtained by Sakane in the maximal cases. The equations in the cases Al and Dl were obtained by Arvanitoyeorgos. We calculate all the t-roots on the generalized ag manifolds of the classical Lie groups. Thus we extend to these manifolds certain results on invariant structures Hermitian obtained by San Martin, Cohen and Negreiros. / Doutorado / Geometria Diferencial / Doutor em Matemática

La catalyse constitue un élément clé en synthèse chimique et la recherche actuelle tend à rendre les procédés catalytiques plus propres dans le contexte de la chimie verte. Dans cet esprit, cette thèse a impliqué la recherche de catalyseurs nanoparticulaires utilisés en milieu aqueux, sans ligand toxique et en très faible quantité. La synthèse des nanoparticules (NPs) catalytiques a utilisé des ions des éléments de transition de la droite du tableau périodique et des réducteurs capables de réduire rapidement ces cations en atomes de degré d’oxydation nul s’agrégeant en petites NPs métalliques très actives en catalyse. Les réducteurs choisis ont été des composés réservoirs d’électron organique (naphthyl sodium) ou organométalliques (complexes sandwichs à 19 électrons de valence du fer tel que [Fe(I)Cp*(ŋ6-C6Me6)] ou du cobalt tel que [Co(II)Cp*2], (Cp* = ŋ5-C5Me5)). Les supports limitant l’agrégation des NPs métalliques ont été le solvant (polyéthylène glycol, 1ère partie de la thèse), les cations des réservoirs d’électron organométalliques (2ème partie de la thèse) ou un réseau zéolitique imidazolate (MOF de type ZIF-8, 3ème partie de la thèse). Au lieu d’un cation métallique, il a aussi été possible d’utiliser un cluster tel que [Au25(SR)18] (R = CH2CH2Ph) comme précurseur, auquel cas la réduction peut se limiter à un simple transfer d’électron produisant un cluster anionique stabilisé par le contre-cation sandwich encombré du réservoir d’électron. Les petites NPs ainsi stabilisées se sont avérées d’excellents catalyseurs “verts” de plusieurs réactions de couplage C-C ou C-N et de production d’hydrogène par hydrolyse d’hydrures métalliques en milieu aqueux dans des conditions très douces. Cette dernière réaction a été efficacement catalysée par des NPs bimétalliques Ni2Pt NP@ZIF-8 avec une synergie spectaculaire entre les deux métaux. / Catalysis is a key element in chemical synthesis, and current research is focusing on making catalytic processes cleaner in the context of green chemistry. In this spirit, this thesis involves the research of nanoparticle (NP) catalysts used in aqueous medium, without toxic ligand and in very small quantities toward a variety of useful processes. The synthesis of the catalytic NPs used cations of the transition elements of the right of the periodic table and of reducing agents capable of rapidly reducing these cations to atoms of zero oxidation state aggregating into small catalytically active metal NPs. The chosen reducing agents were organic (naphthyl sodium) or organometallic (19-electron) sandwich complexes of iron such as [Fe(I)Cp*(ŋ6-C6Me6)] or cobalt such as [Co(II)Cp*2], (Cp* = ŋ5-C5Me5)) used as electron reservoirs. The supports limiting the aggregation of the metal NPs were the solvent (polyethylene glycol, first part of the thesis), the cations of the organometallic electron reservoirs (2nd part of the thesis) or a zeolitic imidazolate framework (MOF of ZIF-8 type, 3rd part of the thesis). Instead of a metal cation, it has also been possible to use a cluster such as [Au25(SR) 18] (R = CH2CH2Ph) as a precursor, in which case the reduction was limited to a simple electron transfer producing an anionic cluster stabilized by the congested sandwich counter cation of the electron reservoir. The small NPs thus stabilized proved to be excellent "green" catalysts for several C-C or C-N reactions and hydrogen production by hydrolysis of metal hydrides in an aqueous medium under very mild conditions. This latter reaction was efficiently catalyzed by Ni2Pt@ZIF-8 bimetallic NPs with a spectacular synergy between the two metals.

abstract: Modern manufacturing systems are part of a complex supply chain where customer preferences are constantly evolving. The rapidly evolving market demands manufacturing organizations to be increasingly agile and flexible. Medium term capacity planning for manufacturing systems employ queueing network models based on stationary demand assumptions. However, these stationary demand assumptions are not very practical for rapidly evolving supply chains. Nonstationary demand processes provide a reasonable framework to capture the time-varying nature of modern markets. The analysis of queues and queueing networks with time-varying parameters is mathematically intractable. In this dissertation, heuristics which draw upon existing steady state queueing results are proposed to provide computationally efficient approximations for dynamic multi-product manufacturing systems modeled as time-varying queueing networks with multiple customer classes (product types). This dissertation addresses the problem of performance evaluation of such manufacturing systems. This dissertation considers the two key aspects of dynamic multi-product manufacturing systems - namely, performance evaluation and optimal server resource allocation. First, the performance evaluation of systems with infinite queueing room and a first-come first-serve service paradigm is considered. Second, systems with finite queueing room and priorities between product types are considered. Finally, the optimal server allocation problem is addressed in the context of dynamic multi-product manufacturing systems. The performance estimates developed in the earlier part of the dissertation are leveraged in a simulated annealing algorithm framework to obtain server resource allocations. / Dissertation/Thesis / Doctoral Dissertation Industrial Engineering 2020

No description available.

Cette thèse porte sur la dynamique de fronts de réaction-diffusion en présence de défauts localisés. Nous étudions des non-linéarités bistable et monostable pour lesquelles il existe des solutions exactes en milieu homogène. L'équation aux dérivées partielles est résolue numériquement et la solution est approchée en utilisant des solutions exactes. Parallèlement, nous développons une analyse en coordonnées collectives, position et largeur du front, basée sur des lois d'équilibre. Pour les deux non-linéarités, l'analyse approchée est en bon accord avec la solution numérique. Il est de plus possible de prédire l'arrêt du front dans le cas bistable. L'étude révèle des différences qualitatives entre les deux types de non linéarités. Elle montre l'importance des dimensions caractéristiques du défaut et du front. Enfin, elle fournit un modèle standardisé qui peut servir en théorie du contrôle ou pour la détermination de paramètres à partir de séries temporelles. / We study reaction-diffusion fronts in presence of a localized defect. We consider bistable and monostable nonlinearities for which exact solutions exist in the homogeneous case. The partial differential equation is solved numerically and the solution is fitted using these exact solutions. We also develop a collective coordinate analysis for the position and width of a front, based on balance laws. For both non linearities, the approximate analysis agrees well with the numerical solution. We cab predict the pinning of the front in the bistable case. The sudy reveals qualitative differences between the two nonlinearities. It shows the importance of the characteristic lenghts of the defect and the front. Finally it provides a reduced model, useful for control theory or for the determination of parameters from time-series.