Neste trabalho foram efetuados estudos importantes e pioneiros sobre as propriedades estruturais, eletrônicas e termodinâmicas dos nitretos e de suas ligas, através de dois métodos de primeiros princípios distintos, o FLAPW (\"Full-potential Linear Augmented Plane Wave\") e o pseudopotencial combinado com a aproximação quasequímica generalizada. Na primeira parte, utilizando o método FLAPW, calculamos as estruturas de bandas para os nitretos cúbicos do grupo-IH, BN, AIN, GaN e InN. Foram obtidos valores para a constante de rede e \"bulk modulus\" para os nitretos do grupo-III através de cálculos relativísticos da energia total. Através das estruturas de bandas e analisando o topo da banda de valência e o fundo da banda de condução perto do ponto r ou, no ponto k correspondente ao mínimo da banda de condução, derivamos os respectivos valores para as massas efetivas de elétron e de buraco pesado, leve e de \"split-off\' e correspondentes parâmetros de Luttinger. Todos os resultados são comparados com dados experimentais e teóricos existentes na literatura. Na segunda parte, utilizando o método FLAPW, estudamos a influência da impureza de Mg na estrutura eletrônica do GaN cúbico (c-). Realizamos cálculos da otimização da geometria, incluindo deslocamentos dos primeiros e segundos vizinhos, para os casos da impureza com estados de carga neutro e negativo. Obtivemos o valor de 190 meV para o deslocamento de Franck-Condon da energia térmica, o qual apresenta um bom acordo com os dados experimentais de fotoluminescência e efeito Hall. Nós concluímos que os primeiros e segundos vizinhos desempenham um papel importante na determinação das energias do nível aceitador resultante da dopagem do c-GaN com Mg. Na terceira parte, nós apresentamos cálculos das propriedades eletrônicas, estruturais e termodinâmicas de ligas cúbicas envolvendo os nitretos do grupo-III, InxGa1-xN, InxAl-xN, AlxGal-xN, BxGal-xN e BxA1-xN. Nós combinamos o método de expansão de \"clusters\" através da aproximação quasequímica generalizada (\"Generalized Quasichemical Approximation -GQCA\") com cálculos de pseudopotenciais \"ab initio\" DFT-LDA. Para todas a ligas, exceto a de AlxGal-xN, encontramos separação de fase para temperaturas próximas das temperaturas de crescimento. Generalizamos o método de expansão de \"c1usters\" para estudar a influência da tensão biaxial. Encontramos uma significativa supressão da separação de fase induzida pela tensão para as ligas de InxGal-xN e InxAh-xN, sendo no caso da liga de InxGal_xN confirmado experimentalmente. Observamos também que flutuações da energia do \"gap\" da liga de InxGal-xN permitem definir valores mínimo e médio para a energia do \"gap\" com diferentes valores para o \"bowing\". Observamos que a tensão biaxial reduz as flutuações da energia do \"gap\", resultando em uma diminuição do valor do \"bowing\". Através deste estudo mostramos uma possível explicação para a discrepância experimental para valores do \"bowing\". / In this work we performed a pioneer theoretical study of structural, electronic and thermodynamic properties of the group-III nitrides and their alloys, by using two distinct first principles methods, the FLAPW full potential linear augmented plane wave and the pseudopotential-plane-wave method combined with the generalized quasichemical approximation. In the first part of our work, by using the FLAPW, we present the electronic band structures ofthe zinc-blende-type group-III nitrides compounds, BN, AIN, GaN, and InN. Lattice constant and bulk modulus are obtained from fuH relativistic total-energy calculations. Electron, heavy-, light-, and split-off-hole effective masses and corresponding Luttinger parameters are extracted from the band-structure calculations. A comparison with other available theoretical results and experimental data is made. In the second part of our work, by using the FLAPW method, the electronic structure of Mg impurity in zinc-blende (c-) GaN is investigated. Full geometry optimization calculations, including nearest and next-nearest neighbor displacements, were performed for the impurity in the neutral and negatively charged states. A value of 190 meV was obtained for the Franck-Condon shift to the thermal energy, which is in good agreement with that observed in recent low temperature photoluminescence and Hall-effect measurements. We conclude that the nearest and the next-nearest neighbors of the Mg impurity replacing Ga in c-GaN undergo outward relaxations which play an important role in the determination ofthe center acceptor energies. In the third part of our work, we present a study of electronic, structural, and thermodynamic properties of the cubic group-III nitrides alloys, InxGal-xN, InxAll_xN, AlxGal-xN, BxGal-xN e BxAll-xN. We combined the generalized quasichemical approximation (GQCA) with an ab initio pseudopotential-plane-wave method. For alI alIoys, except the AlxGal-xN, we observe a miscibility gap for temperatures near those of the growth. The cluster treatment is generalized to study the influence of biaxial strain. We find a remarkable suppression ofphase separation in InxGal-xN and InxAll-xN induced by strain which is confirmed by experiments on the InxGal_xN alloy. We also observed that the gap fluctuations in the InxGal-xN alloy allow the definition of a minimum gap and an average gap with different bowing parameters. Biaxial strain drastically reduces the gap fluctuations, resulting in a reduction of the bowing. The different gaps and the strain influence investigated provide an explanation for the discrepancies found in the experimental values of the bowing parameter.

Si les diagrammes de phase des alliages binaires massifs sont désormais bien connus, il en va tout autrement de ceux des nanoalliages, particules constituées de quelques centaines à quelques milliers d’atomes. Le développement des nanotechnologies est un puissant moteur pour développer la connaissance des diagrammes de phase des nanoparticules. Dans ce but, nous avons étudié le polyèdre de Wulff de 405 atomes (PW405) dans le système Cu-Ag caractérisé par une forte différence de rayons atomiques entre les constituants et une tendance à la démixtion, du moins dans sa forme massive. L’étude est réalisée par simulations Monte Carlo utilisant un potentiel interatomique à N-corps, en tirant parti de la richesse de deux ensembles thermodynamiques, à savoir les ensembles pseudo-grand canonique, i.e. à différence de potentiel chimique fixée, et canonique, i.e. à concentration nominale fixée. Nous montrons tout d’abord qu’un diagramme de phase de nanoalliage est constitué d’un ensemble de diagrammes de phase attachés à des classes de sites de surface ou des couches internes. Ainsi, pour le PW405, nous distinguons trois diagrammes de phase : ceux des facettes (100) et (111) et celui des couches de cœur. Chacun de ces trois diagrammes est attaché à des phénomènes physiques très différents. Ainsi, le diagramme de phase des facettes (100) est relatif à une transition structuro-chimique, alors que celui des facettes (111) caractérise une transition démixtion – désordre sans évolution structurale. Dans l’ensemble p-GC, cela se traduit par une bistabilité collective pour les facettes (100) et individuelle pour les facettes (111). Une conséquence de ce dernier point est la possibilité d’observer des facettes (111) pures en cuivre et d’autres pures en argent au sein d’un même agrégat !L’utilisation de la méthode de Widom pour déterminer dans l’ensemble canonique s’est révélée déterminante pour pouvoir séparer les régimes monophasés de ségrégation superficielle ou sub-superficielle des régimes biphasés. La différence de rayons atomiques joue un rôle important, tant dans la stabilisation dans l’état biphasé de la configuration Janus dissimulée par une coquille d’argent, que dans l’abaissement de la température critique par un facteur d’environ 2 par rapport au diagramme de phase massif.Une analyse des simulations Monte Carlo par une approche en champ moyen sur réseau effectif s’est révélée d’une richesse insoupçonnée pour déterminer les fores motrices de chaque type de transition, tant à la surface que dans les couches de cœur. Cela nous a permis de plus de montrer que les arêtes jouent le rôle d’une véritable nano-armature pour l’agrégat. Ce travail s’achève par ce qui constitue, à notre connaissance, la proposition du premier diagramme de phase de nanoalliage. / If bulk phase diagrams of binary alloys are now well known, it is not the case for nanoalloys, which are particles consisting in a few hundred to a few thousand of atoms. The development of nanotechnologies is a powerful driving force to develop the interest in the phase diagrams of nanoparticles. For this purpose, we have studied the Wulff polyhedron of 405 atoms (PW405) in the Cu-Ag system characterized by a large difference in atomic radii between the components and a tendency to phase separation, at least in the bulk. The study is carried out by Monte Carlo simulations using N-body interatomic potentials, taking advantage of the complementarity of two thermodynamic ensembles, namely the semigrand canonical ensemble (sGC), i.e. at fixed difference in chemical potentials and the canonical ensemble, i.e. at fixed nominal concentration.We first show that a phase diagram of nanoalloy consists in a set of phase diagrams related to the various classes of surface sites or to the internal layers. Thus, for the PW405, we distinguish three phase diagrams: the (100) facets diagram, the (111) facets diagram and the phase diagram for the layers of the core. Each of these diagrams is linked to very different physical phenomena. Thus, the phase diagram of the (100) facets is related to a structural and chemical transition, while the one of the (111) facets characterizes a transition between phase separation and disorder, without structural evolution. In the sGC ensemble, this results in a collective bistability for the (100) facets and an individual bistability for the (111) facets. A consequence of this last point is the possibility to observe some (111) facets pure in copper and other (111) facets pure in silver within the same nanoparticle !The use of the Widom method to determine in the canonical ensemble is decisive in order to discriminate between a single-phase regime (with surface or subsurface segregation) and a two-phase regime. Furthermore, the difference of atomic radii between Cu and Ag plays an important role, both for stabilizing the Janus configuration (with an Ag shell) in the two-phase state, and for lowering the critical temperature of the core layers by a factor of about 2 with respect to the bulk phase diagram.An analysis of the Monte Carlo simulations within an effective lattice formalism and a mean-field approximation is very powerful to determine the driving forces at the origin of each type of transition, both for the different facets of the surface layers and for the core. Moreover, this allowed us to show that the edges act as a nano-armature for the nanoparticles. This work ends with the establishment of what should be considered as the first phase diagram of nanoalloys, to the best of our knowledge.

The commercial polycrystalline superalloy RR1000 developed for turbine disc applications contains a large number of alloying elements. This complex alloy chemistry is required in order to produce appropriate microstructures and the required mechanical properties, such that the most important strengthener γʹ displays complex alloy chemistry. The broad aim of this project is to develop an approach to measuring the composition of γʹ precipitates at a broad range of length scales from nanometres to hundreds of nanometres, and subsequently develop a better understanding of the role of thermodynamics and diffusion kinetics on γʹ phase separation and precipitate growth. A solution of the absorption-corrected EDX spectroscopy to spherical particles was developed in our work, therefore enabling the quantitative analysis of precipitates' composition using an absorption-corrected Cliff-Lorimer approach. By performing this quantification, size-dependent precipitate compositional variations were obtained. Examination of this quantitative approach was compared to thermodynamic calculations of primary γ' precipitates possessing equilibrium compositions. Given the development of semi-quantitative compositional measurements for spherical γʹ precipitates and that cooling is one of the most common and critical regimes in physical metallurgy of Ni-based superalloys, this approach was then applied to study the local compositional variations that are induced in γ' precipitates when the alloy RR1000 undergoes different cooling rates. These measured compositions have been compared to detailed thermodynamic calculations and provide new experimental evidence of the importance of the dominant role of aluminium antisite diffusion in determining the low-temperature growth kinetics of fine-scale γ' precipitates. We have applied a similar analysis approach to study the compositional variations of γʹ cores within the class of secondary precipitates upon cyclic coarsening and reversal coarsening. It was shown that supersaturated Co in secondary γʹ exhibits an overall trend towards the equilibrium but Co content can significantly increase as γʹ coarsens. It was demonstrated that the limited elemental diffusivity in γ and γʹ compared to the observed coarsening rate in the coarsening regime results in the long-lasting Co supersaturation in γʹ and builds up elemental enhancements or depletions. These inhomogeneous elemental distributions produce compressive elastic constraints on large-scale secondary γʹ, therefore inducing morphological instability of these γʹ and causing the reversal coarsening. These results enable us to better understand the role that both thermodynamics and limited diffusion kinetics plays in controlling the complex microstructures of γ' precipitates.

<p>The interactions between anionic, crosslinked gels and cationic surfactants have been investigated. When exposed to oppositely charged surfactant, the gel collapses into a dense complex of polyion and micelles. During deswelling, the gel phase separates into a micelle-rich, collapsed surface phase, and a swollen, micelle-free core, both still part of the same network. As more surfactant is absorbed, the surface phase grows at the expense of the core, until the entire gel has collapsed. Polyacrylate (PA) gels with dodecyl- (C₁₂TAB), and cetyltrimethylammonium bromide (C₁₆TAB), as well as hyaluronate gels with cetylpyridinium chloride, have been studied. </p><p>Kinetic experiments have been performed on macro- as well as microgels, using micromanipulator assisted light microscopy for the latter. A surfactant diffusion controlled deswelling model has been employed to describe the deswelling. The deswelling kinetics of PA microgels have been shown to be controlled by surfactant diffusion through the stagnant layer surrounding the gel, as the surface phase is relatively thin for the major part of the deswelling. For macroscopic PA gels the surface phase is thicker, and the kinetics with C₁₂TAB were therefore also influenced by diffusion through the surface phase, while for C₁₆TAB they were dominated by it. </p><p>Relevant parameters have also been determined using equilibrium experiments. An irregular, balloon-forming deswelling pattern, mainly found for macrogels, as well as unexpectedly long lag times and slow deswelling for microgels, are reported and discussed. </p><p>The microstructure of fully collapsed PA/C₁₂TAB complexes has been studied using small-angle X-ray scattering. A cubic <i>Pm3n</i> structure was found at low salt concentration, which melted into a disordered micellar phase as the salt concentration was increased. Further increasing the salt concentration dissolved the micelles, resulting in no ordering.</p>

The interactions between anionic, crosslinked gels and cationic surfactants have been investigated. When exposed to oppositely charged surfactant, the gel collapses into a dense complex of polyion and micelles. During deswelling, the gel phase separates into a micelle-rich, collapsed surface phase, and a swollen, micelle-free core, both still part of the same network. As more surfactant is absorbed, the surface phase grows at the expense of the core, until the entire gel has collapsed. Polyacrylate (PA) gels with dodecyl- (C12TAB), and cetyltrimethylammonium bromide (C16TAB), as well as hyaluronate gels with cetylpyridinium chloride, have been studied. Kinetic experiments have been performed on macro- as well as microgels, using micromanipulator assisted light microscopy for the latter. A surfactant diffusion controlled deswelling model has been employed to describe the deswelling. The deswelling kinetics of PA microgels have been shown to be controlled by surfactant diffusion through the stagnant layer surrounding the gel, as the surface phase is relatively thin for the major part of the deswelling. For macroscopic PA gels the surface phase is thicker, and the kinetics with C12TAB were therefore also influenced by diffusion through the surface phase, while for C16TAB they were dominated by it. Relevant parameters have also been determined using equilibrium experiments. An irregular, balloon-forming deswelling pattern, mainly found for macrogels, as well as unexpectedly long lag times and slow deswelling for microgels, are reported and discussed. The microstructure of fully collapsed PA/C12TAB complexes has been studied using small-angle X-ray scattering. A cubic Pm3n structure was found at low salt concentration, which melted into a disordered micellar phase as the salt concentration was increased. Further increasing the salt concentration dissolved the micelles, resulting in no ordering.

The overall objective of this PhD thesis research is to understand and control phase separation in mixed perfluorinated fatty acid-phospholipid surfactant systems that have applications as pulmonary surfactant (PS) mixtures, with an ultimate view of controlling film composition, morphology and mechanical properties. In this context the interaction between perfluorooctadecanoic acid (C18F), 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), the major component of native PS extract, and 1,2-dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG) has been explored in Langmuir monolayers and Langmuir–Blodgett (LB) films using a combination of atomic force microscopy (AFM), fluorescence microscopy (FM) and Brewster angle microscopy (BAM) measurements. Thermodynamic and morphological studies of binary and ternary mixed films made of C18F, DPPC and DPPG indicated that both the phospholipids and C18F were miscible over a wide range of compositions. The mixed phospholipid-C18F films contained multimolecular aggregates that were highly enriched in the phospholipids. Furthermore, it was found that the magnitude of the DPPC-C18F interaction could be modulated by altering the concentration of sodium ions in the underlying subphase. Using a highly simplified lung mimic fluid (pH 7.4, 150mM NaCl), DPPC and C18F became fully immiscible. Moreover, the performance characteristics of the mixed films demonstrated the usefulness of C18F as an additive for PS formulations. The effectiveness of a PS protein mimicking peptide was evaluated against DPPC to allow comparison with previous measurements of DPPC-C18F mixed system. The mixing thermodynamics of the peptide and DPPC in Langmuir monolayer implied a repulsive interaction between the film components. The hysteresis response of the mixed monolayer films indicated that the lipid-protein mixture improved the re-spreading of DPPC films. Moreover, molecular-level organization of the mixed films explored by both FM and BAM confirmed the formation of liquid-expanded DPPC domains in the presence of minute amount of the peptide. In order to obtain a thorough understanding of the effect of the deposition process and surfactant tail polarities on the interfacial behavior of perfluorocarbon-hydrocarbon mixed monolayer films, both BAM and AFM measurements of arachidic acid (C20) with perfluorotetradecanoic acid (C14F) and palmitic acid (C16) with C18F mixed monolayer were performed. These measurements revealed that film morphology was minimally perturbed upon its deposition onto solid substrates. Coarse grained molecular dynamics (MD) simulations of films comprised of DPPC molecules with tails of various polarities suggested that the phase separation between the monolayer components could be controlled by varying surfactant tail polarities.

The phase separation of mixed Langmuir-Blodgett (LB) monolayers was investigated using a combination of atomic force microscopy (AFM), X-ray photoelectron emission microscopy (X-PEEM) and confocal fluorescent microscopy measurements. Shapes of phase-separated domains that formed on solid substrate surfaces depended on a competition between line tension and dipole-dipole interactions. In the mixed LB film of arachidic acid (C19H39COOH) (C20) and perfluorotetradecanoic acid (C13F27COOH) (F14), the components phase separated into elevated hexagonal domains of C20 surrounded by a continuous domain primarily consisting of F14. The underlying molecular arrangement of C20 was found to be an oblique packing. The domains in this system grew via Ostwald ripening and the kinetics of their growth was modeled by twodimensional LifshitzSlyozov equation. In the stearic acid (C17H35COOH) (C18) and F14 mixed films, the C18 domains formed a linear pattern where the F14 molecules filled the areas in between the lines occupied by C18. For the mixed film of palmitic acid (C15H31COOH) (C16) and perfluorooctadecanoic acid (C17F35COOH) (F18), the surfactants phaseseparated into elevated hexagonal domains with hairy extensions radiating from them. These domains were composed of F18 and surrounded by C16. Ostwald ripening was found to be the mechanism of domain growth. Phase separation was controlled by different forces such as line tension and dipole interactions, as well as the diffusion of the molecules, solubility of the surfactant in the sub-phase, temperature and surface pressure. Simple mechanisms regarding phase separation and pattern formation were discussed in these mixed systems. It was observed that all fatty acid / F14 systems in this study were immiscible at all molar fractions examined. The fatty acid / F18 systems were immiscible at short chains of fatty acids (myristic acid (C13H27COOH) C14, C16, C18), whereas at longer fatty acid chains (C20, C22 behenic acid (C21H43COOH)) the components of the mixed system became miscible. When perfluorocarboxylic acid chain combined with fatty acids, the domains changed from large hexagonal domains into narrow lines as the fatty acid chain decreased in length.

The phase separation of mixed Langmuir-Blodgett (LB) monolayers was investigated using a combination of atomic force microscopy (AFM), X-ray photoelectron emission microscopy (X-PEEM) and confocal fluorescent microscopy measurements. Shapes of phase-separated domains that formed on solid substrate surfaces depended on a competition between line tension and dipole-dipole interactions. In the mixed LB film of arachidic acid (C19H39COOH) (C20) and perfluorotetradecanoic acid (C13F27COOH) (F14), the components phase separated into elevated hexagonal domains of C20 surrounded by a continuous domain primarily consisting of F14. The underlying molecular arrangement of C20 was found to be an oblique packing. The domains in this system grew via Ostwald ripening and the kinetics of their growth was modeled by twodimensional LifshitzSlyozov equation. In the stearic acid (C17H35COOH) (C18) and F14 mixed films, the C18 domains formed a linear pattern where the F14 molecules filled the areas in between the lines occupied by C18. For the mixed film of palmitic acid (C15H31COOH) (C16) and perfluorooctadecanoic acid (C17F35COOH) (F18), the surfactants phaseseparated into elevated hexagonal domains with hairy extensions radiating from them. These domains were composed of F18 and surrounded by C16. Ostwald ripening was found to be the mechanism of domain growth. Phase separation was controlled by different forces such as line tension and dipole interactions, as well as the diffusion of the molecules, solubility of the surfactant in the sub-phase, temperature and surface pressure. Simple mechanisms regarding phase separation and pattern formation were discussed in these mixed systems. It was observed that all fatty acid / F14 systems in this study were immiscible at all molar fractions examined. The fatty acid / F18 systems were immiscible at short chains of fatty acids (myristic acid (C13H27COOH) C14, C16, C18), whereas at longer fatty acid chains (C20, C22 behenic acid (C21H43COOH)) the components of the mixed system became miscible. When perfluorocarboxylic acid chain combined with fatty acids, the domains changed from large hexagonal domains into narrow lines as the fatty acid chain decreased in length.

Combinatorial techniques have changed the paradigm of materials research by allowing a faster data acquisition in complex problems with multidimensional parameter space. The focus of this thesis is to demonstrate biomaterials design and characterization via preparation of two dimensional combinatorial libraries with chemically-distinct structured patterns. These are prepared from blends of biodegradable polymers using thickness and temperature gradient techniques. The desired pattern in the library is chemically-distinct cell adhesive versus non-adhesive micro domains that improve library performance compared to previous implementations that had modest chemical differences. Improving adhesive contrast should minimize the competing effects of chemistry versus physical structure. To accomplish this, a method of blending and crosslinking cell adhesive poly(季aprolactone) (PCL) with cell non-adhesive poly(ethylene glycol) (PEG) was developed. We examine the interaction between MC3T3-E1 osteoblast cells and PCL-PEG libraries of thousands of distinct chemistries, microstructures, and roughnesses. These results show that cells grown on such patterned biomaterial are sensitive to the physical distribution and phases of the PCL and PEG domains. We conclude that the cells adhered and spread on PCL regions mixed with PEG-crosslinked non-crystalline phases. Tentatively, we attribute this behavior to enhanced physical, as well as chemical, contrast between crystalline PCL and non-crystalline PEG.

The objective of this study is to illustrate that phase separation improves the efficiency of an anaerobic system which digests semi-solid organic wastes. Organic fraction of municipal solid waste (OFMSW) was the semi-solid organic waste investigated. In the first part of the study, the optimum operational conditions for acidified reactor were determined by considering the volatile solid (VS) reductions and average acidification percentages at the end of two experimental sets conducted. Organic loading rate (OLR) of 15 g VS/L.day, pH value of 5.5 and hydraulic retention time (HRT) of 2 days were determined to be the optimum operational conditions for the acidification step. Maximum total volatile fatty acid and average acidification percentage were determined as 12405 mg as HAc/L and 28%, respectively in the reactor operated at optimum conditions. In the second part, an acidification reactor was operated at the optimum conditions determined in the first part. The effluents taken from this reactor as well as the waste stock used to feed this reactor were used as substrate in the biochemical methane potential (BMP) test. The results of BMP test revealed that the reactors fed by acidified samples indicated higher total chemical oxygen demand (tCOD) removals (39%), VS reductions (67%) and cumulative gas productions (265 mL). The result of this study indicated that the separation of the reactors could lead efficiency enhancement in the systems providing that effective control was achieved on acidified reactors.