A presente tese fornece uma revisão bibliográfica dos trabalhos mais importantes sobre as reduções catódicas e anódicas do grupo carbonila em aldeídos, cetonas, dicetonas, cetonas α, β-insaturadas, cetoácidos, cetoésteres, e ésteres α, &#946 -insaturados. São descritas as sínteses por nós efetuadas de 13 hidroxi e cetoésteres metílicos, a saber: mandelato , 4-carbometoxi, 4-cloro, 4-nitro e 4-metoximandelatos, α-fenil-β-hidroxibutirato, fenilglioxilato, 4-carbometoxi, 4-cloro, 4-nitro e 4-metoxifenilglioxilatos,benzilpiruvato e piruvato. A pureza desses compostos foi testada por meio de espectroscopia no I.V .e R.M.N., cromatografia de gás ou em camada delgada e, em alguns casos, por análise elementar. 4-nitrofenilglioxilato, 4-carbometoxifenilglioxilato e 4-carbometoximandelato, compostos ainda não descritos na literatura, foram caracterizados e analisados. O trabalho fornece as eletrólises, com eletrodos de platina em metanol, de 10 cetoésteres que incluem, além dos acima enumerados, o α -cetobutirato de metila, 4-cetopimelato de etila e o cetomalonato de etila. Os produtos de reação obtidos foram identificados por métodos cromatográficos e espectroscópicos. São apresentadas dois tipos de experiências eletrolíticas: as simples e as que foram acompanhadas em tempos periódicos pela análise cromatográfica, sendo as primeiras, em alguns casos, repetidas, variando-se o tempo da eletrólise. Os resultados obtidos demonstram que são reduzidos aos hidroxiésteres correspondentes os seguintes cetoésteres contendo anel aromático: fenilglioxilato, 4-cloro e 4-carbometoxifenilglioxilatos e benzilpiruvato, como também, os 2 cetoésteres alifáticos -α -cetobutirato e piruvato. Entretanto, verifica-se que tanto 4-nitro e 4-metoxifenilglioxilatos como cetomalonato e 4-cetopimelato não sofrem redução. Estes resultados comparados com os anteriores do nosso laboratório, permitem sugerir a seguinte ordem de velocidade da redução: α -cetovalerato > α-cetobutirato > piruvato > fenilglioxilato) > 4-clorofenilglioxilato > 4-carbometoxifenilglioxilato. São fornecidas provas de que a falta de reatividade de 4-nitrofenilglioxilato e cetomalonato e a diminuição de reatividade de 4-cloro e 4-carbometoxifenilglioxilatos, como também, de piruvato são devidas à existência do equilíbrio cetoéster-semiacetaléster nestes compostos, em metanol. Finalmente, é apresenta.da uma discussão do mecanismo da redução de cetoésteres que justificaria a diminuição de reatividade de fenilglioxilato em relação aos α-cetoésteres alifáticos e a falta de reatividade. de 4-cetopimelato e 4-metoxifenilglioxilatonilglioxilato. / The present work describes the syntheses of some aliphatic and aromatic α-hydroxy - and α-keto-methyl esters and reports the electrolyses of the latter, in methanol, at a platinum cathode Some simple electrolytic experiments, varying the experimental conditions as well as those in which the transformations occurring during the electrolyses were followed by gas chromatography, are described. The reactivities towards cathodic reduction are reported and the following order of the relative rate for the reduction is suggested: α-oxovalerate > α-oxobutyrate > pyruvate> phenylglyoxylate > 4-chlorophenylglyoxylate > 4-methoxycarbonylphenylglyoxylate .No reduction is found to occur with 4-methoxyphenylglyoxylate 4-nitrophenylglyoxylate or 4-oxopymelate .An explanation for these differences in reactivities is suggested on the basis of electronic effects, steric inhibition of coplanarity, and hemiacetal formation

Nosso objetivo principal é aproveitar a bem conhecida diastereosseletividade π-facial das (SS)-2-tolilsulfinil-1,4-benzoquinonas, em reações de Diels-Alder, para obter, via metátese olefínica fototérmica, precursores enantiopuros do capneleno e da icarugamicina. Devido à formação de uma inseparável mistura de produtos na reação da (SS)-2-tolilsulfinil-1,4-benzoquinona com 1-metilciclopentadieno comercial, esta rota se tornou inviável para obtenção do capneleno. Visando-se a síntese do precursor da icarugamicina, foram preparadas as 5 e 6-metil-2-tolilsulfinil-1,4-benzoquinonas mas, apesar dos bons resultados de seletividade π-facial que estas apresentaram com o 1,3-cicloexadieno, a eliminação espontânea de ácido sulfênico nos adutos de Diels-Alder formados impediu o uso destes últimos. Numa tentativa de sobrepujar este inconveniente, uma nova quinona clorada foi preparada mas esta se mostrou um oxidante frente ao 1,3-cicloexadieno. / Our main objective is to take advantage of the well known π-facial diastereoselectivity of (SS)-2-tolylsulfinyl-1,4-benzoquinones, in the Diels-Alder reactions, to obtain, via photo-thermal olefin metathesis, enantiopure precursors of capnellene and ikarugamycin. The formation of an inseparable mixture of products in the reaction of commercially available methylcyclopentadiene and (SS)-2-tolylsulfinyl-1,4-benzoquinone made this route unsuitable for obtaining capnellene. For the synthesis of ikarugamycin, the 5 and 6-methyl-2-(SS)-tolylsulfinyl-1,4-benzoquinones were prepared but, in spite of the good π-diastereoselectivity they exhibited, in the reaction with 1,3-cyclohexadiene, spontaneous elimination of sulfenic acid from the formed adducts precluded their use. In an attempt to overcome this very easy elimination, a chlorinated quinone was synthesized but it showed oxidizing properties in the presence of 1,3-cyclohexadiene.

One of the state-of-the-art highlights of deep learning in the past ten years is the introduction of generative adversarial networks (GANs), which had achieved great success in their ability to generate images comparable to real photos with minimum human intervention. These networks can generalise to a multitude of desired outputs, especially in image-to-image problems and image syntheses. This thesis proposes a computer graphics pipeline for 3D rendering by utilising generative adversarial networks (GANs). This thesis is motivated by regression models and convolutional neural networks (ConvNets) such as U-Net architectures, which can be directed to generate realistic global illumination effects, by using a semi-supervised GANs model (Pix2pix) that is comprised of PatchGAN and conditional GAN which is then accompanied by a U-Net structure. Pix2pix had been chosen for this thesis for its ability for training as well as the quality of the output images. It is also different from other forms of GANs by utilising colour labels, which enables further control and consistency of the geometries that comprises the output image. The series of experiments were carried out with laboratory created image sets, to pursue the possibility of which deep learning and generative adversarial networks can lend a hand to enhance the pipeline and speed up the 3D rendering process. First, ConvNet is applied in combination with Support Vector Machine (SVM) in order to pair 3D objects with their corresponding shadows, which can be applied in Augmenter Reality (AR) scenarios. Second, a GANs approach is presented to generate shadows for non-shadowed 3D models, which can also be beneficial in AR scenarios. Third, the possibility of generating high quality renders of image sequences from low polygon density 3D models using GANs. Finally, the possibility to enhance visual coherence of the output image sequences of GAN by utilising multi-colour labels. The results of the adopted GANs model were able to generate realistic outputs comparable to the lab generated 3D rendered ground-truth and control group output images with plausible scores on PSNR and SSIM similarity index metrices.

As cianobactérias apresentam distribuição variada e podem ocorrer desde as regiões frias do ártico até os trópicos, em corpos dágua doce e no ambiente marinho. Algumas espécies de cianobactérias produzem compostos com conhecida toxidade, podendo causar efeitos deletérios em seres humanos e animais. Entre estes compostos estão os com atividade neurotóxica devido aos mais variados mecanismos de ação. O objetivo geral deste projeto é a obtenção por via sintética de algumas neurotoxinas e variantes, entre elas a β-N-metil-L-alanina (L-BMAA), anatoxina-a e anatoxina-a(s), bem como algumas aplicações. Para a L-BMAA, foi buscada a determinação de rotas sintéticas viáveis para a obtenção deste aminoácido modificado sob a forma racêmica e enantiomericamente pura, além de um possível produto cíclico que pode ser gerado naturalmente a partir da L-BMAA. Algumas rotas estratégicas foram determinadas com êxito para a síntese destes compostos. Entre as aplicações dos produtos obtidos podem ser citados: (i) a determinação de um método analítico para a determinação deste aminoácido por RMN de 1H; (ii) um método analítico por LC-MS utilizando D3-L-BMAA como padrão interno e (iii) a indicação de um possível mecanismo de neurotoxidade ligado a neurodegeneração via um intermediário produzido naturalmente do equilíbrio entre L-BMAA e íons bicarbonato. Duas rotas sintéticas estratégicas foram traçadas para a anatoxina-a. Na primeira, que não foi bem sucedida, partiu-se de um biciclo já formado e em sua etapa chave deveria ocorrer à expansão de um anel tropânico que conduziria ao biciclo 1:2:4. Na segunda, tentou-se a síntese a partir de uma molécula extremamente simples, o 1,5- ciclooctadieno, e após cinco passos reacionais obteve-se a síntese total da anatoxina-a sob a forma racêmica. A etapa mais importante para essa rota foi a ciclização utilizando-se paládio divalente. Entre as possíveis aplicações dos produtos obtidos sugere-se: (i) o desenvolvimento de um método analítico por LC-MS utilizando D3-anatoxina-a como padrão interno e (ii) ensaios biológicos com os análogos para se determinar potencial agonista, antagonista ou agonista parcial de receptores nicotínicos para esses compostos. A anatoxina-a(s) apresentou um grande desafio para a sua síntese total. Por se tratar de um organofosforado com uma parte alquílica heterocíclica, buscou-se primeiramente a sua síntese parcial. Assim, foi possivel a síntese da guanidina cíclica sob a forma racêmica e quiral, além da obtenção de alguns compostos análogos inéditos. Para a síntese da N-hidroxiguanidina cíclica traçou-se duas diferentes rotas, uma enantioseletiva, partindo do aminoácido asparagina, e outra racêmica, com a utilização de álcool benzílico como material de partida. No entanto, apesar da produção de intermediários e análogos, não foi possível a obtenção do produto final. Entre as aplicações dos produtos sintéticos obtidos estão: (i) o desenvolvimento e validação de um método analítico para a quantificação indireta de anatoxina-a(s) utilizando a N-hidroxiguanidina cíclica como padrão e (ii) a utilização destes intermediários sintéticos em ensaios para se determinar os mecanismos de ação tóxicos, especialmente como inibidores de colinesterases. / Cyanobacteria are aquatic and photosynthetic organisms that can be present in cold areas, such as the Arctic, as well as in tropical waters, in both marine and freshwater environment. They are important species for aquatic life and terrestrial ecosystems since they are on the base of the food web. However, some cyanobacterial species can produce toxic secondary metabolites that have deleterious effects for animals and human. These toxic metabolites are also called cyanotoxins and show different mechanisms of action ranging from hepatotoxic to neurotoxic effects. The aim of this study is the organic synthesis of some common neurotoxins, including some analogues, such as β-N-methyl-L-alanine (L-BMAA), anatoxin-a and anatoxin-a(s). Moreover, focus was given for some possible application of these compounds, especially as analytical standards for water monitoring. Some synthetic routes were carried out in order to produce L-BMAA in both racemic and enantiomerically pure forms. Also, it was synthesized a possible cyclic analogue that may be formed in vivo from L-BMAA. The racemate and the pure L-BMAA were successfully obtained. Some possible application of these compounds were suggested: (i) the development of an analytical method based on 1H NMR analyses for the determination of L-BMAA in environmental and biological samples; (ii) the development of an analytical method based on LC-MS for the determination of L-BMAA, using D3-L-BMAA as an internal standard, in different matrices and (iii) the indication of a cyclic derivative formed in vivo that can be involved in the neurotoxicity of L-BMAA. Two strategies were planned in order to produce anatoxin-a. The first one was not successful and was started by using the bicyclic precursor. The crucial step was the expansion of tropanic ring to produce the 1:2:4 byciclic anatoxin-a precursor. The second strategy was initiated with the 1,5 cyclicoctadiene and after five reaction steps anatoxin-a was finally synthesized in its racemic form. The most important step for this route was the use of palladium. Similar applications were suggested for anatoxin-a: (i) the development of an analytical method based on LC-MS for the determination of anatoxin-a using D3-anatoxin-a as an internal standard in different matrices and (ii) the screening of anatoxin-a and analogues in bioassays based on nicotinic receptors in order to determine their agonistic and antagonistic mechanism of action. The total synthesis of anatoxin-a(s) is still a challenge. This cyanotoxin is a natural organophosphate with an alkylic heterocyclic chain. Therefore, the first step was the production of the cyclic guanidine in both racemic and enantiomerically pure forms. Some cyclic guanidine derivatives were prepared via asparagine and/or benzilic alcohol. However, anatoxin-a(s) was not achieved. The use of the cyclic guanidine can be recommended for: (i) the development of an analytical method based on LC-MS for the indirect determination of anatoxin-a(s) using its alkylic chain as a standard after sample alkalinization and (ii) the screening of anatoxin-a(s) derivatives in cholinesterase assays to determine their toxicity and mechanisms of action.

La détermination de la conformation d’une petite molécule liée à sa cible biologique nous permet de concevoir des drogues de propriété biologique améliorée. Cette détermination peut être difficile dû aux limitations techniques, comme indiqué par le débat sur la conformation de microtubule-lié d’une drogue anticancéreuse – paclitaxel. Les études utilisant la cristallographie des rayons X et la RMN du liquide ne peut pas fournir les informations détaillées sur la conformation espérée. La RMN du solide est un choix raisonnable en mesurant précisément des distances interatomiques de la molécule, et le marquage sélectif au tritium permet de mesurer une distance longue jusqu’à 14,4 Å avec une précision haute grâce au rapport gyromagnetique élevé de ce noyau. Aucune modification structurale n’a été rendue par le marquage au tritium. Ainsi notre sujet ayant pour l’objectif de déterminer la conformation bioactive du paclitaxel comporte la synthèse des 6 isotopomères de paclitaxel ditritiés sur les sites particuliers, suivie par la préparation des complexes de microtubule-paclitaxel marqué. L’analyse de RMN du tritium à l’état solide fournira les distances clés pour la détermination. 2 isotopomères ont été synthétisés par tritier le paclitaxel dibromé et coupler la baccatine tritiée et la chaîne latérale tritiée, respectivement. La stratégie synthétique conçue permet de réaliser la synthèse avec un rendement généralement satisfaisant et une bonne stéréosélectivité. Différentes méthodes de tritiation ont été testées, dont un enrichissement isotopique supérieur à 92% a été obtenu. La synthèse des autres isotopomères ainsi que des complexes de microtubule-paclitaxel est en cours de réaliser dans notre laboratoire. / The determination of the conformation of small molecule bound to its biological target would facilitate people to design improved drugs. This determination can be difficult due to technical limitations, as exemplified by the long standing debate on the microtubule-binding conformation of a natural anticancer drug – paclitaxel. Previous studies using X-ray crystallography and solution-state NMR failed to furnish direct information on the expected conformation. Solid-state NMR may help in this task by providing precise interatomic distances, and the selective labeling on different sites with tritium atoms enables accurate measurement of long-range distances (up to 14.4 Å) owing to the high gyromagnetic ratio of this nucleus, without any structural modification of the molecule. So our project aiming at illustrating the bioactive conformation of paclitaxel consists the syntheses of 6 different paclitaxel isotopomers bearing a pair of tritiums at specified positions, flowing by the preparations of corresponding microtubule-labeled paclitaxel complexes. The solid-state tritium NMR analyses of these complexes would provide key distances for determining the expected conformation. Up to now, 2 paclitaxel isotopomers have been prepared from labelling the dibrominated paclitaxel precursor and from coupling the tritiated taxane rings and the tritiated side chains, respectively. The synthetic strategy allowed us to realize the syntheses in generally high yield and good stereoselectivity. Different tritiation methods have been used, from which an isotopic enrichment of higher than 92% was obtained. The syntheses of other 4 isotopomers, together with the microtubule complexes are currently underway in our lab.

As cianobactérias apresentam distribuição variada e podem ocorrer desde as regiões frias do ártico até os trópicos, em corpos dágua doce e no ambiente marinho. Algumas espécies de cianobactérias produzem compostos com conhecida toxidade, podendo causar efeitos deletérios em seres humanos e animais. Entre estes compostos estão os com atividade neurotóxica devido aos mais variados mecanismos de ação. O objetivo geral deste projeto é a obtenção por via sintética de algumas neurotoxinas e variantes, entre elas a β-N-metil-L-alanina (L-BMAA), anatoxina-a e anatoxina-a(s), bem como algumas aplicações. Para a L-BMAA, foi buscada a determinação de rotas sintéticas viáveis para a obtenção deste aminoácido modificado sob a forma racêmica e enantiomericamente pura, além de um possível produto cíclico que pode ser gerado naturalmente a partir da L-BMAA. Algumas rotas estratégicas foram determinadas com êxito para a síntese destes compostos. Entre as aplicações dos produtos obtidos podem ser citados: (i) a determinação de um método analítico para a determinação deste aminoácido por RMN de 1H; (ii) um método analítico por LC-MS utilizando D3-L-BMAA como padrão interno e (iii) a indicação de um possível mecanismo de neurotoxidade ligado a neurodegeneração via um intermediário produzido naturalmente do equilíbrio entre L-BMAA e íons bicarbonato. Duas rotas sintéticas estratégicas foram traçadas para a anatoxina-a. Na primeira, que não foi bem sucedida, partiu-se de um biciclo já formado e em sua etapa chave deveria ocorrer à expansão de um anel tropânico que conduziria ao biciclo 1:2:4. Na segunda, tentou-se a síntese a partir de uma molécula extremamente simples, o 1,5- ciclooctadieno, e após cinco passos reacionais obteve-se a síntese total da anatoxina-a sob a forma racêmica. A etapa mais importante para essa rota foi a ciclização utilizando-se paládio divalente. Entre as possíveis aplicações dos produtos obtidos sugere-se: (i) o desenvolvimento de um método analítico por LC-MS utilizando D3-anatoxina-a como padrão interno e (ii) ensaios biológicos com os análogos para se determinar potencial agonista, antagonista ou agonista parcial de receptores nicotínicos para esses compostos. A anatoxina-a(s) apresentou um grande desafio para a sua síntese total. Por se tratar de um organofosforado com uma parte alquílica heterocíclica, buscou-se primeiramente a sua síntese parcial. Assim, foi possivel a síntese da guanidina cíclica sob a forma racêmica e quiral, além da obtenção de alguns compostos análogos inéditos. Para a síntese da N-hidroxiguanidina cíclica traçou-se duas diferentes rotas, uma enantioseletiva, partindo do aminoácido asparagina, e outra racêmica, com a utilização de álcool benzílico como material de partida. No entanto, apesar da produção de intermediários e análogos, não foi possível a obtenção do produto final. Entre as aplicações dos produtos sintéticos obtidos estão: (i) o desenvolvimento e validação de um método analítico para a quantificação indireta de anatoxina-a(s) utilizando a N-hidroxiguanidina cíclica como padrão e (ii) a utilização destes intermediários sintéticos em ensaios para se determinar os mecanismos de ação tóxicos, especialmente como inibidores de colinesterases. / Cyanobacteria are aquatic and photosynthetic organisms that can be present in cold areas, such as the Arctic, as well as in tropical waters, in both marine and freshwater environment. They are important species for aquatic life and terrestrial ecosystems since they are on the base of the food web. However, some cyanobacterial species can produce toxic secondary metabolites that have deleterious effects for animals and human. These toxic metabolites are also called cyanotoxins and show different mechanisms of action ranging from hepatotoxic to neurotoxic effects. The aim of this study is the organic synthesis of some common neurotoxins, including some analogues, such as β-N-methyl-L-alanine (L-BMAA), anatoxin-a and anatoxin-a(s). Moreover, focus was given for some possible application of these compounds, especially as analytical standards for water monitoring. Some synthetic routes were carried out in order to produce L-BMAA in both racemic and enantiomerically pure forms. Also, it was synthesized a possible cyclic analogue that may be formed in vivo from L-BMAA. The racemate and the pure L-BMAA were successfully obtained. Some possible application of these compounds were suggested: (i) the development of an analytical method based on 1H NMR analyses for the determination of L-BMAA in environmental and biological samples; (ii) the development of an analytical method based on LC-MS for the determination of L-BMAA, using D3-L-BMAA as an internal standard, in different matrices and (iii) the indication of a cyclic derivative formed in vivo that can be involved in the neurotoxicity of L-BMAA. Two strategies were planned in order to produce anatoxin-a. The first one was not successful and was started by using the bicyclic precursor. The crucial step was the expansion of tropanic ring to produce the 1:2:4 byciclic anatoxin-a precursor. The second strategy was initiated with the 1,5 cyclicoctadiene and after five reaction steps anatoxin-a was finally synthesized in its racemic form. The most important step for this route was the use of palladium. Similar applications were suggested for anatoxin-a: (i) the development of an analytical method based on LC-MS for the determination of anatoxin-a using D3-anatoxin-a as an internal standard in different matrices and (ii) the screening of anatoxin-a and analogues in bioassays based on nicotinic receptors in order to determine their agonistic and antagonistic mechanism of action. The total synthesis of anatoxin-a(s) is still a challenge. This cyanotoxin is a natural organophosphate with an alkylic heterocyclic chain. Therefore, the first step was the production of the cyclic guanidine in both racemic and enantiomerically pure forms. Some cyclic guanidine derivatives were prepared via asparagine and/or benzilic alcohol. However, anatoxin-a(s) was not achieved. The use of the cyclic guanidine can be recommended for: (i) the development of an analytical method based on LC-MS for the indirect determination of anatoxin-a(s) using its alkylic chain as a standard after sample alkalinization and (ii) the screening of anatoxin-a(s) derivatives in cholinesterase assays to determine their toxicity and mechanisms of action.

La lutte contre le cancer est l'un des défis majeurs du XXème siècle. Pour que les tumeurs puissent se développer dans l'organisme, elles ont besoin d'un apport en nutriment par le biais de vaisseaux sanguins pour se faire, elles vont avoir recours au processus angiogénique. Lors de ce processus, les cellules endothéliales qui tapissent la paroi des vaisseaux sanguins vont se multiplier et créer de nouveaux vaisseaux sanguins qui vont permettre la vascularisation des tumeurs. L'angiogenèse constitue donc aujourd'hui un axe de recherche pour la lutte contre la progression tumorale et donc contre le cancer. Lors de ce développement tumoral, une enzyme, l'aminopeptidase neutre APN est surexprimée sur les parois des cellules endothéliales. Différentes études ont été menées et montrent que l'inhibition de cette enzyme bloque la progression tumorale. Mon travail au sein de l'équipe du Pr Céline Tarnus consistait en la conception et la synthèse d'inhibiteurs de l'APN. Une relation structure activité de nos composés vis-à-vis de l'APN a tout d'abord été effectuée. Le développement de synthèse du composé le plus actif ont été faite, puis la synthèse d'inhibiteurs d'APN ayant pour objectif l'utilisation de la BNCT a été abordée. / The fight against the cancer is one of the most important struggles of this century. For the development of the tumors inside the body, they need to receive nutriments by the blood vessels and they use the angiogenic process. During this process, the endothelial cells being shown on the wall of the blood vessel will multiply and design new blood vessel, which will allow the tumor's vascularisation. Today, the angiogenesis is an axis of research for the fight against the cancer. During the tumoral development, the aminopeptidase N (APN) is overexpressed on the wall of endothelial cells. Various studies have shown that the inhibition of this enzyme stops the tumoral progression. My work in the Pr. Céline Tarnus Team consists in the conception and the synthesis of APN's inhibitors. In a first time, a structure activity relationship has been realized. Syntheses of a subnamolar compound have been developed, and then the synthesis of APN's inhibitors with the use of BNCT has been got onto.

This thesis summarizes the methods and major findings of Ni-W(P)/ã-Al2O3 nitride cata-lyst synthesis, characterization, hydrotreating activity, kinetic analysis and correlation of the catalysts activities to their synthesis parameters and properties.<p> The range of parameters for catalyst synthesis were W (15-40 wt%), Ni (0-8 wt%), P (0-5 wt%) and nitriding temperature (TN) (500-900 °C). Characterization techniques used included: N2 sorption studies, chemisorption, elemental analysis, temperature programmed studies, x-ray diffraction, scanning electron microscopy, energy dispersive x-ray, infrared spectroscopy, trans-mission electron microscopy and x-ray absorption near edge structure. Hydrodesulfurization (HDS), hydrodenitrogenation (HDN) and hydrodearomatization (HDA) were performed at: tem-perature (340-380 °C), pressure (6.2-9.0 MPa), liquid hourly space velocity (1-3 h-1) and hydro-gen to oil ratio (600 ml/ml, STP).<p> The predominant species on the catalyst surface were Ni3N, W2N and bimetallic Ni2W3N. The bimetallic Ni-W nitride species was more active than the individual activities of the Ni3N and W2N. P increased weak acid sites while nitriding temperature decreased amount of strong acid sites. Low nitriding temperature enhanced dispersion of metal particles. P interacted with Al2O3 which increased the dispersion of metal nitrides on the catalyst surface. HDN activity in-creased with Ni and P loading but decreased with increase in nitriding temperature (optimum conversion; 60 wt%). HDS and HDA activities went through a maximum with increase in the synthesis parameters (optimum conversions; 88. wt% for HDS and 47 wt% for HDA). Increase in W loading led to increase in catalyst activity. The catalysts were stable to deactivation and had the nitride structure conserved during hydrotreating in the presence of hydrogen sulfide.<p> The results showed good correlation between hydrotreating activities (HDS and HDN) and the catalyst nitrogen content, number of exposed active sites, catalyst particle size and BET surface area.<p> HDS and HDN kinetic analyses, using Langmuir-Hinshelwood models, gave activation energies of 66 and 32 kJ/mol, respectively. There were no diffusion limitations in the reaction process. Two active sites were involved in HDS reaction while one site was used for HDN. HDS and HDN activities of the Ni-W(P)/ã-Al2O3 nitride catalysts were comparable to the corre-sponding sulfides.

Shape and size control as well as the control of the assembly of nanostructures are current challenges in nano sciences. Focussing on metal nanostructures all of these aspects have been addressed in the frame of the present work. It was possible to develop a new aqueous seeded growth method that produces gold nanoparticles with adjustable diameters over a large range of sizes. The spherical particles obtained show very low polydispersities and a good long term stability. Furthermore it was possible to reveal the growth mechanism of these particles utilizing electron microscopy and optical investigations coupled with theoretical calculations. It was found that there is a formation of small nucleation sites on the surface of the seeds in the beginning of the growth process. These sites then subsequently grow into "blackberry-like" intermediate particles. A final intraparticle ripening step leads to smooth and uniform spherical gold nanoparticles. By correcting the dielectric function of gold for charging and the free mean path effect and taking into account the particle size distribution it was possible to accurately model the optical properties of the gold sols obtained using Mie theory. By controlling the concentration of chloride ions it was possible to influence both the ripening of the "blackberry-like" shaped particles and the morphology of gold nanoparticles. An increased concentration of the chloride ions in the standard citrate reduction procedure leads to larger and elongated particles, whereas the complete removal of the chloride ions made it possible to obtain star shaped, decahedral and \"desert-rose\" shaped particle morphologies. Using the layer-by-layer technique gold nanoparticles of different sizes could be immobilized on glass substrates. The surface-enhanced Raman scattering intensity of these mixed films were about 60% higher than compared to a film made of a single particle size. The optical properties were further investigated by comparing experimentally obtained UV/Vis spectra with generalized Mie theory simulations. Additionally it could be shown that tetrazole and its derivatives are suitable stabilizing agents in the aqueous synthesis of silver nanoparticles. It was found that depending on the tetrazole derivative used the tendencies of the nanoparticles to agglomerate vary significantly. Different agglomeration stages have been investigated by UV/Vis and Raman spectroscopy. The removal of the ligands used and a resulting improvement of the applicability of the silver nanostructures as SERS substrates is still a challenge. In the last part of this work the focus was changed from the optical properties of noble metal nanoparticles to their catalytic properties. Therefore gold and palladium nanoparticles have been successfully immobilized on highly porous zinc oxide aerogels. It was possible to synthesize sponge-, flake-, and ribbon-like zinc oxide gels with high specific surface areas. The facile approach of generating mixed metal oxide/noble metal aerogels is very promising for the preparation of highly selective and highly active heterogenous catalysts. First catalytic investigations of a sponge-like palladium loaded zinc oxide aerogel toward the semi-hydrogenation of acetylene showed very high selectivities of up to 85%.

This thesis summarizes the methods and major findings of Ni-W(P)/ã-Al2O3 nitride cata-lyst synthesis, characterization, hydrotreating activity, kinetic analysis and correlation of the catalysts activities to their synthesis parameters and properties.<p> The range of parameters for catalyst synthesis were W (15-40 wt%), Ni (0-8 wt%), P (0-5 wt%) and nitriding temperature (TN) (500-900 °C). Characterization techniques used included: N2 sorption studies, chemisorption, elemental analysis, temperature programmed studies, x-ray diffraction, scanning electron microscopy, energy dispersive x-ray, infrared spectroscopy, trans-mission electron microscopy and x-ray absorption near edge structure. Hydrodesulfurization (HDS), hydrodenitrogenation (HDN) and hydrodearomatization (HDA) were performed at: tem-perature (340-380 °C), pressure (6.2-9.0 MPa), liquid hourly space velocity (1-3 h-1) and hydro-gen to oil ratio (600 ml/ml, STP).<p> The predominant species on the catalyst surface were Ni3N, W2N and bimetallic Ni2W3N. The bimetallic Ni-W nitride species was more active than the individual activities of the Ni3N and W2N. P increased weak acid sites while nitriding temperature decreased amount of strong acid sites. Low nitriding temperature enhanced dispersion of metal particles. P interacted with Al2O3 which increased the dispersion of metal nitrides on the catalyst surface. HDN activity in-creased with Ni and P loading but decreased with increase in nitriding temperature (optimum conversion; 60 wt%). HDS and HDA activities went through a maximum with increase in the synthesis parameters (optimum conversions; 88. wt% for HDS and 47 wt% for HDA). Increase in W loading led to increase in catalyst activity. The catalysts were stable to deactivation and had the nitride structure conserved during hydrotreating in the presence of hydrogen sulfide.<p> The results showed good correlation between hydrotreating activities (HDS and HDN) and the catalyst nitrogen content, number of exposed active sites, catalyst particle size and BET surface area.<p> HDS and HDN kinetic analyses, using Langmuir-Hinshelwood models, gave activation energies of 66 and 32 kJ/mol, respectively. There were no diffusion limitations in the reaction process. Two active sites were involved in HDS reaction while one site was used for HDN. HDS and HDN activities of the Ni-W(P)/ã-Al2O3 nitride catalysts were comparable to the corre-sponding sulfides.