O iodeto de mercúrio em sua fase vermelha, \"alfa\"-\"HgI IND.2, é um material semicondutor que desperta grande interesse tecnológico devido à sua potencial aplicação como detector, a temperatura ambiente, de raios-\"gama\" e X. Sua imediata aplicação como detector de radiação, no entanto, ainda sofre algumas restrições devido às dificuldades de controle sobre a concentração de defeitos pontuais e extensos, durante sua síntese, e sua fácil degradação quando exposto ao ambiente. A presença destes defeitos gera uma redução na mobilidade de portadores de carga, diminuindo a eficiência de detecção. Neste trabalho, realizamos uma investigação teórica das propriedades estruturais, eletrônicas e ópticas do \"alfa\"-\"HgI IND.2\" e do \"ZnI IND.\", cuja estrutura cristalina é análoga à do \"alfa\"-\"HgI IND.2\". Estudamos, também, o comportamento de tais propriedades do material contendo vacâncias de Hg e de I para entendermos o papel desempenhado por estes defeitos nas propriedades do material. Na análise dos resultados, sempre que possível, são apresentadas comparações com outros resultados teóricos e com dados experimentais. Nossos cálculos foram efetuados dentro do formalismo do funcional da densidade em combinação com duas diferentes aproximações para o termo de exchange e correlação. Utilizamos o método Augmented Plane Wave plus local orbitals (APW+lo), que é um método de primeiros princípios e inclui todos os elétrons do sistema (all electron). Adicionalmente, foram utilizados esquemas que incluem efeitos relativísticos e polarização de spin. O estudo dos defeitos foi simulado através do esquema da supercélula / Mercuric iodide \"alfa\"-\"HgI IND.2 in its red tetragonal crystalline phase is a semiconducting material of great technological interest due to the potential applications as a detector for y- and X-ray spectroscopy to be operated at room temperatures.

La théorie de la fonctionnelle de la densité dépendante du temps (TDDFT) est une méthode basée sur la densité pour calculer les états excités. Bien que la TDDFT soit une théorie exacte, on doit en pratique partir d'une approximation de la fonctionnelle d'échange-corrélation, qui reste inconnue. L'approximation adiabatique est l'approximation de la fonctionnelle la plus courante. Cette approximation donne de très bons résultats pour les propriétés spectroscopiques, mais elle est inexacte pour les simulations en photochimie. Dans cette thèse, on montre que l'origine du problème réside dans l'approximation de la fonctionnelle de corrélation. Le résultat principal de la thèse consiste en un noyau de corrélation, qui peut être utilisé dans la formulation de la réponse linéaire, noyau dérivée à partir de la théorie des perturbations à plusieurs corps. Le noyau inclut de façon générale les excitations doubles qui donnent l'effet principal à la corrélation dans les états excités. La comparaison de ce noyau avec la fonctionnelle adiabatique nous a permis d'identifier les termes manquants à ce dernier. Nous avons testé la possibilité d'ajouter ces termes comme une correction à l'approximation adiabatique. Le noyau pourrait potentiellement être appliqué à des systèmes de grosse taille. / Time-dependent density-functional theory (TDDFT) is a density-functional method for calculating excited states. TDDFT is formally exact, though in practice one has to approximate the unknown exchange-correlation functional, which contains all the unknown many-body effects. The adiabatic functionals are the most commonly used. Although they are very successful for spectroscopy, the adiabatic functionals are too inaccurate to be applied to photochemistry. In this thesis, we show that the main problem is due to the approximations in the correlation functional. The main result of the thesis is a correlation kernel for linear-response TDDFT, derived using many-body perturbation theory techniques, which generally includes double excitations, thus introducing the leading correlation effects in the excited states. The comparison of this kernel with the adiabatic functionals allowed us to identify which correlation effects are missing in these approximation. We tested the possibility of improving the description of correlation by adding the missing terms from many-body theory to the adiabatic functionals. This mixed kernel is more efficient than the full many-body kernel, and can potentially be applied to systems of medium to large size.

Des études expérimentales et théoriques du mécanisme d'électrodéposition de différents revêtements à base de cobalt : le cobalt métallique, les alliages binaires Co-Ni, Co-Cu et l'alliage ternaire Co-Ni-Cu sont menés.Les propriétés de ces films dépendent fortement des paramètres expérimentaux tels que le pH de l'électrolyte, la nature du substrat, etc.. Un champ magnétique jusqu'à 12T est superposé parallèlement à la surface de l'électrode de travail. La convection forcée induite dans l'électrolyte modifie les conditions hydrodynamiques et par conséquent, influe sur la structure et la morphologie des films obtenus. Par ailleurs, la cinétique des processus et la croissance des cristaux sont améliorées dans des conditions magnéto électrochimique. Les microstructures obtenues suite au procédé électrochimique en présence d'un champ magnétique sont caractérisées par différentes propriétés magnétiques trouvant un intérêt potentiel en tant que matériaux magnétiques doux et / ou dur. Dans le cadre du projet ANR COMAGNET, les matériaux subissent ensuite un traitement thermique en présence d'un champ magnétique, des phénomènes de recristallisation et interdiffusion sont mis en évidence induisant des modifications des propriétés magnétiques.Enfin, des calculs de l'énergie d'adsorption de l'atome d'hydrogène et de l'ion H+ sur différentes faces cristallographiques du cobalt sont menés par modélisation DFT (Théorie de la Fonctionnelle de la Densité). Ces résultats, mis en relation avec l'étude expérimentale, montrent le lien entre la formation de certaines structures cristallographiques et l'évolution simultanée de l'hydrogène à la surface de l'électrode pendant l'électrolyse. / The present work is focused on the experimental and theoretical studies of the electrodeposition mechanism of Co-based films, including single Co, binary Co-Ni, Co-Cu and ternary Co-Ni-Cu alloy films. The modeling and structure-properties relationship by a multi-scale approach is investigated.The preliminary study involves determination and optimization of the electrodeposition parameters in order to obtain nanocrystalline films with satisfied surface quality and promising magnetic properties. It is shown that the electrolytic pH, to-be-deposited type of substrate and deposition time are of high importance in the fabrication of nanoscale materials. Among them, the electrolytic pH is the one, which has the greatest effect on the structure formation. The film growth as well as its quality can be strongly affected by a superimposition of an external magnetic field. The electrodeposition process was carried out under parallel to the electrode surface magnetic fields with magnets strength up to 12T. The results reveal that the induced forced convection in the electrolyte changes the hydrodynamic conditions and thus, affects the structure and morphology of the obtained films. Furthermore, the process kinetics and crystal growth are enhanced under magnetic electrodeposition conditions.Microstructure formed by the electrochemical processing is characterized by some significant magnetic properties, which may result in soft and/or hard magnetic materials, depending on their application approach. Additionally, the microstructure of films has been improved by the magnetic annealing treatment. Thus, the recrystallization and interdiffusion phenomena are observed, and modification of the magnetic properties is induced.Considering the electrodeposition process carried out in aqueous solutions the secondary process, which is the hydrogen evolution reaction (HER), needs to be taken into account. The theoretical study based on the tools of quantum mechanics (QM) and density functional theory (DFT) is used to determine the adsorption energies of hydrogen. In this work, the calculation results are related with the experimental study and may explain the structure formation assisted by the simultaneous evolution of hydrogen at the electrode surface.

En raison de l'augmentation de la demande d'énergie, ressources écologiques respectueux de l'environnement et durables (solaires, éoliennes) doivent être développées afin de remplacer les combustibles fossiles. Ces sources d'énergie sont discontinues, étant corrélés avec les conditions météorologiques et leur disponibilité est fluctuant dans le temps. En conséquence, les dispositifs de stockage d'énergie à grande échelle sont devenus incontournables, pour stocker l'énergie sur des échelles de temps longues avec une bonne compatibilité environnementale. La conversion d'énergie électrochimique est le mécanisme clé pour les développements technologiques des sources d'énergie alternatives. Parmi ces systèmes, les batteries Lithium-ion (LIB) ont démontré être les plus robustes et efficaces et sont devenus la technologie courante pour les systèmes de stockage d'énergie de haute performance. Ils sont largement utilisés comme sources d'énergie primaire pour des applications populaires (ordinateurs portables, téléphones cellulaires, et autres). La LIB typique est constitué de deux électrodes, séparés par un électrolyte. Celui-ci joue un rôle très important dans le transfert des ions entre les électrodes fournissant la courante électrique. Ce travail de thèse porte sur les matériaux complexes utilisés comme électrolytes dans les LIB, qui ont un impact sur les propriétés de transport du ion Li et les performances électrochimiques. Habituellement l'électrolyte est constitué de sels de Li et de mélanges de solvants organiques, tels que les carbonates cycliques ou linéaires. Il est donc indispensable de clarifier les propriétés structurelles les plus importantes, et leurs implications sur le transport des ions Li+ dans des solvants purs et mixtes. Nous avons effectué une étude théorique basée sur la théorie du fonctionnelle densité (DFT) et la dynamique moléculaire (MD), et nous avons consideré des carbonates cyclique (carbonate d'éthylène, EC, et carbonate de propylène, PC) et le carbonate de diméthyle, DMC, linéaire. Les calculs DFT ont fourni une image détaillée des structures optimisées de molécules de carbonate et le ion Li+, y compris les groupes pures Li+(S)n (S =EC,PC,DMC et n=1-5), groupes mixtes binaires, Li+(S1)m(S2)n (S1,S2=EC,PC,DMC, m+n=4), et ternaires Li+(EC)l(DMC)m(PC)n (l+m+n=4). L'effet de l'anion PF6 a également été étudié. Nous avons aussi étudié la structure de la couche de coordination autour du Li+, dans tous les cas. Nos résultats montrent que les complexes Li+(EC)4, Li+(DMC)4 et Li+(PC)3 sont les plus stables, selon les valeurs de l'énergie libre de Gibbs, en accord avec les études précédentes. Les énergies libres de réactions calculés pour les mélanges binaires suggèrent que l'ajout de molécules EC et PC aux clusters Li+ -DMC sont plus favorables que l'addition de DMC aux amas Li+-EC et Li+-PC. Dans la plupart des cas, la substitution de solvant aux mélanges binaires sont défavorables. Dans le cas de mélanges ternaires, la molécule DMC ne peut pas remplacer EC et PC, tandis que PC peut facilement remplacer EC et DMC. Notre étude montre que PC tend à substituer EC dans la couche de solvation. Nous avons complété nos études ab-initio par des simulations MD d'une ion Li immergé dans les solvants purs et dans des mélanges de solvants d'intérêt pour les batteries, EC:DMC(1: 1) et EC:DMC:PC(1:1:3). MD est un outil très puissant et nous a permis de clarifier la pertinence des structures découvertes par DFT lorsque le ion est entouré par des solvants mélangés. En effet,la DFT fournit des informations sur les structures les plus stables de groupes isolés, mais aucune information sur leur stabilité ou de la multiplicité (entropie) lorsqu'il est immergé dans un environnement solvant infinie. Les données MD, ainsi que les calculs DFT nous ont permis de donner une image très complète de la structure locale de mélanges de solvants autour le ion lithium, sensiblement amélioré par rapport aux travaux précédents. / Due to the increasing global energy demand, eco-friendly and sustainable green resources including solar, or wind energies must be developed, in order to replace fossil fuels. These sources of energy are unfortunately discontinuous, being correlated with weather conditions and their availability is therefore strongly fluctuating in time. As a consequence, large-scale energy storage devices have become fundamental, to store energy on long time scales with a good environmental compatibility. Electrochemical energy conversion is the key mechanism for alternative power sources technological developments. Among these systems, Lithium-ion (Li+) batteries (LIBs) have demonstrated to be the most robust and efficient, and have become the prevalent technology for high-performance energy storage systems. These are widely used as the main energy source for popular applications, including laptops, cell phones and other electronic devices. The typical LIB consists of two (negative and positive) electrodes, separated by an electrolyte. This plays a very important role, transferring ions between the electrodes, therefore providing the electrical current. This thesis work focuses on the complex materials used as electrolytes in LIBs, which impact Li-ion transport properties, power densities and electrochemical performances. Usually, the electrolyte consists of Li-salts and mixtures of organic solvents, such as cyclic or linear carbonates. It is therefore indispensable to shed light on the most important structural (coordination) properties, and their implications on transport behaviour of Li+ ion in pure and mixed solvent compositions. We have performed a theoretical investigation based on combined density Functional Theory (DFT) calculations and Molecular Dynamics (MD) simulations, and have focused on three carbonates, cyclic ethylene carbonate (EC) and propylene carbonate (PC), and linear dimethyl carbonate (DMC). DFT calculations have provided a detailed picture for the optimized structures of isolated carbonate molecules and Li+ ion, including pure clusters Li+(S)n (S=EC, PC, DMC and n=1-5), mixed binary clusters, Li+(S1)m(S2)n (S1, S2 =EC, PC, DMC, with m+n=4), and ternary clusters Li+(EC)l(DMC)m(PC)n with l+m+n=4. Pure solvent clusters were also studied including the effect of PF6- anion. We have investigated in details the structure of the coordination shell around Li+ for all cases. Our results show that clusters such as Li+(EC)4, Li+(DMC)4 and Li+(PC)3 are the most stable, according to Gibbs free energy values, in agreement with previous experimental and theoretical studies. The calculated Gibbs free energies of reactions in binary mixtures suggest that the addition of EC and PC molecules to the Li+-DMC clusters are more favourable than the addition of DMC to Li+-EC and Li+-PC clusters. In most of the cases, the substitution of solvent to binary mixtures are unfavourable. In the case of ternary mixtures, the DMC molecule cannot replace EC and PC, while PC can easily substitute both EC and DMC molecules. Our study shows that PC tends to substitute EC in the solvation shell. We have complemented our ab-initio studies by MD simulations of a Li-ion when immersed in the pure solvents and in particular solvents mixtures of interest for batteries applications, e.g. , EC:DMC (1:1) and EC:DMC:PC(1:1:3). MD is a very powerful tool and has allowed us to clarify the relevance of the cluster structures discovered by DFT when the ion is surrounded by bulk solvents. Indeed, DFT provides information about the most stable structures of isolated clusters but no information about their stability or multiplicity (entropy) when immersed in an infinite solvent environment. The MD data, together the DFT calculations have allowed us to give a very comprehensive picture of the local structure of solvent mixtures around Lithium ion, which substantially improve over previous work.

En raison de l'augmentation de la demande d'énergie, ressources écologiques respectueux de l'environnement et durables (solaires, éoliennes) doivent être développées afin de remplacer les combustibles fossiles. Ces sources d'énergie sont discontinues, étant corrélés avec les conditions météorologiques et leur disponibilité est fluctuant dans le temps. En conséquence, les dispositifs de stockage d'énergie à grande échelle sont devenus incontournables, pour stocker l'énergie sur des échelles de temps longues avec une bonne compatibilité environnementale. La conversion d'énergie électrochimique est le mécanisme clé pour les développements technologiques des sources d'énergie alternatives. Parmi ces systèmes, les batteries Lithium-ion (LIB) ont démontré être les plus robustes et efficaces et sont devenus la technologie courante pour les systèmes de stockage d'énergie de haute performance. Ils sont largement utilisés comme sources d'énergie primaire pour des applications populaires (ordinateurs portables, téléphones cellulaires, et autres). La LIB typique est constitué de deux électrodes, séparés par un électrolyte. Celui-ci joue un rôle très important dans le transfert des ions entre les électrodes fournissant la courante électrique. Ce travail de thèse porte sur les matériaux complexes utilisés comme électrolytes dans les LIB, qui ont un impact sur les propriétés de transport du ion Li et les performances électrochimiques. Habituellement l'électrolyte est constitué de sels de Li et de mélanges de solvants organiques, tels que les carbonates cycliques ou linéaires. Il est donc indispensable de clarifier les propriétés structurelles les plus importantes, et leurs implications sur le transport des ions Li+ dans des solvants purs et mixtes. Nous avons effectué une étude théorique basée sur la théorie du fonctionnelle densité (DFT) et la dynamique moléculaire (MD), et nous avons consideré des carbonates cyclique (carbonate d'éthylène, EC, et carbonate de propylène, PC) et le carbonate de diméthyle, DMC, linéaire. Les calculs DFT ont fourni une image détaillée des structures optimisées de molécules de carbonate et le ion Li+, y compris les groupes pures Li+(S)n (S =EC,PC,DMC et n=1-5), groupes mixtes binaires, Li+(S1)m(S2)n (S1,S2=EC,PC,DMC, m+n=4), et ternaires Li+(EC)l(DMC)m(PC)n (l+m+n=4). L'effet de l'anion PF6 a également été étudié. Nous avons aussi étudié la structure de la couche de coordination autour du Li+, dans tous les cas. Nos résultats montrent que les complexes Li+(EC)4, Li+(DMC)4 et Li+(PC)3 sont les plus stables, selon les valeurs de l'énergie libre de Gibbs, en accord avec les études précédentes. Les énergies libres de réactions calculés pour les mélanges binaires suggèrent que l'ajout de molécules EC et PC aux clusters Li+ -DMC sont plus favorables que l'addition de DMC aux amas Li+-EC et Li+-PC. Dans la plupart des cas, la substitution de solvant aux mélanges binaires sont défavorables. Dans le cas de mélanges ternaires, la molécule DMC ne peut pas remplacer EC et PC, tandis que PC peut facilement remplacer EC et DMC. Notre étude montre que PC tend à substituer EC dans la couche de solvation. Nous avons complété nos études ab-initio par des simulations MD d'une ion Li immergé dans les solvants purs et dans des mélanges de solvants d'intérêt pour les batteries, EC:DMC(1: 1) et EC:DMC:PC(1:1:3). MD est un outil très puissant et nous a permis de clarifier la pertinence des structures découvertes par DFT lorsque le ion est entouré par des solvants mélangés. En effet,la DFT fournit des informations sur les structures les plus stables de groupes isolés, mais aucune information sur leur stabilité ou de la multiplicité (entropie) lorsqu'il est immergé dans un environnement solvant infinie. Les données MD, ainsi que les calculs DFT nous ont permis de donner une image très complète de la structure locale de mélanges de solvants autour le ion lithium, sensiblement amélioré par rapport aux travaux précédents. / Due to the increasing global energy demand, eco-friendly and sustainable green resources including solar, or wind energies must be developed, in order to replace fossil fuels. These sources of energy are unfortunately discontinuous, being correlated with weather conditions and their availability is therefore strongly fluctuating in time. As a consequence, large-scale energy storage devices have become fundamental, to store energy on long time scales with a good environmental compatibility. Electrochemical energy conversion is the key mechanism for alternative power sources technological developments. Among these systems, Lithium-ion (Li+) batteries (LIBs) have demonstrated to be the most robust and efficient, and have become the prevalent technology for high-performance energy storage systems. These are widely used as the main energy source for popular applications, including laptops, cell phones and other electronic devices. The typical LIB consists of two (negative and positive) electrodes, separated by an electrolyte. This plays a very important role, transferring ions between the electrodes, therefore providing the electrical current. This thesis work focuses on the complex materials used as electrolytes in LIBs, which impact Li-ion transport properties, power densities and electrochemical performances. Usually, the electrolyte consists of Li-salts and mixtures of organic solvents, such as cyclic or linear carbonates. It is therefore indispensable to shed light on the most important structural (coordination) properties, and their implications on transport behaviour of Li+ ion in pure and mixed solvent compositions. We have performed a theoretical investigation based on combined density Functional Theory (DFT) calculations and Molecular Dynamics (MD) simulations, and have focused on three carbonates, cyclic ethylene carbonate (EC) and propylene carbonate (PC), and linear dimethyl carbonate (DMC). DFT calculations have provided a detailed picture for the optimized structures of isolated carbonate molecules and Li+ ion, including pure clusters Li+(S)n (S=EC, PC, DMC and n=1-5), mixed binary clusters, Li+(S1)m(S2)n (S1, S2 =EC, PC, DMC, with m+n=4), and ternary clusters Li+(EC)l(DMC)m(PC)n with l+m+n=4. Pure solvent clusters were also studied including the effect of PF6- anion. We have investigated in details the structure of the coordination shell around Li+ for all cases. Our results show that clusters such as Li+(EC)4, Li+(DMC)4 and Li+(PC)3 are the most stable, according to Gibbs free energy values, in agreement with previous experimental and theoretical studies. The calculated Gibbs free energies of reactions in binary mixtures suggest that the addition of EC and PC molecules to the Li+-DMC clusters are more favourable than the addition of DMC to Li+-EC and Li+-PC clusters. In most of the cases, the substitution of solvent to binary mixtures are unfavourable. In the case of ternary mixtures, the DMC molecule cannot replace EC and PC, while PC can easily substitute both EC and DMC molecules. Our study shows that PC tends to substitute EC in the solvation shell. We have complemented our ab-initio studies by MD simulations of a Li-ion when immersed in the pure solvents and in particular solvents mixtures of interest for batteries applications, e.g. , EC:DMC (1:1) and EC:DMC:PC(1:1:3). MD is a very powerful tool and has allowed us to clarify the relevance of the cluster structures discovered by DFT when the ion is surrounded by bulk solvents. Indeed, DFT provides information about the most stable structures of isolated clusters but no information about their stability or multiplicity (entropy) when immersed in an infinite solvent environment. The MD data, together the DFT calculations have allowed us to give a very comprehensive picture of the local structure of solvent mixtures around Lithium ion, which substantially improve over previous work.

Des films nanoporeux de dioxyde de titane peuvent être formés au-dessus d’un substrat de titane par oxydation anodique, c’est-à-dire lorsque le métal est plongé dans un électrolyte fluoré et est soumis à une tension ou une densité de courant constante. Le dopage par des ions fluorures permet également d’améliorer les propriétés photocatalytiques de TiO2 utiles pour de nombreuses applications. L’objectif de cette thèse consiste à étudier, par la théorie de la fonctionnelle de la densité (DFT), l’influence des ions fluorures sur les mécanismes élémentaires ayant lieu lors de la croissance des couches de TiO2. Nous nous attacherons notamment à comprendre l’influence de la phase et de l’orientation cristallographique des surfaces de TiO2 fluoré, sur leurs stabilités et leurs propriétés photocatalytiques. Tous les calculs DFT ont été effectués avec le logiciel VASP qui permet de simuler des systèmes périodiques. Dans un premier temps, nous avons étudié l’adsorption de F et de F2 sur différentes surfaces de TiO2 parfaites et défectueuses. Deux groupes de surfaces ont pu être distingués vis-à-vis de leur réactivité et de leurs propriétés. Nous nous sommes ensuite intéressés au F-dopage à et sous différentes surfaces. Cette étude est inédite car le dopage a été essentiellement étudié au niveau du massif de TiO2. L’influence de l’hydroxylation des surfaces sur la réactivité de TiO2 vis-à-vis du fluor a été étudié afin d’obtenir des informations sur la croissance des couches d’oxyde. / Nanoporous titanium dioxide films can be formed over a titanium substrate by anodic oxidation, ie, when the metal is immersed in a fluorinated electrolyte and a constant voltage or a current density is applied. Doping with fluoride ions also improve the photocatalytic properties of TiO2 which is crucial for many applications. The goal of this thesis is to study the influence of fluoride ions on the elementary mechanisms during the growth of the TiO2 layers by the density functional theory (DFT). In particular, we will try to understand the influence of the phase and the crystallographic orientation of fluorinated TiO2 surfaces on their stability and their photocatalytic properties. All DFT calculations were carried out using VASP software which allows to simulate periodic systems. In a first step, we studied the adsorption of F and F2 on different perfect and defective TiO2 surfaces. Two groups of surfaces have been distinguished from their reactivity and their properties. Then we focused on F-doping at and under different surfaces. This study is pioneering because F-doping was mainly studied in bulk TiO2. The influence of surface hydroxylation on the reactivity of TiO2 towards fluoride ions has been studied in order to obtain information on the growth of the oxide layers.

Neste trabalho foi feito o estudo teórico de complexos de rutênio de fórmula: cis-[RuCl2(NN)(PP)], [Ru(NS)(bipy)(PP)]+, [Ru(pic)(bipy)(PP)]+ e [RuCl(bCN)(bipy)(PP)]+, onde NN = 2,2\'-bipiridina (bipy), 4,4\'-dimetil-2,2\'-bipiridina (Mebipy) e 1,10-fenantrolina (phen); PP = 1,1-Bis(difenilfosfino)metano (dppm), 1,2-Bis(difenilfosfino)etano (dppe), 1,3-Bis(difenilfosfino)propano (dppp), 1,4-Bis(difenilfosfino)butano (dppb) e 1,1\'-Bis(difenilfosfino)ferroceno (dppf), NS = 2-mercaptopiridina (pyS), 2-mercaptopirimidina (prm), 4,6-dimetil-2-mercaptopirimidina (dmpm); pic = 2-picolinato; bCN = benzonitrila em nível da teoria DFT utilizando o funcional B3LYP. Os complexos tiveram suas estruturas otimizadas, as quais foram comparadas às estruturas obtidas experimentalmente por difração de raios X e apresentaram grande concordância com estas. A análise de decomposição de carga revelou que os ligantes NN, PP, NS, pic, bCN e Cl doam elétrons para o centro metálico tornando-o mais negativo, isso reflete na energia dos orbitais HOMO, que por sua vez pode ser relacionado com os valores dos potencias de oxidação dos complexos. Pode-se observar que quanto maior o valor do potencial de oxidação, menor a energia do HOMO. Além disso, nos complexos cis-[RuCl2(NN)(PP)] os valores de potenciais de oxidação são mais anódicos porque os dois átomos de cloro somados doam mais para o rutênio que os ligantes NS e pic e bCN. Os espectros eletrônicos foram calculados e as bandas observadas em 420 e 290 nm nos complexos [Ru(NS)(NN)(PP)]+ puderam ser atribuídas a transições dos tipos MLCT e LLCT devido a análise da composição dos orbitais. As bandas de baixa intensidade observadas nos complexos cis-[RuCl2(NN)(PP)] em torno de 460 nm, mostraram ser transições do tipo MLCT do HOMO-2 para o LUMO, sendo que esses orbitais estão concentrados nos átomos de rutênio e nos ligantes NN, respectivamente. Os espectros vibracionais calculados estão de acordo com os dados experimentais. As bandas observadas em torno de 1380, 1160 e 760 cm-1 foram atribuidas aos estiramentos vC-S e as bandas em 1715 cm-1 ao estiramento νC=O; no ligante livre essa última banda é observada em 1654 cm-1. A banda referente ao estiramento vC≡N, que aparece em 2229 cm-1 na benzonitrila não coordenada, nos complexos [RuCl(bCN)(NN)(PP)]+ aparece em 2250 cm-1. / Ruthenium complexes with formulas: cis-[RuCl2(NN)(PP)], [Ru(NS)(bipy)(PP)]+, [Ru(pic)(bipy)(PP)]+ e [RuCl(bCN)(bipy)(PP)]+, where NN = 2,2\'-bipyridine (bipy), 4,4\'-dimethyl-2,2\'-bipyridine (Mebipy) e 1,10-phenanthroline (phen); PP = 1,1-Bis(diphenylphosphino)methane (dppm), 1,2- Bis(diphenylphosphino)ethano (dppe), 1,3- Bis(diphenylphosphino)propane(dppp), 1,4-Bis(diphenylphosphino)buthane (dppb) e 1,1\'-Bis(diphenylphosphino)ferrocene (dppf), NS = 2-mercaptopyridine (pyS), 2-mercaptopyrimidine (prm), 4,6-dimethyl-2-mercaptopyrimidina (dmpm); pic = 2-picolinate; bCN = benzonitrile were studied in this thesis applying DFT theory with B3LYP functional. The complexes had their structures optimized and comparison with experimental structures determined by X ray crystallography show that they compare well. Charge decomposition analysis revealed that NN, PP, NS, pic, bCN and Clligands donate eletrons to the metal center makig it more negative, what reflects in the HOMO energies, that can be related to oxidation potencials of the complexes. The higher the oxidation potential of the complexes the lower the energy of HOMO. In complexes cis-[RuCl2(NN)(PP)] the oxidation potentials were more anodic because the two chlorine ligands togheterdonate more than NS, pic or bCN ligands. Electronic spectra of the complexes [Ru(NS)(NN)(PP)]+ were calculated and the bands around 420 and 290 nm could be assigned to transitions of the type MLCT, LLCT based on the analysis of the orbitals composition. The bands of low intensity around 460 nm observed in the spectra of cis-[RuCl2(NN)(PP)] were also MLCT transitions from HOMO-2 to LUMO, these orbitals are concentrated in ruthenium and NN ligands, respectively. The calculated vibrational spectra are in good agreement with experimental data, bands around 1380, 1160 e 760 cm-1 were assigned to vC-S stretching and bands around 1715 cm-1 to νC=O, in pic not coordinated this band appears at 1654 cm-1. In freebenzonitrile vC≡N band is observed at 2229 cm-1, but in complexes [RuCl(bCN)(NN)(PP)]+ this band is upshifted to 2250 cm-1.

Nesta tese investigamos a partir de cálculos de primeiros princípios, dispositivos e componentes de dispositivos baseados em grafeno. Abordamos os campos da nanoeletrônica e da spintrônica. Dentro da nanoeletrônica investigamos: (i) propriedades de transporte de um nanotransistor de bicamada de grafeno na presença de um gate duplo. Demonstramos que sobre a ação de um campo elétrico externo, mesmo utilizando um gate da ordem de 10 nm, à temperatura ambiente e 4.5K uma corrente nula nunca é exibida. Esses resultados são explicados por um regime de tunelamento; (ii) propriedades eletrônicas e de transporte de multicamadas de grafeno em função do número de camadas e tipo de empilhamento entre elas. Mostramos que a estrutura eletrônica do sistema depende fortemente desse novo grau de liberdade de empilhamento. Na presença de um campo elétrico externo aplicado perpendicular ao sistema, o empilhamento do tipo Bernal nunca exibe um gap de energia, ao contrário do empilhamento romboédrico que exige um gap ajustável através da intensidade do campo. Mostramos também que é possível diferenciar os tipos de empilhamentos através da resistência do sistema e variando-se a temperatura; (iii) dentro das componentes de um nanotransistor mais realista, estudamos as propriedades eletrônicas e estruturais de: (a) bicamadas de grafeno sobre um substrato de nitreto de boro hexagonal. Neste sistema o limite de voltagens que podem ser aplicadas depende fortemente do número de camadas de h-BN e da direção do campo, onde quanto menos camadas maior é a voltagem que pode ser aplicada; (b) heteroestruturas compostas de bicamadas de grafeno, nitreto de boro hexagonal e cobre. Demonstramos que para uma aplicação direta em um dispositivo a configuração com uma bicamada de grafeno depositada sobre um substrato de h-BN e este conjunto sobre a superfície de cobre é a mais favorável. Nessa configuração é possível tanto controlar o gap na bicamada como a dopagem do sistema, sem a abertura de canais de condução através do dielétrico (h-BN). Dentro do campo da spintrônica estudamos: (i) propriedades de transporte das nanofitas de grafeno (GNR) (3,0) pristinas e dopadas com boro e nitrogênio. Para as GNR pristinas mostramos com os eletrodos em um alinhamento de spin anti-paralelo o sistema apresenta um comportamento de filtro de spin, onde para tensões de bias positivos/negativos somente o canal up/down conduz. Para as GNR dopadas com boro e nitrogênio, mostramos que as correntes para os diferentes canais de spin são não degeneradas ao longo de todo o intervalo de tensões aplicadas, apresentando desse modo um comportamento de filtro de spin; (ii) finalmente estudamos as propriedades de transporte de uma junção túnel magnética, composta de GNR intercaladas por uma nanofita de nitreto de boro hexagonal. Mostramos que esse sistema pode ser utilizado tanto como filtros de spin como elementos para dispositivos de magnetoresistência gigante, onde para este último a sua eficiência é muito mais pronunciada. / In this thesis we investigated by first principle calculations, devices and components of devices based on graphene. We covered the fields of nanoelectronics and spintronics. On the field of nanoelectronics we investigated: (i) the transport properties of a dual gate bilayer graphene nanotransistor. We showed that under the action of an external electrical field, even with a gate length of 10 nm, at room temperature and 4.5K a zero current is never exhibited. These results could be explained by a tunneling regime; (ii) the electronic and transport properties of few layer graphene, as function of the number and type of stacking of the layers. We show that the electronic structure strong deppends of the stacking order. On the presence of a external electrical field applied to the system, the Bernal stacking never presents a gap, contrary to the rombohedrical one, that posses a tuneable energy gap. Also we showed that is possible to differentiate the types of stacking by the resistance of the system and varying the temperature;(iii) for the components of a more realistic nanodevice, we study the structural and electronic properties of: (a) bilayer graphene over a hexagonal boron nitride substrate. We show that the voltages that could be applied to the system strongly depends of the number 0 layers and the direction of the field, where with more layers, smaller is the field; (b) heterostructures composed with bilayer graphene, hexagonal boron nitride and cooper. We show that for a direct application on a device, the better configuration is with a bilayer graphene over the hexagonal boron nitride, and this set over a cooper. In this configuration is possible to control both the gap and the doping of the system, without the creation of conducting channels through the dielectric (h-BN). On the field of spintronics, we study: (i) the transport properties (3,0) graphene nanoribbons pristines and doped with nitrogen and boron. For the pristine GNR we show that for the electrodes in an anti-parallel alignment the system presents a spin filter behavior, where for positive/negative bias the transport is only by up/down channel. For the GNR doped with nitrogen and boron we show that the current is non-degenerated in all range of voltages applied, presenting a spin filter behavior; (ii) finally, we study the transport properties of a magnetic tunnel junction, consisting of a GNR intercalated with a hexagonal boron nitride nanoribbon. We show that such system could be used both as a spin filter as a device that uses the the giant magnetoresistance effect, where for the last the system if more efficient.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / A transmissÃo de impulsos nervosos à feita por meio das sinapses, envolvendo neurotransmissores e receptores. Os receptores ionotrÃpicos glutamatÃrgicos (GluRs) sÃo importantes canais iÃnicos do sistema nervoso central, encontrados em sinapses de excitaÃÃo rÃpida, e estÃo relacionados a funÃÃes cerebrais importantes como aprendizado e memÃria. AlÃm disso, os GluRs tambÃm estÃo associados com certas doenÃas neurolÃgicas e psiquiÃtricas, como por exemplo: a doenÃa de Alzheimer, o mal de Parkinson, a epilepsia, o acidente vascular cerebral, a esclerose lateral amiotrÃfica e a esquizofrenia. Neste trabalho, tiramos vantagem dos dados disponÃveis na literatura da co-cristalizaÃÃo dos seguintes agonistas glutamato (C5H9NO4) e AMPA (C7H10N2O4), do agonista parcial cainato (C10H15NO4) e do antagonista DNQX (C8H2N4O6) com o receptor GluR2 com resoluÃÃo de 1,9 Ã, 1,7 Ã, 1,9 à e 1,8 Ã, respectivamente, para estudar a interaÃÃo destes quatro ligantes com a GluR2 por meio de mÃtodos computacionais ab initio. Os hidrogÃnios ausentes dos dados de difraÃÃo de raios-X foram colocados atravÃs de um processo semi-clÃssico de minimizaÃÃo da energia total GluR2-ligante. A seguir, as simulaÃÃes foram feitas usando a Teoria do Funcional da Densidade (DFT), tanto ao nÃvel da aproximaÃÃo da densidade local (LDA), como da aproximaÃÃo do gradiente generalizado (GGA), para descriÃÃo dos efeitos de troca e correlaÃÃo. A utilizaÃÃo do mÃtodo de fragmentaÃÃo molecular com capas conjugadas (MFCC) tornou possÃvel analisar a interaÃÃo dos ligantes com cada um dos resÃduos prÃximos e pÃs-prÃximos do GluR2. Considerou-se tambÃm a relevÃncia da blindagem dos resÃduos pÃs-prÃximos que interagem com os ligantes, bem como se fez uma anÃlise da energia de interaÃÃo dos resÃduos (prÃximos e pÃs-prÃximos) considerados com os Ãtomos dos ligantes (resultados apresentados nos grÃficos BIRD), sem e com mediaÃÃo das molÃculas de Ãgua existentes no sÃtio de ligaÃÃo (o que permite se determinar ab initio a relevÃncia da Ãgua na energÃtica da interaÃÃo ligante-GluR2). Obteve-se a energia total de interaÃÃo GluR2-ligante em funÃÃo da distÃncia dos centroides dos ligantes aos resÃduos, o que permitiu correlacionÃ-la à intensidade de ativaÃÃo e antagonismo dos neurotransmissores em questÃo. Demonstrou-se que ela segue a ordem AMPA > glutamato > cainato > DNQX somente quando um raio do sÃtio de ligaÃÃo suficientemente grande à considerado, o que explica dados experimentais publicados sobre a ativaÃÃo e antagonismo do receptor glutamatÃrgico GluR2, sugerindo que os resÃduos pÃs-prÃximos podem ser importantes para determinar o funcionamento do receptor. Para o glutamato, os resultados obtidos indicam que os resÃduos atrativos mais relevantes sÃo: Arg485, Lys730 (mediado pela Ãgua W39), Ser654, Leu650 mediado por W69, e Lys656 mediado por W22; os resÃduos repulsivos mais relevantes para o glutamato sÃo Glu402 (pÃs-prÃximo) mediado por W36, Glu657 e Asp651 (pÃs-prÃximos). Para o AMPA os resÃduos atrativos mais relevantes sÃo: Arg485, Thr655 mediado por W134, Lys730 mediado por W137, Lys656 mediado por W138, Lys449 e Arg684 (pÃs-prÃximos); os resÃduos repulsivos mais relevantes para o AMPA sÃo Glu402 mediado por W3, Asp651 mediado por W96 e W139 (pÃs-prÃximo), e Glu657 (pÃs-prÃximo) mediado por W140. Para o cainato os resÃduos atrativos mais relevantes sÃo Arg485, Ser654, Thr655 e Arg684 (pÃs-prÃximo); os resÃduos repulsivos mais relevantes para o Cainato sÃo Glu402, Glu657 mediado por W78 (pÃs-prÃximo) e Asp651. Para o DNQX os resÃduos atrativos mais relevantes sÃo Arg485, Glu705 e Tyr450 mediado por W26 e W137; o resÃduo repulsivo mais relevante para o DNQX à Leu498. Uma plÃiade de perspectivas relacionadas aos resultados obtidos reluz e dentre elas podemos destacar a possibilidade do desenvolvimento de agonistas e antagonistas glutamatÃrgicos com especificidades voltadas à diminuiÃÃo de efeitos colaterais quando utilizados no tratamento de doenÃas relacionadas à neurotransmissÃo glutamatÃrgica. / The transmission of nerve impulses occurs through the synapses, involving neurotransmitters and receptors. The ionotropic glutamate receptors GluRs are important ionic channels of the central nervous system, founded in rapid excitation synapses, and related to important cerebral functions like learning and memory. Besides this, GluRs are also associated with important neurological and psychiatric diseases like Alzheimer, Parkinson, epilepsy, cerebral ischemia, amyotrophic lateral sclerosis, and schizophrenia. In this work, we take advantage of the available data in the literature of co-crystallization of the following full agonists glutamate (C5H9NO4) and AMPA (C7H10N2O4), the partial agonist kainate (C10H15NO4) and the antagonist DNQX (C8H2N4O6) with the GluR2 receptor with resolution of 1.9 Ã, 1.7 Ã, 1.9 à and 1.8 Ã, respectively to study the interaction of these four ligands with GluR2 by means of ab initio computational methods. The absent hydrogens in the GluR2-ligand X-ray diffraction data were inserted through a semi-classical total energy minimization process. Next, the simulations were performed within the scope of the Density Functional Theory (DFT), both in the local density approximation (LDA) as generalized gradient approximation (GGA) for the description of exchange-correlation effects. The use of the molecular fragmentation method with conjugated caps (MFCC) allowed to analyze the interaction between the ligands with each one close and next-closed GluR2 residues. It was also considered the relevance of the screening of the next-closed residues with interact with the ligands, and it was performed an analysis of the interaction energy between the focused residues (close and next-closed) with the atoms of the ligands (results depicted in the BIRD panels), without and with the mediation of water molecules existing in the binding pocket (which allows to determine ab initio the relevance of water in the GluR2-ligands energetic). It was obtained the GluR2-ligand total energy interaction as a function of the distance between the ligand centroid and the residues, which allowed to correlate it to the activation strength and antagonism of the ligands focused. It was demonstrated that it follows the order AMPA > glutamate > kainite > DNQX only when a large enough binding pocket radius is taken into account, explaining the experimental data published on the activation and antagonism of the glutamatergic receptor GluR2 and suggesting the next-closed residues can be important to determine the receptor functioning. For the glutamate, the obtained results point that the most important attractive residues are Arg485, Lys730 (water W39 mediated), Ser654, Leu650 (water W69 mediated), and Lys656 (water W22 mediated); the most important repulsive residues for the glutamate are Glu402 (next-closed water W36 mediated), Glu657 and Asp651 (nex-closed). For AMPA, the most important attractive residues are Arg485, Thr655 (water W134 mediated), Lys730 (water W137 mediated), Lys656 (water W138 mediated), Lys449 and Arg684 (next-closed); the most important repulsive residues for AMPA are Glu402 (water W3 mediated), Asp651 (next-closed, water W96 and W139 mediated), and Glu657 (next-closed, water W140 mediated). For kainate the most important attractive residues are Arg485, Ser654, Thr655 and Arg684 (next-closed); the most important repulsive residues for kainite are Glu402, Glu657 (next-closed, water W78 mediated) and Asp651. For DNQX, the most important attractive residues are Arg485, Glu705 and Tyr450 (water W26 and W137 mediated); the most important repulsive residue for DNQX is Leu498. A pleiade of perspectives related with the obtained results shines, among which one can highlight the possibility to develop glutamatergic agonists and antagonists with specificities related to decrease side effects when used in the treatment of maladies related with the glutamatergic neurotransmission.

Submitted by Irene Nascimento (irene.kessia@ufpe.br) on 2016-07-21T15:56:16Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Machado CMB_Configuração Absoluta e Deslocamentos Químicos de Moléculas Organicas por GIAO DFT.pdf: 4055624 bytes, checksum: 78784f0f10b901b4372dd2e2a038b173 (MD5) / Made available in DSpace on 2016-07-21T15:56:16Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Machado CMB_Configuração Absoluta e Deslocamentos Químicos de Moléculas Organicas por GIAO DFT.pdf: 4055624 bytes, checksum: 78784f0f10b901b4372dd2e2a038b173 (MD5) Previous issue date: 2016-04-06 / Facepe / Nesta dissertação, estudamos quatro moléculas quirais através do emprego de quatro funcionais da densidade diferentes com vinte funções de base para encontrar uma combinação que melhor descrevesse o ângulo de rotação óptica destas moléculas. Analisamos também como a ligação de hidrogênio pode influenciar no ângulo de rotação óptica e quais seriam os efeitos dessa ligação de hidrogênio nas propriedades de diferentes sistemas envolvendo compostos nitrogenados cíclicos e ácidos carboxílicos. Os funcionais empregados foram: B3LYP, PBE, PBE0 e ωB97XD; e as funções de base foram: 6-31G, 6-31+G, 6-31++G, 6-31G*, 6-31+G*, 6-31++G*, 6-31G**, 6-31+G**, 6-31++G**, cc-pVDZ, cc-pVTZ, cc-pVQZ, aug-cc-pVDZ, aug-cc-pVTZ, aug-cc-pVQZ, def2-TZVP, def2-TZVPD, def2-TZVPP, def2-TZVPPD e 6-311+G**. As moléculas quirais estudadas para a atribuição da configuração absoluta foram a R,R-dimetiloxirano, S-metiloxirano, S-halotano e (S)(+)4-fenil-2-oxazolidinona. As moléculas estudadas para a espectroscopia de RMN de 1H e cálculos GIAO foram a 1,10-fenantrolina, 2,2-bipiridina, 4,4’-difenil-2,2’-bipiridina e os ácidos carboxílicos: fórmico, acético e benzoico. As geometrias de todas as moléculas e complexos foram otimizadas para que não houvesse nenhuma frequência harmônica imaginária e, a partir delas, o ângulo de rotação óptica foi calculado. Para a molécula (S)(+)4-fenil-2-oxazolidinona, foi observada a possibilidade de formação de dímeros que podem influenciar no valor do ângulo de rotação óptica. As conformações possíveis para esse dímero foram encontradas utilizando mecânica molecular (MMFF - Merck Molecular Force Field) pelo programa SPARTAN 14, tendo suas geometrias sido na sequência otimizadas. As energias de estabilização de todos os complexos foram corrigidas levando em consideração os erros de sobreposição do conjunto de base (BSSE) e da energia do ponto zero (ZPE). O cálculo do ângulo de rotação óptica foi feito com o uso da média de Boltzmann. Todas as moléculas quirais foram estudadas no modo isolado e em clorofórmio através do método PCM. Observamos que moléculas simples, rígidas e sem possibilidade de formação de ligação de hidrogênio, não precisam de funcionais sofisticados para a otimização de sua geometria, mas precisam destes para o cálculo do ângulo de rotação óptica. Para moléculas flexíveis, é necessário um funcional mais sofisticado também para a otimização de geometria. Assim, a melhor combinação de método e base para moléculas mais rígidas para a otimização de geometria foi ωB97XD com a base 6-31G**. Já para moléculas flexíveis, para a otimização de geometria, a melhor base foi def2-TZVP. Em ambos os casos, para o cálculo do ângulo de rotação óptica, a melhor combinação foi ωB97XD com a base def2-TZVP. Para os complexos de ligação de hidrogênio estudados, o deslocamento químico do núcleo do hidrogênio ácido nos complexos envolvendo a 1,10-fenantrolina, como aceitador de elétrons foi mais fortemente afetado em comparação aos complexos correspondentes da 2,2’-bipiridina e 4,4-difenil-2,2’-bipiridina. Os cálculos GIAO estão de acordo com os experimentos de RMN de 1H na previsão da blindagem eletrônica do hidrogênio ácido durante a ligação de hidrogênio existente nos complexos estudados. / In this dissertation, we studied four chiral molecules using four distinct density functionals, and twenty basis functions, searching for a combination that could best describe the optical rotation angle of these molecules. We also analyzed how hydrogen bonding influences the optical rotation angle and what are the effects of hydrogen bonding on the properties of different systems involving cyclic nitrogenated compounds and carboxylic acids. The functionals employed were B3LYP, PBE, PBE0 and ωB97XD. The basis functions employed were: 6-31G, 6-31+G, 6-31++G, 6-31G* 6-31+G*, 6-31++G*, 6-31G**, 6-31+G**, 6-31++G**, cc-pVDZ, cc-pVTZ, cc-pVQZ, aug-cc-pVDZ, aug-cc-pVTZ, aug-cc-pVQZ, def2- TZVP, def2-TZVPD, def2-TZVPP, def2-TZVPPD and 6-311+G**. The chiral molecules studied for the assignment of their absolute configuration were R,R-dimethyloxirane, S-methyl oxirane, S-halothane and (S)(+)4-phenyl-2-oxazolidinone. The molecules studied for the GIAO calculation of 1H NMR spectroscopy were 1,10-phenanthroline, 2,2-bipyridine, 4,4'-diphenyl-2,2'-bipyridine and the carboxylic acids: formic, acetic and benzoic. The geometries of all molecules and complexes were optimized so that there were no imaginary harmonic frequencies and, from them, the optical rotation angle was calculated. For the (S)(+)4-phenyl-2-oxazolidinone molecule, the possibility of dimer formation was observed which can influence the value of the optical rotation angle. The possible conformations for this dimer were found using molecular mechanics by SPARTAN program, and their geometries optimized. The energies of stabilization of all complexes were corrected for the basis set superposition error (BSSE) and zero point energy (ZPE). The calculation of optical rotation angle was made using a Boltzmann average. All chiral molecules were studied, both isolated and in chloroform by the PCM method.We note that simple molecules, rigid and with no possibility of hydrogen bond formation, do not need sophisticated functionals for geometry optimization. However, these sophisticated functionals are indeed needed for the optical rotation angle to be calculated. For flexible molecules, more sophisticated functionals are also necessary for the geometry optimization step. Thus, the best combination method and basis set for the geometry optimization of more rigid molecules found was ωB97XD to the base 6-31G **. As for the geometry optimization of flexible molecules, the best basis set was def2-TZVP. On the other hand, for calculating the optical rotation angle in both cases, the best combination found was ωB97XD with the def2-TZVP.For the hydrogen bonding complexes studied, the chemical shift of the acidic hydrogen nucleus in the complex involving 1,10-phenanthroline as electron acceptor was more strongly affected compared to the corresponding complexes of 2,2'-bipyridine and 4,4'-diphenyl-2,2'-bipyridine. The GIAO calculations agree with the NMR 1H experiments in predicting the electronic shielding of the acidic hydrogen during hydrogen bonding in the studied complexes.