In this thesis, we focus on theoretical investigations on metal interfaces where many heterogeneous chemical reactions take place. Surface-enhanced Raman scattering (SERS) spectroscopy and the modern electrochemical methods are important in-situ techniques that have been widely employed for a variety of applications. Theoretical simulations have become an indispensable tool to infer the molecular details of interfacial structures that are not directly accessible from experimental measurements. In this context, we have proposed several new theoretical models for both SERS and interfacial electrochemistry, which allow us to provide molecular-level understanding of the interfacial structures under the realistic experimental conditions.   The first part of the thesis has addressed the basic theory of SERS that offers the vibrational structure of the interfacial molecules. It is well known that the huge enhancement of Raman intensity in this technique can be attributed to two independent factors, namely the physical and chemical enhancements. The former is resulted from the enhanced electromagnetic field induced by the plasmonic excitations, while the latter comes from the changing of interaction between the molecule and the surface. The interplay between these two enhancement factors, which has long been an issue of debate, is revealed in this thesis. They are coupled through molecular polarizability. A practical computational approach is proposed and used to demonstrate the importance of the coupling on different molecular systems. It is found that for certain systems the coupling factor can be as large as 106. This finding is of great importance towards a comprehensive understanding of the SERS mechanisms and a quantitative prediction of the enhancement factor.   The other part of the thesis is devoted to the theory of interfacial electrochemistry, in particular the effects of water solution. A novel protocol that combines classical molecular dynamics (MD) and the first principles density functional theory (DFT) calculations is proposed to address the statistical behavior of interfacial properties. Special attention has been paid to the work function of Pt(111) surface and CO adsorption energy on Pt(111) surface in aqueous solution. It has been found that in this case the work function of Pt surface illustrates a surprisingly broad distribution under the room temperature, sheds new light on the understanding of reaction activity of the surface. The proposed protocol is able to provide results in very good agreement with experiments and should be applied routinely in future studies. / <p>QC 20120515</p>

A computational analysis on the first hyperpolarizability (£]) of 5-6-5 heterocyclic rigid-rod benzobisthiazole-based chromophores and their derivatives has been systematically investigated in this study by the software Gaussian03, in an effort to provide a molecular design criterion for better opto-electonic properties. The specific objectives of this study have been focused on (1) understanding of 5-6-5 heterocyclic rigid-rod chromophores and other similar compounds in nonlinear optical property, (2) control on the nonlinear optical properties through the manipulation of the conjugated composition of the chromophores, (3) understanding the effect of the donor and acceptor of chromophore on the nonlinear optical property. The results show that £] value of a chromophore with its 5-6-5 heterocyclic rigid-rod structure near the acceptor is 30 % higher than the one with the same structure near the donor; this reflects a unique property of 5-6-5 heterocyclic rigid-rod structure quite different from other chromophores. In conjugated chromophores, those having double bonds or thiophenes, which have soft structures or heteroatoms, would offer higher £] values. For example, a 5-6-5 heterocyclic rigid-rod chromophore with 15 double bonds has a £] value up to 1.44 x 10-26esu. In the study of multi-donors and acceptors, from 1 donor- 1 acceptor to 2 donors- 2 acceptors, the difference in £] could be up to 116%; however, this difference reduces to only 5% if the chromophere structure becomes unstable. From the band-resolved analysis, it is found that £] values of chromophores are affected by the electronic donor via its HOMO part, the electronic acceptor via its LUMO part, while the conjugated bridge via its molecular orbital exclusive of HOMO and LUMO, depending on the type and degree of conjugation, i.e., HOMO-2 and LUMO+1 for double bonds and HOMO-2, HOMO-1, and LUMO+1 for phenyl moieties. For the 5-6-5 heterocyclic rigid-rod structure, £] value is mainly affected by those molecular orbital below HOMO, i.e., HOMO-2 for those containing double bonds. Further detailed elucidation on the relation between molecular orbital and molecular structure of chromophores can be ambiguously depicted through analyses of electronic distribution on the molecular orbital.

Research into determining what intrinsic characteristics cause materials to be hard is imperative if one would like to design future materials with a hardness exceeding that of diamond. Measuring the hardness of materials in order to obtain a fundamental quantity independent of extrinsic factors is difficult, if not impossible. However, many theories have been proposed pertaining to the quantification of hardness as a fundamental property. While it is clear that the hardness of a material will strongly depend on its crystal structure, another fundamental quantity, the electronic band gap, has also been linked to the intrinsic hardness of materials. The electronic band gap is a seemingly simple quantity, but is difficult to de- termine for novel or complicated materials. Core-level spectroscopy techniques that probe the occupied and unoccupied density of states separately allow for an indirect determination of the electronic band gap. These methods have several advantages over conventional tech- niques in that they do not strongly depend on the extrinsic material properties such as defects and impurities. The electronic band gap has been determined in this way for several novel materials. These include group 14 nitrides with spinel structure that were recently studied over the last decade. The electronic band gap of three synthesized binary spinel nitrides γ-Si3N4, γ-Ge3N4 and γ-Sn3N4 are determined using core-level spectroscopy to be 4.8 ± 0.2 eV, 3.5 ± 0.2 and 1.6 ± 0.2 eV, respectively. These measurements agree with the calculated values of 4.97, 3.59 and 1.61 eV for γ-Si3N4, γ-Ge3N4 and γ-Sn3N4, respectively. We have also extended these measurements and calculations to include the solid solutions γ-(GexSi1−x)3N4 and γ-(SnxGe1−x)3N4 showing these spinel-structured nitrides form a multi-functional class of semiconductors. This band gap measurement technique has also been applied successfully to the phosphor converting light emitting diode material Ba3Si6O12N2 and the novel semicon- ductor MnNCN. This shows that using core-level spectroscopy is a very effective method to determine the electronic band gap where there are no other feasible techniques. Aside from the electronic band gap, core-level spectroscopy is also a complementary technique to deter- mine the crystal structure, which is also an important parameter with regard to hardness. The crystal structure, particularly aspects such as anion ordering and vacancy ordering, have been determined for the spinel-structured oxonitride Ga3O3N and a novel phase of calcium nitride Ca3N2. These results show that core-level spectroscopy is a powerful technique to determine the anion ordering in oxonitrides and was further applied to the material class β-sialons, allowing for the determination of the electronic band gap as well as ascertaining both the anion and cation ordering. Combining all of these aspects we show that the electronic band gap is not only useful for predicting the hardness of materials, but in some cases can be used to predict the existence of certain materials. We use theoretical methods, combined with experimental measurements, to calculate the hardness and electronic band gap of all possible binary and ternary group 14 spinel-structured nitrides. Through the correlation of the hardness and electronic band gap we show that only the three already synthesized binary group 14 spinel-structured nitrides are stable along with their solid solutions and that the elusive spinel-structured carbon nitride γ-C3N4 will be never synthesized.

Discrete Fourier Transform (DFT) and Finite Impulse Response (FIR) filters are extensively used in Digital Signal Processing (DSP) and Image Processing. As a result, there is a strong motivation to come up with area- and delay-efficient hardware realizations of DFT and FIR filters. In this thesis, we propose an arithmetic Sum-of-Product (SOP) based approach to implement area- and delay-efficient Discrete Fourier Transform (DFT) and FIR filter circuits. Our SOP based engine uses an improved column compression algorithm, and handles the sign of the input efficiently. The partial products of the computation are compressed down to 2 operands, which are then added using a single hybrid adder (which is comprised of a ripple carry adder for the early-arriving lower-order bits, a Kogge-Stone adder for the slower middle bits, and a carry-select adder for the early-arriving higher order bits). The DFT and FIR filters can also be cast as instances of the Multiple Constant Multiplication (MCM) problem. RAG-n is one of the best known algorithms for realizing an MCM block with the minimum number of adders. We compare our SOP-based implementations with the RAG-n algorithm. We implement both approaches using a 45 nm cell library, and demonstrate that our approach yields a faster DFT circuit (by about 12-13%), with a small (about 5%) area penalty and a significantly better algorithmic runtime. We also demonstrate that our approach realizes FIR filters with hard-to-implement coefficients with a 4.4× speedup and 1.38× area penalty as compared to two recent adder cascade based approaches. For a set of symmetric and asymmetric filters, we compare the area-delay curves of the circuits generated by using our SOP based approach with that of the circuits generated by using a Common Sub-expression Elimination (CSE) based algorithm, which tries to minimize the number of adders utilized under a maximum adder cascade length constraint. We show that for a large range of delays, the circuits generated by using our approach have the smallest area. Finally, we propose a new hybrid form realization for FIR filters. The hybrid form realization attempts to perform the computation using both the Direct and Transposed Direct form realization styles. We discuss conditions under which it would improve on both the Direct and Transposed Direct form realizations, in terms of circuit area and delay.

Ces travaux portent sur la recherche de matériaux alternatifs d’électrolyte ou d’électrodes de piles à combustible à oxyde solide. Une méthodologie basée sur la composition de l’oxyborate La26O27(BO3)8, développé à l’IMN, a permis de mettre en évidence deux matériaux prometteurs : Ba3Ti3O6(BO3)2 et K3Sb4O10(BO3). La synthèse de poudres de Ba3Ti3O6(BO3)2 et de phases substituées sur les sites du Ba ou du Ti ont été réalisés par voie solide à 950 °C. Les mesures de conductivité ont été effectuées par EIS sur des échantillons denses (compacité ≥ 90 %). Sous air, la conductivité est purement anionique et dépasse 10-4 S.cm-1 à 700 °C. Elle augmente pour les composés substitués par un élément de valence supérieure, et inversement. Sous atmosphère hydrogénée, une forte augmentation de conductivité est observée (x 200), liée à l’apparition d’une contribution électronique. Une étude couplant DRX, XPS et ATG montre que cette contribution est due à une réduction de 5 % du Ti4+ en Ti3+ et que cette réaction est réversible. Les calculs DFT ont permis de déterminer les énergies de formation et de migration des défauts dans le matériau. L’oxyborate K3Sb4O10(BO3) a été obtenu sous forme de monocristaux et de poudre. Une étude approfondie de la densification a été nécessaire afin d’obtenir des échantillons denses (compacité ≈ 90 %), en utilisant un broyage planétaire et/ou une aide au frittage. La conductivité du matériau sous air est de l’ordre de 10-3 S.cm-1 à 700 °C. Ces travaux mettent en évidence pour la première fois des niveaux de conductivité (ioniques et/ou électroniques) importants dans les oxyborates. Cette approche peut être appliquée à la recherche de matériaux alternatifs pour SOFC. / This work focuses on the search for alternative electrolyte or electrodes materials for solid oxide fuel cells. A methodology based on the composition of the La26O27(BO3)8 oxyborate, developed at the IMN, revealed two promising materials: Ba3Ti3O6(BO3)2 and K3Sb4O10(BO3). Syntheses of powders of Ba3Ti3O6(BO3)2 and substituted phases on the Ba or Ti atomic site were carried out by solid state reaction at 950 °C. Conductivity measurements were carried out by electrochemical impedance spectroscopy on dense samples (relative density ≥ 90 %). Under air, the conductivity is purely anionic and exceeds 10-4 S.cm-1 at 700 °C. Conductivity increases for compounds substituted with a supervalent element, and vice versa. In a hydrogen containing atmosphere, a large increase of conductivity is observed (x 200), linked to the appearance of an electronic contribution. A study combining XRD, XPS and TGA shows that this contribution is due to the reduction of 5 % of the Ti4+ in Ti3+ and that this reaction is reversible. DFT calculations allowed to determine the formation energies and the migration barriers of the defects in the material. K3Sb4O10(BO3) oxyborate was obtained as single crystals and powder. A thorough study of the densification of the material was necessary in order to obtain dense samples (relative density ≈ 90 %), using ball milling and/or sintering aid. The conductivity of the material in air is about 10-3 S.cm-1 at 700 °C. This work highlights significant conductivity levels (ionic and/or electronic) observed for the first time in oxyborates. This approach can be applied to find alternative materials for SOFC.

Ce manuscrit décrit l’application de la RMN du solide de l’17O à l’étude structurale d’espèces oxo de tungstène bien définies en lien avec des catalyseurs industriels pour la métathèse des oléfines. Dans un premier temps, pour compenser la faible réceptivité de l’17O, des techniques d'amélioration de signal tels que DFS (Double Frequency Sweep) et HS (Hyperbolic Secant) ont été évaluées de façon critique pour ces systèmes comportant des interactions anisotropes d’amplitudes très diverses. La méthode DFS s’est avérée la plus robuste, avec un gain en signal de 2-2,4. Dans une seconde étape, des complexes oxo de tungstène moléculaires et supportés sur silice ont été étudiés par RMN MAS 17O, après enrichissement isotopique du groupement oxo. Les paramètres RMN de W=O sont très sensibles à la sphère de coordination du métal : combiné avec des calculs DFT, ceci permet une évaluation de la structure des espèces greffées. Le marquage sélectif en 17O de la surface de silice a également amené à une meilleure compréhension de ces systèmes, plus particulièrement en ce qui concerne les interactions métal-support. Des informations supplémentaires ont été obtenues grâce à l'application des méthodes de haute résolution (17O MQ MAS) et de corrélation hétéronucléaire (1H-17O HMQC). Quelques perspectives de ce travail sont avancées, plus particulièrement sur la chimie de surface du molybdène. / This manuscript aims at the use of 17O solid state NMR for accessing the structure of well-defined silica-supported oxo-tungsten species related to industrial olefin metathesis catalysts. As a first step, to compensate for the low receptivity of 17O, signal enhancement techniques such as DFS (Double Frequency Sweep) and HS (Hyperbolic Secant) were critically assessed for such systems featuring large range of anisotropic interactions. DFS proved to be the most robust method, providing a signal enhancement of 2-2.4. In a second stage, series of molecular and silica-supported tungsten-oxo complexes have been studied by 17O MAS NMR, following isotopic enrichment of the oxo moiety. The W=O NMR parameters showed a high sensitivity to the metal coordination sphere, thus allowing structural assessment of grafted species when combined with DFT calculations. Silica-surface selective 17O labelling also afforded deeper understanding of these systems, most particularly regarding metal-support interactions. Further elements were obtained thanks to application of methods for high resolution (17O MQ MAS) and heteronuclear correlation (1H-17O HMQC). Some perspectives of this work are drawn, most particularly on the related molybdenum surface chemistry.

In this dissertation, I present four projects on the fundamental study of the surface configurations and reactivity of the metal oxides using density functional theory computational method. In the first project, we studied the formaldehyde adsorption and diffusion on rutile TiO2 (110) surface. By comparing the adsorption of formaldehyde on stoichiometric and defective TiO2 surfaces under the same condition, we evaluated the effect of surface oxygen vacancy on their interaction with formaldehyde. The project involved close collaboration with Dr. Zhenrong Zhang’s group of Baylor University who studied the formaldehyde chemistry on rutile TiO2(110) surface using a combination of STM and other surface science techniques. In the second project, we compared the surface chemistries of formaldehyde and formic acid on rutile TiO2 and SnO2, two structurally similar but chemically different oxides. We analyzed the oxidation of formaldehyde to formic acid on two oxides and assessed the role of surface oxygen in the oxidation. In the third project, we studied the oxygen evolution reaction (OER) catalyzed by γ-FeOOH (010) under the alkaline condition. The OER process was divided into four elementary steps and the potential energy profiles of these steps on three terminations of the γ-FeOOH (010) surface were mapped out. Based on the computed reaction energies, we determined the most probable OER reaction pathway on each surface termination. We found that partially exposed surface Fe sites were the active sites for the OER process. In the fourth project, we studied the potential of iron oxides (FeOx) and iron-titanium mixed oxides (FeTiOx) as solid oxygen carriers for the chemical looping combustion (CLC) process. As oxygen carriers for CLC, FeOx and FeTiOx in fully oxidized forms went through a series of reduction steps by reacting with the fuel molecules. The reduced oxides were then re-oxidized in an air reactor to restore their oxygen. By studying the surface oxygen vacancy formation and oxygen diffusion, we gained insights into the initial stage of reduction process and activities of FeOx and FeTiOx as well as the effect of Ti on oxygen carrying properties of FeTiOx for CLC.

As a reliable, convenient, and advantageous tool in the theoretical investigations of bioorganic, inorganic, and organometallic chemistry, density functional theory (DFT) computations have provided chemists with numerous significant insights. The understanding of mechanisms of chemical reactions, and the design and development of catalysts have been greatly promoted by the employment of DFT. In this dissertation, the applications of DFT computations in the catalytic bioorganic, inorganic, and organometallic systems were studied. Phosphoramidate hydrolysis catalyzed by human histidine triad nucleotide binding protein 1 (hHint1) was investigated using a cluster-model DFT approach, and the key involvement of the histidine triad as a proton shuttle was discussed in the proposed mechanism. The IEFPCM-Bondi-B3LYP/BS1 methodology was demonstrated as a reliable, and time-saving model in computing the reduction potentials of transition metal complexes. Moderate accuracy (MAD = 0.233 V, mean absolute deviation) and good linear correlation (R2 = 0.93) between computed and experimental reduction potentials of the 49 studied species are osberved. The fluxionality of cyclohexenyl manganese tricarbonyl [(C6H9)Mn(CO)3] was investigated using DFT computations, which uncovered a previously uncharacterized “closed” Cs agostomer. The intramolecular oxidative amination of an alkene catalyzed by the extreme π-loading N-heterocyclic carbene pincer Tantalum(V) bis(imido) complex was also computationally analyzed, and the mechanisms of the formation of oxidative amination product, reduction product, and hydroamination product were investigated. The computational results are consistent with the experimentally observed product ratios and selectivity.

Desde finales del siglo XX, la crisis ambiental ha tomado relevancia tanto en espacios académicos como en la sociedad. La utilización de combustibles fósiles y su consecuente emisión de gases de efecto invernadero, provocan una creciente inquietud a nivel mundial. Una solución a este problema, es la utilización de hidrógeno como vector energético, debido a su notable rendimiento. Sin embargo, el principal desafío radica en el almacenamiento y liberación eficiente para su posterior utilización. Con el objetivo de abordar esta problemática, la presente tesis doctoral se enfoca en el estudio de dos sistemas prometedores vinculados a esta cuestión. El primer sistema estudiado es hidruro de magnesio dopado con niobio (MgH2+Nb) con vacancias cargadas. Se analizaron los cambios estructurales, magnéticos y energéticos. En una segunda parte, se analizó el efecto de la adsorción de distintos metales alcalinos sobre nanotubos de carbono de pared simple (8,0) prístinos y con una vacancia de carbono. Los metales alcalinos aumentan la cantidad de sitios activos para la adsorción de hidrógeno. Un estudio detallado de la adsorción de hidrógeno en nanotubos de carbono, requiere un entendimiento completo del sustrato para lograr mejoras significativas en las propiedades de adsorción y así maximizar el potencial de almacenamiento. En este sistema se estudiaron las geometrías de adsorción, la naturaleza de los enlaces y la estructura electrónica. Todos los cálculos fueron realizados utilizando la teoría del funcional de la densidad (DFT) con spín polarizado implementados por el código VASP. / Since the end of the 20th century, the environmental crisis has taken on relevance both in academic spaces and in society. The use of fossil fuels and their consequent emission of greenhouse gases, cause a growing concern worldwide. A solution to this problem is the use of hydrogen as an energy vector, due to its remarkable performance. However, the main challenge lies in efficient storage and release for later use. In order to address this problem, this doctoral thesis focuses on the study of two promising systems related to this issue. The first system studied is magnesium hydride doped with niobium (MgH2+Nb) with charged vacancies. Structural, magnetic and energetic changes were analyzed. In a second part, the effect of the adsorption of different alkali metals on pristine single-walled carbon nanotubes (8,0) and with a carbon vacancy, was analyzed. Alkali metals increase the number of active sites for hydrogen adsorption. A detailed study of hydrogen adsorption on carbon nanotubes requires a thorough understanding of the substrate in order to achieve significant improvements in adsorption properties and thus maximize storage potential. In this system, the adsorption geometries, the nature of the bonds and the electronic structure were studied. All calculations were performed using spin polarized density functional theory (DFT) implemented by the VASP code.

No description available.