Le français est la troisième langue enseignée au Yémen après l'arabe et l'anglais. Cette vitalité du français nécessite la création de supports didactiques, afin d'aider à son apprentissage. L'objet de la thèse est de mettre en évidence les zones du français, qui sont sources de difficultés pour les étudiants tant sur le plan de la morphologie que sur celui de la syntaxe, de les classer et de les décrire. L'étude des erreurs s'appuiera sur deux corpus de travaux écrits effectués, l'un par trente étudiants de quatrième année de licence du département de français de l'université de Dhamar, l'autre par quinze enseignants de lycée diplômés du même département. Cette étude a pour finalité d'initier une réflexion, au sein des départements de français des universités yéménites, sur les stratégies et les objectifs de l'enseignement du français en tant que langue, dans un milieu plurilingue non francophone. Elle vise également à offrir des ressources didactiques en matière d'étude de la langue aux enseignants, en tentant de mettant en évidence des priorités. / French is the third language taught in Yemen after Arabic and English. This vitality of French requires the creation of didactic supports, in order to help in its learning. The purpose of the thesis is to highlight the areas of French, which are sources of difficulty for students, in terms of morphology and syntax, to classify and to describe them. The study of errors is based on two corpus of written work, one which has been done by thirty students in the fourth year of the license of the French Department of Thamar University, the other one by fifteen high school teachers, graduated in the same department. This study aims to introduce a reflection, in the French departments of the Yemenite universities, about strategies and objectives of the teaching of French as a language, in a non-French-speaking multilingual environment. It also aims to provide didactic resources for the study of the language to the teachers, trying to highlighting the priorities.

In this thesis, we introduce EPOSIC channels, a class of SU(2) -covariant quantum channels. For each of them, we give a Stinespring representation, a Kraus representation, its Choi matrix, a complementary channel, and its dual map. We show that these channels are the extreme points of all SU(2) -irreducibly covariant channels. As an application of these channels to the theory of quantum information, we study the minimal output entropy of EPOSIC channels, and show that a large class of these channels is a potential example of violating the well-known problem, the additivity problem. We determine the cases where their minimal output entropy is not zero, and obtain some partial results on the fulfillment of their entanglement breaking property. We find a bound of the minimal output entropy of the tensor product of two SU(2) -irreducibly covariant channels. We also get an example of a positive map that is not completely positive.

Hypertrophic cardiomyopathy (HCM), a heart-related abnormality, is the most prevalent cause of sudden death in young athletes at sporting events. A cluster of cardiomyopathy mutations are localized in β-cardiac myosin at the N-terminal region of subfragment-2. Using resonance energy transfer probes, a synthetic peptide model system was developed to study stability of the coiled coil (S2 fragment) structure by determining monomer-dimer equilibrium of the peptide. Fluorescence resonance energy transfer and MacroModel software suite were used to obtain distance measurements along with measurement of coiled coil formation. The model peptide was used to characterize the effects of disease-causing-mutations and examine potential candidate drugs (polyamines) to counteract effects of mutations causing HCM. Distance measurements between donor and acceptor probes obtained by computational simulation and fluorescence resonance energy transfer (FRET) were consistent. Measurements also agreed with simulations of unlabeled wildtype, indicating coiled coil structural stability of the peptide. Interaction of the site-specific antibody with the peptide strongly inhibited dimerization and destabilized coiled coil structure of the peptide. Presence of negatively charged glutamate residues in the region of subfragment-2 strongly suggested a potential interaction site for positively charged polyamines. Binding of certain polyamines, such as poly-L-Lysine 11 residues and poly-D-Lysine 17 residues, demonstrated the ability to enhance dimerization and improve stability of the coiled coil structure, while some other polyamines were shown to have insignificant impact on the structure. In an attempt to characterize the effect of HCM-causing-mutations, peptides containing E924K mutation and lethal mutation E930 deletion were synthesized. Fluorescence resonance probes were conjugated to the mutant peptides to determine coiled coil formation. Results obtained from both dynamic simulations and resonance energy transfer experiments indicated that these mutations strongly inhibit dimerization, and thus, destabilize coiled coil structure of the peptide. Further experiments were conducted using heterodimers containing a chain of wildtype and a chain of mutant peptide. Both E924K & Edel930 mutations destabilized coiled coil formation and prevented dimerization. This peptide model system would provide a promising tool for drug development targeting HCM-causing-mutations along the S2 region of myosin.

This work investigates the defect energies, band alignments, and charge carrier recombination in polycrystalline Cu(In1-xGax)(Se1-ySy)2 chalcopyrite thin films and the interrelationship with the alloy composition. Photoluminescence spectroscopy of investigated Cu-poor Cu(In,Ga)(Se,S)2 layers generally shows broad emission lines with the corresponding maxima shifting towards higher energies under decreasing temperature or under increasing excitation power. Admittance spectroscopy of Cu-poor ZnO/CdS/Cu(In,Ga)(Se,S)2 chalcopyrite devices shows that the activation energies of the dominant defect distributions involving donors at the CdS/absorber interface and deep acceptors in the chalcopyrite bulk, increase upon alloying CuInSe2 with S. The band alignments within the Cu(In1-xGax)(Se1-ySy)2 system are determined using the energy position of the bulk acceptor state as a reference. The band gap enlargement under Ga alloying is accommodated almost exclusively in the rise of the conduction band edge, whereas the increase of band gap upon alloying with S is shared between comparable valence and conduction band offsets. The extrapolated band discontinuities [delta]EV(CuInSe2/CuInS2) = -0.23 eV, [delta]EC(CuInSe2/CuInS2) = 0.21 eV, [delta]EV(CuInSe2/CuGaSe2) = 0.036 eV, and [delta]EC(CuInSe2/CuGaSe2) = 0.7 eV are in good agreement with theoretical predictions. Current-voltage analysis of Cu-poor ZnO/CdS/Cu(In,Ga)(Se,S)2 devices reveals recombination barriers which follow the band gap energy of the absorber irrespective of alloy composition, as expected for dominant recombination in the chalcopyrite bulk. In turn, the recombination at the active junction interface prevails in Cu-rich devices which display substantially smaller barriers when compared to the band gap energy of the absorber. The result indicates that the Cu-stoichiometry is the driving compositional parameter for the charge carrier recombination in the chalcopyrite heterojunctions under investigations.

Made available in DSpace on 2015-05-14T13:21:17Z (GMT). No. of bitstreams: 1 Arquivototal.pdf: 2697470 bytes, checksum: 43857465b95f5bc745fef7fb91591233 (MD5) Previous issue date: 2012-08-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This study was performed in order to synthesize and evaluate a new chiral tertiary amine, (S)-2-hydroxypropanoate 2-N, N-dimetilaminoetila according to the concept of multifunctional catalyst, which allows one same molecule presents various characteristics such as Bronsted acid and Lewis base. Allied to the interests of organocatalysis of simplicity and low operating costs, resulting in products that do not have contaminants like metals, designed in order to be able to catalyze / promote the reaction of Morita-Baylis-Hillman (MBH). The synthesis was performed starting from cheap raw materials and good commercial availability, S-lactic acid and 2-N, Ndimetilaminoetan-1-ol in one step reaction (98%). Based on current mechanistic proposal of the MBH reaction, the mechanism of Cantillo and Kappe, we investigated the reaction conditions, where there is a hydrogen transfer in aprotic media and also in the presence of phenols as additives, checking the influence speeds, reaction yields and asymmetric induction. We investigated the reaction between pnitrobenzaldehyde and acrylonitrile as reaction MBH pattern, varying experimental conditions such as solvent, quantity of (S)-2-hydroxypropanoate 2-N, Ndimetilaminoetila time, temperature and addition of phenolic additives. In addition, a new method was developed for the separation of enantiomers for the racemic compound antiparasitic 2 - [Hydroxy (4-nitrophenyl) methyl] acrylonitrile by gas chromatography with high resolution capillary columns packed with chiral phases. Chemical yields were obtained 2 - [hydroxy (4-nitrophenyl) methyl] <1 to 98% and enantiomeric excesses up to 20.5%. Mechanistic aspects are discussed in this paper, based on the results and computations, using the theory DFT and QTAIM. / Este trabalho foi realizado com o intuito de sintetizar e avaliar uma nova amina terciária quiral, o (S)-2-hidroxipropanoato de 2-N,N-dimetilaminoetila de acordo ao conceito de catalisador multifuncional, o qual permite que uma mesma molécula apresente várias características como os de ácido de Bronsted e base de Lewis. Aliado aos interesses da organocatálise de simplicidade e baixo custo operacional, resultando em produtos que não apresentam metais como contaminantes, projetamos no intuito de ser capaz de catalisar/promover a reação de Morita-Baylis-Hilman (MBH). A síntese de foi efetuada a partir de matérias-primas baratas e de boa disponibilidade comercial; o ácido S-lático e o 2-N,Ndimetilaminoetan- 1-ol em uma etapa reacional (98%). Baseado na atual proposta mecanística da reação de MBH, o mecanismo de Cantillo e Kappe, foram investigados as condições reacionais, onde há a transferência de hidrogênio em meio aprótico e também na presença de fenóis como aditivos, verificando a influência nas velocidades, rendimentos de reação e indução assimétrica. Investigamos a reação entre o p-nitrobenzaldeído e a acrilonitrila como reação de MBH padrão, variando condições experimentais como: solvente, quantidade do(S)-2-hidroxipropanoato de 2-N,N-dimetilaminoetila, tempo, temperatura e adição de aditivos fenólicos. Em adição, foi desenvolvido um novo método de separação de enantiômeros para o composto racêmico antiparasitário 2-[Hidroxi(4-nitrofenil)metil] acrilonitrila via cromatografia gasosa com colunas capilares de alta resolução empacotadas por fases quirais. Foram obtidos rendimentos químicos do 2-[Hidroxi(4- nitrofenil)metil] <1 à 98% e excessos enantioméricos até 20,5%. Aspectos mecanísticos são discutidos neste trabalho, baseados nos resultados obtidos e em cálculos computacionais, usando a teoria DFT e QTAIM.

L'unité structurale de la chromatine, nommée nucléosome, est composée d'un double brin d'ADN enroulé autour d'un octamère d'histone, et subit une pléthore de modifications post-traductionnelles. Les conséquences biologiques de l’acétylation des lysines et de l’isomérisation des liaisons peptidyl-prolyl ont été étudiées à travers une analyse à l’échelle atomique par RMN de systèmes d'intérêt reconstitués in vitro. Les liaisons peptidyl-prolyl du domaine N-terminal de l'histone H3 sont substrats in vitro d’une isomérase chez S. cerevisiae nommée Fpr4p, laquelle exerce un contrôle catalyse-dépendant de la transcription. La résolution de la structure du domaine catalytique de Fpr4p, à partir de contraintes géométriques mesurées par RMN, révéla un domaine canonique de la famille FKBP (FK506-binding protein). Grâce à l'analyse de la séquence primaire et aux expériences RMN, nous proposons un modèle structural préliminaire de Fpr4p entière. L'analyse fonctionnelle est réalisée grâce à trois décapeptides construits à partir de la séquence primaire de H3 chez S. cerevisiae. Ils sont tous substrats de Fpr4p et la catalyse est équivalente pour Pro16 et Pro30. La proportion à l'équilibre du conformère cis fut déterminée pour les trois peptides et celle-ci n'est pas affectée par l'activité catalytique de Fpr4p. Les structures en solution des substrats en conformation trans ont été résolues par spectroscopie RMN, et seront utilisées pour des appariements moléculaires in silico sur le domaine catalytique de Fpr4p. Pour étudier le rôle biologique de l'acétylation des histones, une méthodologie de production de protéines acétylées a été développée. Le protocole repose sur la mutation d'une lysine en cystéine d'une protéine recombinante, suivie d'une alkylation contrôlée exploitant la nucléophilie du groupe thiol préalablement introduit. La production de l'agent alkylant adéquat est simple, rapide, réalisable dans un laboratoire de biologie et permet différents marquages isotopiques du groupe acétyle. L'alkylation d'une protéine repliée fut réalisée avec succès en conditions natives. Le dimère d'histone H2A-H2B, un intermédiaire de l'assemblage du nucléosome et siège d'acétylation in vivo, fut reconstruit in vitro. Les déplacements chimiques des domaines N et C-terminaux de H2A sont cohérents avec un état intrinsèquement déstructuré bien que leurs dynamiques moléculaires ne soient pas équivalentes. / The structural unit of chromatin, the nucleosome, is composed of double-stranded DNA wrapped around a histone octamer and is subject to a plethora of post-translational modifications. The biological consequences of peptidyl-prolyl isomerization and lysine acetylation were investigated at atomic scale through analysis of in vitro reconstituted systems by NMR. Peptidyl-prolyl bonds of histone H3 N-terminal domain are substrates in vitro of an isomerase from S. cerevisiae named Fpr4p, which underlies transcriptional control dependent on its catalytic activity. The solution structure of the catalytic domain of Fpr4p was calculated based on restraints from NMR spectroscopy, and reveals a canonical catalytic domain belonging to the FK506-binding protein (FKBP) family. Based on primary sequence analysis and NMR experiments, a preliminary structural model of full length Fpr4p is also presented. Functional analyses were performed with three decapeptides designed from the primary sequence from the N-terminal tail of S. cerevisiae histone H3. All three constitute substrates of Fpr4p, with equivalent catalysis observed for Pro16 and Pro30. The equilibrium proportion of the cis-proline conformer has been determined for all three decapeptides, and these populations are unaffected by Fpr4p catalytic activity. Structural ensembles of the substrates with proline in the trans conformation were determined by using NMR spectroscopy, and will be subsequently used for in silico molecular docking onto Fpr4p. To study a second form of histone regulation, a semi-synthetic method to produce acetylated protein was developed. The protocol relies on the site-specific mutation of lysine to cysteine in recombinant proteins followed by controlled alkylation thanks to nucleophilicity of the introduced thiol. The production of the required alkylation reagent is easy, quick, and suitable for biology laboratory and allows diverse isotopic labeling within the acetyl group. Alkylation of folded proteins has also been achieved in native conditions. As one target of acetylation in vivo, the histone H2A-H2B dimer is an intermediate of nucleosome assembly and was reconstituted in vitro. Chemical shift values of the N- and C-terminal domains of H2A are in agreement with an intrinsically disordered state although they display differences in dynamic mobility.

This work investigates the defect energies, band alignments, and charge carrier recombination in polycrystalline Cu(In1-xGax)(Se1-ySy)2 chalcopyrite thin films and the interrelationship with the alloy composition. Photoluminescence spectroscopy of investigated Cu-poor Cu(In,Ga)(Se,S)2 layers generally shows broad emission lines with the corresponding maxima shifting towards higher energies under decreasing temperature or under increasing excitation power. Admittance spectroscopy of Cu-poor ZnO/CdS/Cu(In,Ga)(Se,S)2 chalcopyrite devices shows that the activation energies of the dominant defect distributions involving donors at the CdS/absorber interface and deep acceptors in the chalcopyrite bulk, increase upon alloying CuInSe2 with S. The band alignments within the Cu(In1-xGax)(Se1-ySy)2 system are determined using the energy position of the bulk acceptor state as a reference. The band gap enlargement under Ga alloying is accommodated almost exclusively in the rise of the conduction band edge, whereas the increase of band gap upon alloying with S is shared between comparable valence and conduction band offsets. The extrapolated band discontinuities [delta]EV(CuInSe2/CuInS2) = -0.23 eV, [delta]EC(CuInSe2/CuInS2) = 0.21 eV, [delta]EV(CuInSe2/CuGaSe2) = 0.036 eV, and [delta]EC(CuInSe2/CuGaSe2) = 0.7 eV are in good agreement with theoretical predictions. Current-voltage analysis of Cu-poor ZnO/CdS/Cu(In,Ga)(Se,S)2 devices reveals recombination barriers which follow the band gap energy of the absorber irrespective of alloy composition, as expected for dominant recombination in the chalcopyrite bulk. In turn, the recombination at the active junction interface prevails in Cu-rich devices which display substantially smaller barriers when compared to the band gap energy of the absorber. The result indicates that the Cu-stoichiometry is the driving compositional parameter for the charge carrier recombination in the chalcopyrite heterojunctions under investigations.

Experiments to determine the fate and transport of the chemical warfare agent (CWA) simulants diisopropyl fluorophosphate (DIFP), O,S-diethyl methylphosphonothioate (OSDEMP), and 2-Chloroethyl ethyl sulfide (CEES) exposed to complex matrix systems are reported here. The aforementioned simulants were used in place of O-isopropyl methylphosphonofluoridate (GB), O-Ethyl S-(2-diisopropylaminoethyl) methylphosphonothiolate (VX), and Bis (2-chloroethyl) sulfide (HD), respectively. At ambient temperature, simulant pH (2.63 to 12.01) and reaction time (1 minute to 24 hours) were found to have significant influence on the recovery of simulants from charcoal, plastic, and TAP (butyl rubber gloves) in aqueous media. Buffer systems used included, phosphate, acetate, borate, and disodium tetraborate. Organic extractions were carried out using a 90:10 (v/v) dichloromethane / 2-propanol solution. All extracts were analyzed with a gas chromatograph equipped with flame ionization and flame photometric detectors (GC-FID-FPD). The FPD was used to determine the amount of simulant recovery.

At the Dawn of the 21st century, the field of specialty optical fibers experienced a scientific revolution with the introduction of the stack-and-draw technique, a multi-steps and advanced fiber fabrication method, which enabled the creation of well-controlled micro-structured designs. Since then, an extremely wide variety of finely tuned fiber structures have been demonstrated including novel materials and novel designs. As the complexity of the fiber design increased, highly-controlled fabrication processes became critical. To determine the ability of a novel fiber design to deliver light with properties tailored according to a specific application, several mode analysis techniques were reported, addressing the recurring needs for in-depth fiber characterization. The first part of this dissertation details a novel experiment that was demonstrated to achieve modal decomposition with extended capabilities, reaching beyond the limits set by the existing mode analysis techniques. As a result, individual transverse modes carrying between ~0.01% and ~30% of the total light were resolved with unmatched accuracy. Furthermore, this approach was employed to decompose the light guided in Large-Mode Area (LMA) fiber, Photonic Crystal Fiber (PCF) and Leakage Channel Fiber (LCF). The single-mode performances were evaluated and compared. As a result, the suitability of each specialty fiber design to be implemented for power-scaling applications of fiber laser systems was experimentally determined. The second part of this dissertation is dedicated to novel specialty fiber laser systems. First, challenges related to the monolithic integration of novel and complex specialty fiber designs in all-fiber systems were addressed. The poor design and size compatibility between specialty fibers and conventional fiber-based components limits their monolithic integration due to high coupling loss and unstable performances. Here, novel all-fiber Mode-Field Adapter (MFA) devices made of selected segments of Graded Index Multimode Fiber (GIMF) were implemented to mitigate the coupling losses between a LMA PCF and a conventional Single-Mode Fiber (SMF), presenting an initial 18-fold mode-field area mismatch. It was experimentally demonstrated that the overall transmission in the mode-matched fiber chain was increased by more than 11 dB (the MFA was a 250 ?m piece of 50 ?m core diameter GIMF). This approach was further employed to assemble monolithic fiber laser cavities combining an active LMA PCF and fiber Bragg gratings (FBG) in conventional SMF. It was demonstrated that intra-cavity mode-matching results in an efficient (60%) and narrow-linewidth (200 pm) laser emission at the FBG wavelength. In the last section of this dissertation, monolithic Multi-Core Fiber (MCF) laser cavities were reported for the first time. Compared to existing MCF lasers, renown for high-brightness beam delivery after selection of the in-phase supermode, the present new generation of 7-coupled-cores Yb-doped fiber laser uses the gain from several supermodes simultaneously. In order to uncover mode competition mechanisms during amplification and the complex dynamics of multi-supermode lasing, novel diagnostic approaches were demonstrated. After characterizing the laser behavior, the first observations of self-mode-locking in linear MCF laser cavities were discovered.