La mémoire NAND Flash représente une part importante dans le marché des circuits intégrés et a bénéficié de la traditionnelle miniaturisation de l’industrie des sémiconducteurs lui permettant un niveau d’intégration élevé. Toutefois, cette miniaturisation semble poser des sérieux problèmes au-delà du noeud 22 nm. Dans un souci de dépasser cette limite, des solutions mémoires alternatives sont proposées parmi lesquelles la mémoire résistive (RRAM) se pose comme un sérieux candidat pour le remplacement de NAND Flash. Ainsi, dans cette thèse nous essayons de répondre à des nombreuses questions ouvertes sur les dispositifs RRAM à base d’oxyde d’hafnium (HfO2) en particulier en adressant le manque de compréhension physique détaillée sur leur fonctionnement et leur fiabilité. L’impact de la réduction de taille des RRAM, le rôle des électrodes et le processus de formation et de diffusion des défauts sont étudiés. L’impact de l’alliage/dopage de HfO2 avec d’autres matériaux pour l’optimisation des RRAM est aussi abordé. Enfin, notre étude tente de donner quelques réponses sur la formation du filament conducteur, sa stabilité et sa possible composition. / Among non-volatile memory technologies, NAND Flash represents a significant portion in the IC market and has benefitted from the traditional scaling of semiconductor industry allowing its high density integration. However, this scaling seems to be problematic beyond the 22 nm node. In an effort to go beyond this scaling limitation, alternative memory solutions are proposed among which Resistive RAM (RRAM) stands out as a serious candidate for NAND Flash replacement. Hence, in this PhD thesis we try to respond to many open questions about RRAM devices based on hafnium oxide (HfO2), in particular, by addressing the lack of detailed physical comprehension about their operation and reliability. The impact of scaling, the role of electrodes, the process of defects formation and diffusion are investigated. The impact of alloying/doping HfO2 with other materials for improved RRAM performance is also studied. Finally, our study attempts to provide some answers on the conductive filament formation, its stability and possible composition.

Knowledge of the dissociation constants of the ionizable protons of weak acids in aqueous media is of fundamental importance in many areas of chemistry and biochemistry. The pKa value, or equilibrium dissociation constant, of a molecule determines the relative concentration of its protonated and deprotonated forms at a specified pH and is therefore an important descriptor of its chemical reactivity. Considerable efforts have been devoted to the determination of pKa values by deferent experimental techniques. Although in most cases the determination of pKa values from experimental is straightforward, there are situations where interpretation is difficult and the results ambiguous. It is, therefore, not surprising that the capability to provide accurate estimates of the pKa value has been a central goal in theoretical chemistry and there has been a large effort in developing methodologies for predicting pKa values for a variety of chemical systems by differing quantum chemical techniques. A prediction accuracy within 0.5 pKa units of experiment is the desirable level of accuracy. This is a non-trivial exercise, for an error of 1 kcal/mol in estimates of the free energy value would result in an error of 0.74 pKa units. In this thesis ab initio Car-Parrinello molecular dynamics (CPMD) has been used for investigating the Brϕnsted acid-base chemistry of weak acids in aqueous solution. A key issue in any dissociation event is how the solvating water molecules arrange themselves spatially and dynamically around the neutral and dissociated acid molecule. Ab initio methods have the advantage that all solvent water molecules can, in principle, be con- sidered explicitly. One of the factors that has inhibited the widespread use of ab initio MD methods to study the dissociation reaction is that dissociation of weak acids are rare events that require extremely long simulation times before one is observed. The metady- namics formalism provides a solution to this conundrum by preventing the system from revisiting regions of configuration space where it has been in the past. The formalism allows the system to escape the free-energy minima by biasing the dynamics with a history dependent potential (or force) that acts on select degrees of freedom, referred to as collective variables. The bias potentials, modeled by repulsive inverted Gaussians that are dropped during propagation, drive the system out of any free-energy minima and allow it to explore the configuration space by a relatively quick and efficient sampling. The the- sis deals with a detailed investigation of the Brϕnsted acid-base chemistry of weak acids in aqueous solutions by the CPMD-metadynamics procedure. In Chapter 1, current approaches for the theoretical estimation of pKa values are summarized while in Chapter 2 the simulation methodology and the metadynamics sampling techniques used in thisstudy are described. The potential of the CPMD-metadynamics procedure to provide estimates of the acid dissociation constant (pKa) is explored in Chapter 3, using acetic acid as a test sys- tem. Using the bond-distance dependent coordination number of protons bound to the dissociating carboxylic groups as the collective variable, the free-energy profile for the dissociation reaction of acetic acid in water was computed. Convergence of the free-energy profiles and barriers for the simulations parameters is demonstrated. The free-energy profiles exhibit two distinct minima corresponding to the dissociated and neutral states of the acid and the deference in their values provides the estimate for pKa. The estimated value of pKa for acetic acid from the simulations, 4.80, is in good agreement with the experiment at value of 4.76. It is shown that the good agreement with experiment is a consequence of the cancellation of errors, as the pKa values are computed as the difference in the free energy values at the minima corresponding to the neutral and dissociated state. The chapter further explores the critical factors required for obtaining accurate estimates of the pKa values by the CPMD-metadynamics procedure. It is shown that having water molecules sufficient to complete three hydration shells as well as maintaining water density in the simulation cell as close to unity is important. In Chapter 4, the CPMD-metadynamics procedure described in Chapter-3 has been used to investigate the dissociation of a series of weak organic acids in aqueous solutions. The acids studied were chosen to highlight some of the major factors that influence the dissociation constant. These include the influence of the inductive effect, the stabilization of the dissociated anion by H-bonding as well as the presence of multiple ionizable groups. The acids investigated were aliphatic carboxylic acids, chlorine-substituted carboxylic acids, cid and trans-butenedioic, the isomers of hydroxybenzoic acid and phthalic acids and its isomers. It was found that in each of these examples the CPMD-metadynamics procedure correctly estimates the pKa values, indicating that the formulism is capable of capturing these influences and equally importantly indicating that the cancellation of errors is indeed universal. Further, it is shown that the procedure can provide accurate estimates of the successive pKa values of polypro tic acids as well as the subtle deference in their values for deterrent isomers of the acid molecule. Changes in protonation-deprotonation of amino acid residues in proteins play a key role in many biological processes and pathways. It is shown that CPMD simulations in conjunction with metadynamics calculations of the free energy profile of the protonation- deprotonation reaction can provide estimates of the multiple pKa values of the 20 canonical α-amino acids in aqueous solutions in good agreement with experiment (Chapter 5). The distance-dependent coordination number of the protons bound to the hydroxyl oxygen of the carboxylic and the amine groups is used as the collective variable to explore the free energy profiles of the Brϕnsted acid-base chemistry of amino acids in aqueous solutions. Water molecules, sufficient to complete three hydration shells surrounding the acid molecule were included explicitly in the computation procedure. The method works equally well for amino acids with neutral, acidic and basic side chains and provides estimates of the multiple pKa values with a mean relative error with respect to experimental results, of 0.2 pKa units. The tripeptide Glutathione (GSH) is one of the most abundant peptides and the major repository for non-protein sulfur in both animal and plant cells. It plays a critical role in intracellular oxidative stress management by the reversible formation of glutathione disulfide with the thioldisulfide pair acting as a redox buffer. The state of charge of the ionizable groups of GSH can influences the redox couple and hence the pKa value of the cysteine residue of GSH is critical to its functioning. In Chapter 6, it has been reported that ab initio Car-Parrinello Molecular Dynamics simulations of glutathione solvated by 200 water molecules, all of which are considered in the simulation. It is shown that the free-energy landscape for the protonation - deprotonation reaction of the cysteine residue of GSH computed using metadynamics sampling provides accurate estimates of the pKa and correctly predicts the shift in the dissociation constant values as compared to the isolated cysteine amino acid. The dissociation constants of weak acids are commonly determined from pH-titration curves. For simple acids the determination of the pKa from the titration curves using the Henderson-Hasselbalch equation is relatively straightforward. There are situations, however, especially in polypro tic acids with closely spaced dissociation constants, where titration curves do not exhibit clear inflexion and equivalence stages and consequently the estimation of multiple pKa values from a single titration curve is no longer straightfor- ward resulting in uncertainties in the determined pKa values. In Chapter 7, the multiple dissociation constant of the hexapeptide glutathione disulfide (GSSG) with six ionizable groups and six associated dissociation constants has been investigated. The six pKa values of GSSG were estimated using the CPMD-metadynamics procedure from the free-energy profiles for each dissociation reaction computed using the appropriate collective variable. The six pKa values of GSSG were estimated and the theoretical pH-titration curve was then compared with the experimentally measured pH-titration curve and found to be in excellent agreement. The object of the exercise was to establish whether interpretation of pH-titration curves of complex molecules with multiple ionizable groups could be facilitated using results of ab initio molecular dynamics simulations.

As most of spectroscopic measurements occur in solution, the solvent effect in the spectral properties of molecular systems is of great scientific interest. In particular, when the formation of hydrogen bond between the solute and solvent molecules occurs, the traditional effective dielectric solvent models provides inadequate results in some situations of interest. This is the case of diazines in aqueous solution. In this work, we study the solvent effects on the low-lying n=>π* absorption transition and on the 15N chemical shielding of diazines in water using the combination of Monte Carlo simulations and Quantum Mechanical calculations. As the classical simulations and the Quantum Mechanical calculations are uncoupled, a special attention is devoted to the solute polarization. Firstly, this is included by a previously developed iterative procedure where the solute is electrostatically equilibrated with the solvent. In addition, we verify the simple yet unexplored alternative of combining the polarizable continuum model (PCM) and the hybrid QM/MM method. In this procedure, we use PCM to obtain the average solute polarization and include this in the MM part of the sequential QM/MM methodology, PCM-MM/QM. These procedures are compared and further used in discrete and explicit solvent models. As the solute geometry is unchanged during all process, we study the influence of small differences on the molecular conformation in the spectroscopic properties using as prototype some recently sintetized quinone based molecules with antitrypanocidal activity. This study showed that small changes in the molecular conformation provide a small impact in the studied spectroscopic properties. The two implemented procedures in this work were systematically performed for each diazine and they were used in discrete and explicit solvent models for the calculations of the spectroscopic properties of each diazine in water as solvent. The results show good agreement between the values obtained by the two polarization methods. Considering the simplicity of the PCM over the iterative polarization method, this is na important aspect and the computational savings point to the possibility of dealing with larger solute molecules. This PCM-MM/QM approach conciliates the simplicity of the PCM model with the reliability of the combined QM/MM approaches. / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Conselho Nacional de Desenvolvimento Científico e Tecnológico / O efeito de solvente nas propriedades espectrais de sistemas moleculares é de grande interesse científico uma vez que grande parte das medidas espectroscópicas são realizadas na presença de um meio solvente. Métodos tradicionais que consideram o solvente como um meio dielétrico efetivo oferecem resultados insatisfatórios em algumas situações de interesse, em particular quando há a formação de ligações de hidrogênio entre o soluto e o solvente. Este é o caso de soluções aquosas de diazinas, que são as moléculas básicas para a formação dos ácidos nucleicos. Neste trabalho estudamos o efeito do solvente na energia de absorção da primeira transição n =>π* e a blindagem magnética nuclear do 15N das diazinas em água utilizando a combinação de simulações Monte Carlo e cálculos de Mecânica Quântica. No método utilizado neste trabalho, as simulações clássicas são desacopladas dos cálculos quânticos. Portanto, foi dada uma atenção especial à polarização do soluto que foi incluída de duas maneiras distintas. Inicialmente, a polarização é obtida utilizando um procedimento iterativo onde o soluto é submetido ao equilíbrio eletrostático com o solvente. Adicionalmente, é utilizada uma abordagem simples onde realizamos uma combinação entre o modelo contínuo (PCM) e os métodos híbridos QM/MM. Nesse último procedimento, que denominamos como PCM-MM/QM, nós utilizamos o método PCM para obter a polarização média do soluto, a qual é incluída na parte clássica do método sequencial QM/MM. Como a geometria do soluto é mantida inalterada durante todo o processo, nós realizamos um estudo do efeito da conformação molecular nas propriedades espectroscópicas, utilizando como protótipo moléculas recentemente sintetizadas derivadas de quinonas que apresentam atividade antimalárica. Este estudo mostra que pequenas variações na conformação molecular têm um pequeno impacto nas propriedades espectroscópicas estudadas. Os dois procedimentos implementados neste trabalho foram realizados de maneira sistemática para cada diazina e utilizados nos modelos discreto e explícito de solvente para os cálculos das propriedades espectroscópicas de cada diazina em água. Os resultados obtidos mostram uma boa concordância entre os valores obtidos através dos dois métodos de polarização. Considerando a simplicidade da polarização PCM se comparado com o processo iterativo, o método PCM-MM/QM abre a possibilidade de tratar sistemas compostos por moléculas do soluto maiores. Além disso, o método PCM-MM/QM concilia a simplicidade do modelo PCM com a confiabilidade dos métodos híbridos QM/MM.

Les méthodes classiques dans l'étude de la dynamique de réseau, comme la diffusion inélastique des neutrons et des rayons X, sont et vont rester limitées en flux. En conséquence les mesures sont coûteuse en temps. Pour optimiser le rendement de ces techniques, la stratégie de mesure doit être préparée avant l'expérience. Cette stratégie peut être élaborée et optimisé par des calculs de la dynamique de réseau et de la diffusion diffuse. La mesure de la diffusion diffuse thermique est une expérience simple où les régions étendues dans l'espace réciproque peuvent être explorées rapidement et en détails, ce qui permet d'identifier les caractéristiques dans la dynamique de réseau. Une méthode spectroscopique peut être appliquée ensuite sur les régions d'intérêt donnant accès à l'énergie et à l'intensité des vibrations individuelles. Dans certains cas, la diffusion diffuse thermique devient tellement contraignante pour les calculs de la dynamique de réseau (quasi)harmonique, que l'expérience de diffusion inélastique n'est plus un ingrédient nécessaire dans la reconstruction d'une image cohérente de la dynamique. Dans le cadre de ce travail la combinaison des techniques utilisant la diffusion thermique, la diffusion inélastique des rayons X et les calculs réalisés à partir des premiers principes ab initio est proposée pour l'étude de la dynamique de réseau de monocristaux. Les intensités de diffusion diffuse ainsi que les spectres inélastiques observés sont comparés à ceux calculés ab initio. Ces techniques combinées donnent accès à la description complète de la dynamique de réseau en approximation harmonique, et fournissent des informations supplémentaires précieuses. Le lecteur sera initié au formalisme de la dynamique de réseau et à celui de la diffusion inélastique et thermique. Les méthodes de calculs des propriétés vibrationnelles issues des calculs ab initio vont être introduites suivant un ensemble d'étapes menant à la convergence et donc à la validation de l'ensemble des calculs. Les techniques expérimentales utilisées tout au long de cette étude ainsi que les nouvelles possibilités s'ouvrant désormais grâce aux études combinées, serons présentées. La méthodologie sera illustrée par plusieurs systèmes de référence. Dans le cadre des systèmes à liaisons covalentes, deux polymorphes de silice - coésite et cristobalite - sont à l'étude. Les calculs expérimentaux validés sont utilisés pour l'étude des vecteurs propres, des valeurs propres et de leur contributions a la densité d'états vibrationnels (partiel et total). La comparaison avec le polymorphe de silice le plus abondant - alpha-quartz - et l'oxide de germanium en structure de alpha-quartz - révèle des ressemblances et des différences distinctes dans les propriétés vibrationnelles à basse énergie. Les polymorphes d'étain ont été choisis pour étudier l'impact du sous-système électronique sur les interactions inter-ioniques et la dynamique de réseau de monocristaux. L'étain manifeste des propriétés structurelles intéressantes, et une surface de Fermi relativement complexe. On observe une asymétrie inhabituelle au niveau de la diffusion diffuse, propriété qui peut s'expliquer dans le cadre d'une approximation harmonique de la dynamique de réseau. Enfin, la méthode élaborée est appliquée à de la glace, ce qui démontre non seulement une diffusion thermique caractéristique mais également des contributions statiques provenant du désordre de l'hydrogène. La méthodologie proposée fournit un outil puissant pour l'étude de la dynamique de réseau et sera applicable à une large variété de systèmes. Les études peuvent être étendues à des conditions extrêmes impliquant de très hautes pressions et une large gamme de températures. Cette méthodologie peut également être utilisée pour étudier les propriétés localisées de vibrations atomiques dans les systèmes avec des symétries brisées, par exemple des systèmes avec du désordre ou des effets topologiques. / The classical methods in the study of lattice dynamics, such as inelastic neutron and x-ray scattering, are and will remain flux-limited, consequently the measurements are time consuming. To maximise the yield of these techniques, measurement strategies need to be established prior to the experiment. These strategies can be elaborated and optimised by lattice dynamics calculations and thermal diffuse scattering. Measuring thermal diffuse scattering is a simple experiment where extended regions of reciprocal space can be rapidly explored in detail and characteristic features of the lattice dynamics identified. Slower spectroscopy measurements can then be applied on the selected regions of interest to gain access to the energy and intensity of individual vibrations. Moreover, in some cases the input of thermal diffuse scattering may become so constraining for the (quasi)harmonic lattice dynamic calculation, that inelastic scattering experiment will not be a necessary ingredient for the recovery of a self-consistent picture of the dynamics. In the frame of this work, the combination of thermal diffuse scattering, inelastic x-ray scattering and lattice dynamics calculations from first principles ab initio is applied to study the lattice dynamics of single crystals. Both diffuse scattering intensities and inelastic spectra determined by experiment are compared to the ones calculated ab initio. The combination of these three techniques gives access to the full lattice dynamics in the harmonic description and permits valuable new insights into the vibrational properties. The reader will be introduced to the key formalism of lattice dynamics, inelastic and thermal diffuse scattering. Methods for the calculation of vibrational properties from first principles are discussed, followed by a guideline for well converged calculations. The experimental techniques used in this work are presented and new possibilities for combined studies examined. The methodology is illustrated for several benchmark systems. Two silica polymorphs - coesite and alpha-cristobalite – were chosen as examples for covalent systems and investigated in detail. The experimentally validated calculation was used for the analysis of eigenvectors and eigenvalues of different modes, and their contribution to the total and partial density of vibrational states. Comparison with the most abundant silica polymorph - alpha-quartz - and germanium oxide in alpha-quartz structure reveals distinct similarities and differences in the low-energy vibrational properties. Metallic tin polymorphs were chosen to study the influence of the electron subsystem on inter-ionic interactions and the lattice dynamics. Tin exhibits both interesting structural properties and a complex Fermi surface. An unusual asymmetry of thermal diffuse scattering is observed which can be explained within the frame of harmonic lattice dynamics. Finally, the established method is applied to ice which exhibits not only characteristic thermal diffuse scattering but also static contributions from the hydrogen disorder. The methodology proposed in the present work provides a powerful tool in the study of lattice dynamics and will be applicable to a large variety of systems. The studies can be extended to extreme conditions involving very high pressures and a large temperature range. It may be also used to study localised properties of atomic vibrations in systems with broken symmetries, e.g. disorder or surface effects.

Nesta tese, utilizamos técnicas de química quântica para o estudo de sistemas contendo átomos de silício e nitrogênio. Nesse contexto, o trabalho aqui apresentado procurou concentrar-se em três tópicos principais: aglomerados de silício e nitrogênio, a superfície Si(100) e a interação de um átomo de nitrogênio com essa superfície. Estudamos inicialmente os aglomerados de fórmula SiNN e Si3N2, onde procuramos caracterizar de forma rigorosa a estabilidade e a natureza das ligações químicas das várias espécies. Com relação ao sistema SiNN, os resultados utilizando cálculos de alto nível mostraram a necessidade de novos dados experimentais para a caracterização inequívoca dessa espécie. Além disso, obtivemos os primeiros resultados para as propriedades de quatro novas estruturas. Estudamos a estrutura de várias espécies com fórmula Si3N2, e obtivemos como mínimo global uma estrutura planar contendo apenas ligações Si-N. Para a simulação da superfície Si(100) utilizamos os aglomerados de fórmula Si9H12 e Si15H16. Apenas a metodologia do Funcional da Densidade indicou a existência de uma estrutura distorcida para o Si9H12. Os cálculos CASSCF mostraram a necessidade do uso de funções multideterminantais. Com relação ao aglomerado Si15H16, obteve-se uma estrutura distorcida com os dímeros alternados como a forma mais estável. Entretanto, o uso do CASSCF mostrou ser a forma simétrica a mais estável. Com base no estudo dos aglomerados Si9H12 e Si15H16, partimos para a investigação dos mecanismos envolvidos no processo de interação e incorporação de átomos de nitrogênio na superfície de silício. Estruturas com simetria de spin quarteto e dupleto foram estudadas, sendo os mecanismos bastante semelhantes, com o nitrogênio interagindo inicialmente com um dos silícios dímeros para em seguida poder formar duas outras estruturas, uma com o nitrogênio ligado aos dois silícios dímeros e outra com o nitrogênio inserido no interior do aglomerado. Esta última estrutura corresponde à forma mais estável. Os resultados mostram claramente que a reação é bastante favorável do ponto de vista energético. Questionamos também o uso de vínculos durante a otimização das estruturas por impedir um devido relaxamento apropriado dos átomos de silício da primeira e segunda camadas. / In this thesis, quantum chemical techniques were used to study molecular systems containing silicon and nitrogen atoms. ln this context, the work was concentrated on three main topics: silicon-nitrogen clusters, the Si(100) surface, and the interaction of a nitrogen atom with that surface. We studied initially the clusters SiNN and Si3N2, where we tried to characterize with rigour the stability and the nature of the chemical bonds of the various species. Concerning the system SiNN, using high level calculation, the results showed the necessity of new experimental data for the unequivocal characterization of this species. Moreover, the properties of four new structures were described for the first time in this study. The structures of various species with formula Si3N2 were studied, and the global minimum corresponds to a planar structure containing only Si-N bonds. For the simulation of Si(100) surface, the clusters Si9H12 and Si15H16 were used. Density Functional Theory predicted the existence of an asymmetric geometry for the Si9H12 cluster. The CASSCF calculations showed the necessity of using multideterminantal wave functions. Concerning the Si15H16 cluster, the most stable structure corresponds to a distorted one with alternated dimers. However, the CASSCF methodology showed that the symmetric structure is the most stable. Based on the results for the Si9H12 and Si15H16 clusters, we started to investigate the mechanism involved in the interaction and incorporation of a nitrogen atom into the silicon surface. Structures with quartet and doublet spin symmetry were studied, being their mechanisms very similar, with the nitrogen atom interacting initially with one silicon dimer; after that two other structures can be formed, one with the nitrogen bonded to two silicon dimers, and the other with the nitrogen bonded to one silicon dimer and two internal silicons. This last structure corresponds to the most stable species. The results show clearly that the reaction is very favorable energetically. The use of geometrical constraints during the optimization was questioned since it prevents the proper relaxation of the silicon atoms in the first and second layers.

O estudo da reatividade de íons em fase gasosa é bem diversificado e têm sido extensivamente explorado com técnicas de espectrometria de massas. Essa tese apresenta resultados experimentais e teóricos em três áreas distintas da química de íons em fase gasosa. Os processos iônicos foram experimentalmente estudados em um espectrômetro de massa por transformada de Fourier (FTMS ou FT-ICR), método de alta resolução que possibilita um acompanhamento temporal de reações de íons em -fase gasosa. Em vários momentos são apresentados resultados de cálculos teóricos, necessários para a elucidação do mecanismo das reações observadas e das estruturas de espécies envolvidas. Um dos assuntos estudados se refere à obtenção de dados termoquímicos a partir da dissociação de íons induzida por radiação infravermelha. O método desenvolvido anteriormente no laboratório, que utiliza como fonte de radiação um filamento aquecido, foi comparado com a dissociação induzida por laser de CO2.Os resultados indicam que o primeiro método é capaz de diferenciar íons através das suas energias de dissociação enquanto que o segundo mostrou ser pouco sensível às variações de energias de dissociação para os íons estudados (derivados de acetofenona e alguns alquilbenzenos). Um outro tema abordado é a reação de substituição nucleofílica aromática em fase gasosa. As reações de nucleófilos como F-, OH- e alguns alcóxidos com nitrobenzeno e halobenzenos foram detalhadas. O mecanismo da reação e a atuação de complexos íon-molécula foram caracterizados. Foi possível estabelecer que reações de substituição nucleofílica aromática podem ser muito rápidas em fase gasosa mesmo na ausência de grupos ativadores no anel aromático. Reações SNAr podem ocorrer em fase gasosa em competição com abstração de próton, ou como reação secundária do produto desprotonado através da formação de um complexo íon-molécula de tempo de vida suficiente para sofrer rearranjos. O último assunto apresentado é a reação de íons de caráter eletrofílico, CF3+,CCl3+ e CClF2<SUP+, com acetofenona e derivados. Vários caminhos de reação foram observados experimentalmente, o que proporcionou um amplo estudo para a elucidação das estruturas e dos mecanismos das reações. Os resultados obtidos com substratos deuterados e os resultados de cálculos teóricos indicam que os diferentes produtos de reação observados são devidos a um ataque inicial do eletrófilo no oxigênio carbonílico da acetofenona. / The study of ionic reactivity in the gas-phase has been widely explored with mass spectrometry techniques. This thesis reports experimental and theoretical results in three different areas of gas-phase ion chemistry. The ionic processes were experimentally studied by Fourier transform mass spectrometry (FTMS or FT-ICR), a high resolution method that allows for the observation of ion/molecule reactions. For almost all cases, extensive theoretical calculations were necessary to clarify the mechanisms of the observed reactions and to establish the structures of the ions. The first part of the thesis deals with the possibility of deriving termochemical data from ion dissociation processes induced by infrared radiation. The method previously developed in our laboratory, which uses a heated tungsten wire as the infrared source, was compared with the dissociation induced by a CO2 laser. The results show that the former method can readily differentiate ions by their dissociation energy while the second one proved to be insensitive for the ions studied in this thesis (acetophenones derivatives and some alkylbenzenes). The second part deals with gas-phase nucleophilic aromatic displacement reactions. Reactions of gas-phase nucleophiles such as F-, OH- and some alcoxides with nitrobenzene and halobenzenes were extensively studied. The reaction mechanism and the role of ion-molecule complexes were established. Nucleophilic aromatic displacement can be very fast in the gas-phase even in the absence of activating groups in the ring. Gas-phase Sn Ar reactions may occur in competition with proton abstraction, or may N result from a secundary reaction following proton abstraction through a ion-molecule complex with life time enough to suffer rearrangements. The last part of the thesis deals with the reaction of eletrophilc ions, such as CFSUB>3+ , CCl3+ e CClF2+ , with acetophenone and substituted 3 3 2 acetophenones. Several reaction paths were experimentally observed, wich motivated a thorough investigation of these reactions to establish the structures of the ions and the mechanisms of the reactions. The combination of experimental results obtained with deuterated substrates and theoretical calculations suggest that all the different products observed result from initial attack of the eletrophilic ion on the carbonyl oxygen atom of acetophenone.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-05-05T17:36:02Z No. of bitstreams: 1 raphaeldasilvaalvin.pdf: 3200596 bytes, checksum: 01129a5b7d70b7b3e44dd871c5c77b1d (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-05-17T13:41:10Z (GMT) No. of bitstreams: 1 raphaeldasilvaalvin.pdf: 3200596 bytes, checksum: 01129a5b7d70b7b3e44dd871c5c77b1d (MD5) / Made available in DSpace on 2017-05-17T13:41:10Z (GMT). No. of bitstreams: 1 raphaeldasilvaalvin.pdf: 3200596 bytes, checksum: 01129a5b7d70b7b3e44dd871c5c77b1d (MD5) Previous issue date: 2009-02-19 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Os organofosforados encontram aplicações na indústria, nas áreas de corantes, vernizes, couro artificial, isolantes elétricos, impermeabilizantes, plásticos, aditivos de petróleo e dissolventes; também são usados na medicina no tratamento de doenças, como o glaucoma; no uso doméstico e na agricultura encontram aplicabilidade como inseticidas e pesticidas. Além dessas várias utilidades civis, os organofosforados são utilizados, também, como armas químicas de destruição em massa, e possuem estrutura similar àquelas dos compostos utilizados como inseticidas e pesticidas. No caso do VX, um organofosforado utilizado como arma química, a quebra da ligação P­S é essencial na sua degradação química. Embora muitas reações químicas possam ser empregadas para decompor agentes químicos de guerra, somente algumas podem ser utilizadas na prática em uma neutralização, porque estas reações precisam ser simples e os reagentes empregados devem ser estáveis, baratos e de baixa massa molecular. No entanto, muitas das reações que podem ser úteis para a neutralização do agente neurotóxico VX ainda seguem em discussão em recentes pesquisas, mas na maioria delas se limita alguns sucessos a hidrólise catalisada. Nesta dissertação foram estudados processos de hidrólise catalisada do agente VX por MgO(001) por meio de cálculos ab initio. Foi utilizado o programa PWscf ­ Plane­Waves Self Consistent Field. O PWscf utiliza a Teoria do Funcional da Densidade, a partir de um conjunto de base de autofunções dado por ondas planas e pseudopotenciais. Entre outros atributos, este código é capaz de calcular a energia do estado fundamental dos orbitais de Kohn­Sham para um elétron, além de forças atômicas em diferentes condições de tensão, otimização estrutural e estado de transição. A molécula de VX foi substituída por uma molécula menor, chamada de p­VX, em que foram substituídos alguns radicais do VX por grupos metila. Isto foi feito para diminuir o tamanho da molécula, que reduzirá o custo computacional, mas sem afetar substancialmente a química do problema, o estudo da quebra da ligação P­S. O mecanismo de hidrólise proposto esta relacionado com a quebra heterolítica da ligação P­S, com a conseqüente formação de íons intermediários R­P+ e R'­S­, estes, por sua vez, estabilizados por quimissorção na superfície de MgO(001). Em conjunto com essa reação, acontece a dissociação de moléculas de água para a formação dos íons H+ e HO­, cujos íons também são estabilizados na superfície de MgO(001). O passo final é a recombinação desses íons, para gerar os produtos de hidrólise R­POH e R'­SH, seguida pelo processo de dessorção destas moléculas da superfície do catalisador. Para a reação global de hidrólise da molécula de p­VX, a variação da energia interna foi calculada em ­5,66kcal/mol. Foram determinadas as estruturas dos íons R­P+ e R'­S­ estabilizados sobre a superfície de MgO(001), com uma energia de formação calculada em ­0,20kcal/mol, indicando que os intermediários teriam boa estabilidade sobre a superfície se comparados com a molécula de p­VX original. Na quimissorção dissociativa de moléculas de água sobre a superfície de MgO(001), verificou­se que os íons formados somente ficam estabilizados se estiverem a uma distância mínima de 4,70Å. Qualquer distância abaixo desta levará a formação da molécula de água novamente. No processo envolvendo duas moléculas de água, apenas uma delas se dissocia, enquanto a outra estabiliza os íons formados via ligação de hidrogênio. A molécula não dissociada também interage com um sítio superficial de magnésio. Este resultado foi comprovado pelos cálculos de diferença de densidade de carga eletrônica do sistema, determinação do caminho de reação, onde este obteve uma barreira energética calculada em 5,55kcal/mol para a reação direta e em 7,53kcal/mol para a reação inversa, e pela dissociação parcial utilizando um trímero de moléculas de água, com energia calculada em ­5,40kcal/mol. Os resultados permitem concluir que o mecanismo proposto para a hidrólise catalisada do agente neurotóxico VX pelo MgO é possível. Os modelos construídos podem ser modificados para testes de novos catalisadores com estrutura tipo MgO, via adição de defeitos ou dopantes à superfície da estrutura cristalina, visando a elaboração de catalisadores mais eficientes para a reação de hidrólise com o mesmo mecanismo. / The organophosphates are used in industry, in the fields of dyes, varnishes, artificial leather, electrical insulation, waterproofing, plastics, oil additives and solvents, are also used in medicine to treat diseases such as glaucoma, in the household and in agriculture are applied as insecticides and pesticides. Besides these various civilian facilities, the organophosphates are used as well as chemical weapons of mass destruction, and have similar structure to those of compounds used as insecticides and pesticides. In the case of VX, an organophosphate used as a chemical weapon, the fall in P­S binding is essential in its chemical degradation. Although many chemical reactions can be used to decompose the chemical agents of war, only some can be used in practice in a breakthrough, because these reactions need to be simple and the reagents used should be stable, inexpensive and low molecular weight. However, many of the reactions that may be useful for the neutralization of the neurotoxic VX agent still follow under discussion in recent polls, but most of them are confined to some successes catalyzed hydrolysis. In this dissertation we studied processes of the VX agent catalyzed hydrolysis by MgO (001) by means of ab initio calculations. We used the program PWscf ­ Plane Waves Self­Consistent Field. The PWscf using the Density Functional Theory from a set of basic autofunction given by plane waves and pseudopotentials. Among other attributes, this code is able to calculate the energy of the ground state of the Kohn­Sham orbital for an electron, and atomic force under different conditions of stress, structural optimization and transition state. The VX molecule was replaced by a smaller molecule, called a p­VX, which replaced some of the radicals VX for methyl groups. This was done to reduce the size of the molecule, which reduces the computational cost, but not substantially affect the chemistry of the problem, the study of breaking the link P­S. The proposed mechanism of hydrolysis is related to the breaking of the link heterolytic P­S, with the consequent formation of intermediate ion R­P+ and R'­S­, they, in turn, stabilized by quimissorption the MgO(0010 surface. Together with this reaction, is the dissociation of water molecules to the formation of ions H+ and HO­, whose ions are stabilized on the MgO(001) surface. The final step is the recombination of these ions, to generate products of hydrolysis and R­POH and R'­SH, followed by the process of desorption of molecules from the surface of the catalyst. For the overall reaction of hydrolysis of the molecule p­VX, the variation of internal energy was calculated to be ­5.66 kcal/mol. Were determined the structures of ion R­P+ and R'­S­ stabilized on the MgO(001) surface, with an formation energy calculated at ­0.20 kcal/mol, indicating that the middlemen have good stability on the surface is compared with the p­VX molecule original. In dissociative quimissorption of water molecules on the MgO(001) surface, it was found that the ions formed are stable only if a minimum distance of 4.70 Å. Any distance below this will lead to formation of the water molecule again. In the process involving two water molecules, only one is dissociated, while the other ions stabilizes by the hydrogen bonding formed. The non­dissociated molecule also interacts with a surface site of magnesium. This result was confirmed by the calculations of load density difference of the system, determining the reaction path, where he obtained a energy barrier calculated at 5.55 kcal/mol for the direct reaction and 7.53 kcal/mol for the reverse reaction and by partial decoupling using a trimer of water molecules, with energy calculated at ­5.40 kcal/mol. The results show that the proposed mechanism for the catalyzed hydrolysis of the neurotoxic agent VX by MgO is possible. The models constructed can be modified for testing of new catalysts with MgO type structure, via addition of doping or defects on the surface of the crystal structure, to the development of more efficient catalysts for the hydrolysis reaction with the same mechanism.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-05-10T15:19:03Z No. of bitstreams: 1 vivianedasilvavaiss.pdf: 3626163 bytes, checksum: 3a5c810252b870e4922ae7564819b604 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-05-17T15:10:36Z (GMT) No. of bitstreams: 1 vivianedasilvavaiss.pdf: 3626163 bytes, checksum: 3a5c810252b870e4922ae7564819b604 (MD5) / Made available in DSpace on 2017-05-17T15:10:36Z (GMT). No. of bitstreams: 1 vivianedasilvavaiss.pdf: 3626163 bytes, checksum: 3a5c810252b870e4922ae7564819b604 (MD5) Previous issue date: 2011-01-27 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Os hidróxidos lamelares, dentre os quais se incluem os compostos tipo brucita e os hidróxidos duplos lamelares (HDL), apresentam inúmeras aplicações devido às suas propriedades, tais como: alta estabilidade térmica, capacidade de troca iônica, porosidade e elevada área específica, eletroquímicas, fotoquímicas e outras. A capacidade de adsorção e troca iônica apresentada por estes compostos faz com que estes materiais sejam amplamente utilizados na remoção de espécies aniônicas e contaminantes. Assim, por exemplo, tais materiais podem ser usados na remoção do excesso de fluoreto de água potável, pois fluoreto em excesso pode causar fluorose esquelética ou dental, e na desativação de agentes neurotóxicos, organofosforados que atacam o sistema nervoso. O presente trabalho, apresenta um estudo detalhado dos mecanismos de reação de H2O, HF e sarin com hidróxidos lamelares. Para tanto, foram utilizados cálculos ab initio baseados na Teoria do Funcional da Densidade com condições de contorno e o método CI-NEB (Climbing Image-Nudging Elastic Band) para calcular o caminho de reação e a barreira energética de cada reação envolvida nos processos. A reação de HF foi baseada num mecanismo proposto composto por quatro reações elementares: a adsorção de HF sobre a superfície dos hidróxidos, a dissociação do HF adsorvido, a dessorção da molécula de água formada na reação de dissociação e a formação do composto M(OH)2−xFx. Os resultados mostram que a formação do composto M(OH)2−xFx é espontânea. Este resultado está de acordo com trabalhos experimentais, que prevêem como estável o composto formado pela substituição de OH- por F- em hidróxidos lamelares. Foram analisadas duas propostas para a formação de hidroxila sobre a superfície de hidróxidos lamelares. Na primeira uma hidroxila da própria lamela do hidróxido migra para a superfície e na segunda uma molécula de água se dissocia sobre a superfície do hidróxido, formando uma molécula de água e hidroxila adsorvidas na superfície, sendo esta última proposta mais favorável segundo os cálculos realizados. Os trabalhos anteriores foram etapas necessárias para o objetivo central da tese: estudar a desativação do agente neurotóxico sarin. No processo de desativação do sarin foram estudados intermediários formados pela dissociação e quimissorção do agente sobre a superfície da brucita. Com base nos resultados dos três processos estudados pôde-se verificar que a reação de desativação do agente sarin utilizando a brucita é termodinamicamente mais estável do que a reação de hidrólise. Além disso, verificou-se que a água pode participar do processo de desativação como um catalisador. Tal estudo está no âmbito do Programa Pró-Defesa do ministério da Defesa em conjunto com a CAPES. / Layered hydroxides, including brucite-like compounds and layered double hydroxides (LDHs), present numerous applications due to their properties such as: high thermal stability, ion exchange capacity, porosity and high surface area, electrochemistry, photochemistry and others. The capacity of adsorption and ion exchange presented by these materials make them to be widely used in the removal of anionic species and contaminants. Thus, for example, such materials can be used in the removal of excess fluoride from drinking water, because fluoride in excess can cause dental or skeletal fluorosis, and degradation of nerve agents, like organophosphates that attack the nervous system. This work presents a detailed study of the reactions H2O, HF and sarin with layered hydroxides. Therefore, were used ab initio calculations based on Density Functional Theory (DFT) with periodic boundary conditions and the CI-NEB method (Climbing Image – nudging Elastic band) to calculate the reaction path and the energetic barrier for each reaction involved in the process. The reaction of HF was based on a proposed mechanism consisting of four elementary steps: the adsorption of HF on the surface of hydroxides, the dissociation of adsorbed HF, the desorption of water molecule formed in the dissociation reaction and the formation of the compound M(OH)2−xFx. The results show that the formation of the compound M(OH)2−xFx is spontaneous. This result agrees with experimental works, that predicts as stable the compound formed by replacing OH- by F- in layered hydroxides. Two proposals were analyzed for the formation of hydroxyl on the surface of layered hydroxides. In the first proposal a hydroxyl of the hydroxide lamella migrates to the surface and in the second proposal a water molecule dissociates on the surface of the hydroxides, forming a water molecule and hydroxyl adsorbed on the surface; the latter proposal is more favorable according to the calculations performed. The previous studies were necessary steps for the central goal of the thesis: to study the deactivation of the nerve agent sarin. In the deactivation process of the nerve agent sarin intermediates formed by dissociation and chemisorption of the agent on the surface of brucite were studied. Based on the results of the three cases studied it could be checked that the deactivation reaction of the agent sarin using brucite is thermodynamically more stable than the hydrolysis reaction. Moreover, it appears that water can participate in the deactivation process as a catalyst. This study has happened into the ambit of the Pro-Defense program of the Defense Menistry in conjunction with the CAPES.

Ce travail se place dans le cadre des réacteurs nucléaires de 4ème génération tels que les réacteurs à neutrons rapides et caloporteur gaz (RNR-G), pour lesquels des matériaux réfractaires comme le nitrure de titane (TiN) enroberont le combustible afin de permettre la rétention des produits de fission. Cette étude a porté sur le comportement thermique intragranulaire du xénon dans des échantillons de TiN obtenus par frittage à chaud sous charge. Le rôle de la microstructure sur le comportement thermique du xénon a été étudié. Plusieurs lots ont ainsi été synthétisés sous différentes conditions de température et de composition de la poudre initiale. Le xénon a été introduit par implantation ionique. Les échantillons ont ensuite subi des traitements thermiques entre 1300°C et 1600°C, soient les températures accidentelles envisagées. Un transport majoritaire du xénon vers la surface a été mis en évidence. Ce dernier est ralenti lorsque la température de frittage augmente. Des différences de comportement ont été observées selon les poudres mises en oeuvre dans la synthèse et selon l'orientation cristalline du grain considéré. Le relâchement du xénon a également été corrélé à l’oxydation de TiN. Des bulles de Xe dès 0,38 % atomique ont été observées. Leur taille est proportionnelle à la concentration en Xe et augmente avec la température de recuit, d’où une certaine mobilité du Xe au sein de TiN. Plusieurs mécanismes pouvant expliquer cette mobilité sont proposés. En complément, des calculs ab initio ont confirmé le caractère fortement insoluble du Xe dans TiN et révélé que les bilacunes sont les plus favorables à l'incorporation du xénon au sein de ce matériau. / This work concerns the generation IV future nuclear reactors such as gas-cooled fast reactor (GFR) for which refractory materials as titanium nitride (TiN) are needed to surround fuel and act as a fission product diffusion barrier. This study is about Xe thermal behavior in sintered titanium nitride. Microstructure effects on Xe behavior have been studied. In this purpose, several syntheses have been performed using differents sintering temperatures and initial powder compositions. Xenon species have been introduced into samples by ionic implantation. Then, samples were annealed in temperature range from 1300°C to 1600°C, these temperatures being the accidental awaited temperature. A transport of xenon towards sample surface has been observed. Transport rate seems to be slow down when increasing sintering temperature. The composition of initial powder and the crystallographic orientation of each considered grain also influence xenon thermal behavior. Xenon release has been correlated with material oxidation during annealing. Xenon bubbles were observed. Their size is proportional with xenon concentration and increases with annealing temperature. Several mechanisms which could explain Xe intragranular mobility in TiN are proposed. In addition with experiments, very low Xe solubility in TiN has been confirmed by ab initio calculations. So, bivacancies were found to be the most favoured Xe incorporation sites in this material.

Nous présentons des études expérimentales et théoriques de disiliciures alcalino-terreux, le disilane (Si2H6) et du carbone à haute pression. Nous étudions les disiliciures et en particulier le cas d’une phase plane de BaSI2 qui a une structure hexagonale avec des liaisons sp3 entre les atomes de silicium. Cet environnement électronique conduit à un gaufrage de feuilles du silicium. Nous démontrons alors une amélioration de la température de transition supraconductrice de 6 à 8.9 K lorsque les couches de silicium s’aplanissent dans cette structure. Des calculs ab initio basés sur DFT ont guidé la recherche expérimentale et permettent d’expliquer comment les propriétés électroniques et des phonons sont fortement affectés par les fluctuations du flambage des plans de silicium. Nous avons aussi étudié les phases cristallines de disilane à très haute pression et une nouvelle phase métallique est proposé en utilisant les méthodes de prédiction de structure cristalline. Les températures de transition calculées donnant un supraconducteur autour de 20 K à 100 GPa. Ces valeurs sont significativement plus faibles comparées à celles avancées dans la littérature. Finalement, nous présentons des études de structures de carbone à haute pression à travers une recherche de structure systématique. Nous avons trouvé une nouvelle forme allotropique du carbone avec une symétrie Cmmm que nous appelons Z-carbone. Cette phase est prévue pour être plus stable que le graphite pour des pressions supérieures à 10 GPa. Des expériences et simulation de rayon-X et spectre Raman sugèrent l’existence de Z-carbone dans des micro-domaines de graphite sous pression / We present experimental and theoretical studies of sp3 materials, alkaline-earth-metal (AEM) disilicides, disilane (Si2H6) and carbon at high pressure. First, we study the AEM disilicides and in particular the case of a layered phase of BaSi2 which has an hexagonal structure with sp3 bonding of the silicon atoms. This electronic environment leads to a natural corrugated Si-sheets. Extensive ab initio calculations based on DFT guided the experimental research and permit explain how electronic and phonon properties are strongly affected by changes in the buckling of the silicon plans. We demonstrate experimentally and theoretically an enhancement of superconducting transition temperatures from 6 to 8.9 K when silicon planes flatten out in this structure. Second, we investigated the crystal phases of disilane at the megabar range of pressure. A novel metallic phase of disilane is proposed by using crystal structure prediction methods. The calculated transition temperatures yielding a superconducting Tc of around 20 K at 100 GPa and decreasing to 13 K at 220 GPa. These values are significantly smaller than previously predicted Tc’s and put serious drawbacks in the possibility of high-Tc superconductivity based on silicon-hydrogen systems. Third, we studied the sp3-carbon structures at high pressure through a systematic structure search. We found a new allotrope of carbon with Cmmm symmetry which we refer to as Z-carbon. This phase is predicted to be more stable than graphite for pressures above 10 GPa and is formed by sp3-bonds. Experimental and simulated XRD, Raman spectra suggest the existence of Z-carbon in micro-domains of graphite under pressure