Une étude expérimentale systématique de la formation des microstructures de solidification d’alliages métalliques (Al-Cu) a été effectuée par application de la radiographie X synchrotron. La radiographie X-Synchrotron nous a donné accès à des observations in situ et en temps réel qui nous ont permis d’analyser de manière quantitative les phénomènes physiques impliqués au cours de la solidification (vitesse de croissance, redistribution du soluté, interaction entre grains équiaxes etc.). Elle nous a également permis de mettre en évidence l’influence de la convection naturelle et de la gravité sur ces différents paramètres physiques et par conséquent sur la formation de la microstructure de solidification. Nous avons comparé nos résultats expérimentaux avec des modèles de prédiction de la croissance dendritique et ensuite avons montré l’intérêt des expériences en microgravité. Nous avons enfin effectué des séries de tests du dispositif expérimental conçu et développé par SSC (Swedish Space Corporation) dans le cadre du projet XRMON (In situ X-Ray MONitoring of advanced metallurgical processes under microgravity and terrestrial conditions) de l'ESA-MAP en vue d’une expérience in situ et en temps réel de solidification en microgravité à bord d’une fusée sonde Maser12. Cette expérience prévue en Novembre 2011. Les résultats obtenus lors des séries de tests valident le dit dispositif en termes de comportement thermique et d’imagerie X par radiographie. / A systematic experimental study of the formation of solidification microstructures of metallic alloys (Al-Cu) was carried out by the application of synchrotron X-ray radiography. Synchrotron X-ray radiography gave access to in situ and real time observations which allowed us to quantitatively analyze the dynamical physical parameters involved in the solidification process (growth rate, solute redistribution, equiaxed grain interaction etc). It also allowed to show the influence of natural convection and gravity on these various physical parameters and consequently on the formation of the solidification microstructures. Our experimental results were compared with models predictions of the dendritic growth and the necessity of microgravity solidification experiments was evidenced. Finally series of tests were carried out on the experimental setup designed by the Swedish Space Corporation (SSC) in the framework of the project XRMON (In situ X-Ray MONitoring of advanced metallurgical processes under microgravity and terrestrial conditions) of ESA-MAP for in situ and real time solidification experiments under microgravity conditions on board a sounding rocket Maser12. Maser12 mission is scheduled for November 2011. The tests results validated the experimental setup in term of imaging and thermal behaviour.

Cette étude s’inscrit dans le cadre d’un projet ADEME (Agence de l'Environnement et de la Maîtrise de l'Energie) et a pour objectif la réduction du frottement entre chemise et segments via l’introduction de nouveaux revêtements. Notre rôle était de proposer à nos partenaires des matériaux de revêtements de type métal-intermétalliques complexes aux propriétés de frottement optimisées. Dans un premier temps, nous avons élaboré par frittage à l’état solide des matériaux composites Al/(AlCuFeB)p contenant des particules intermétalliques complexes (alliages quasicristallins de structure icosaédrique du système AlCuFeB) renforçant une matrice d’aluminium pur. Cette partie de l’étude consiste à étudier les cinétiques de transformations de phases résultantes de la diffusion de l’aluminium provenant de la matrice vers les particules icosaédriques. Il a été montré que la déstabilisation de la phase icosaédrique peut être évitée par la création d’une barrière de diffusion via un prétraitement d’oxydation des particules AlCuFeB. Ensuite, l’étude par microscopie électronique a permis d’identifier une nouvelle phase approximante de la phase icosaédrique du système AlCuFe. Il s’agit d’une phase orthorhombique qui à notre connaissance est observée ici pour la première fois. Enfin, les propriétés mécaniques et de frottement de ces nouveaux matériaux sont présentées. Les matériaux composites Al/(AlCuFeB)p élaborés ont des propriétés améliorées par rapport à l’aluminium non renforcé. L’évolution des propriétés est influencée par le taux de particules AlCuFeB et leur état d’oxydation initial. Les propriétés sont améliorées lorsque la fraction volumique de particules augmente mais de façon moindre quand les particules AlCuFeB sont fortement oxydées. / The present study was performed in the framework of a project funded by the ADEME agency (French Agency for Environment and Energy Management), aiming at the reduction of friction loss in car engines through the introduction of new tribological coatings. Our task was to propose our partners new coating materials based on metal-intermetallic nano-composites with optimized friction properties. In a first part, we have prepared by solid state sintering new Al-based composite materials reinforced by quasicrystalline icosahedral particles Al/(AlCuFeB)p. The kinetics of phase transformations resulting from the diffusion of Al matrix to the quasicrystalline particles was studied. It was shown that the destabilization of the icosahedral phase can be avoided by the creation of a diffusion barrier via an oxidation pre-treatment of the AlCuFeB particles. In a second part, the results of a structural study of the composites by transmission electron microscopy are presented. We also describe a new approximant of the quasicrystalline AlCuFe icosahedral phase. This phase was identified as an orthorhombic phase which, to our knowledge, is observed here for the first time. Finally, the mechanical and friction properties of the composites are presented. We show that the composite materials have improved properties compared to aluminium and that their evolution is influenced by the volume fraction of AlCuFeB particles and their initial state of oxidation. The best properties are obtained when the volume fraction of the particles is increased, but in a less pronounced manner when the AlCuFeB particles are strongly oxidized.

This thesis refers to Fe-based biodegradable materials and their potencial aplications in medicine, especially as temporary bone implants. This work generaly summaries aplications of biomaterial in medicine with more interest kept on biodegradable materials and their in-vivo corrosion. The experimental part refers to conduction of porous Fe-based materials with silica addition. The structure of prepared specimens is identified by EDX and XRD analysis. The imersion test and electrochemical studies were conducted to observe corrosion behaviour with respect to different concentration of silica. Potenciodynamic curves were obtained to determine corrosion potencial and corrosion current density of prepared samples.

Dendritic solidification has been the subject of continuous research, also because of its high importance in metal production. The challenge of predicting macroscopic material properties due to complex solidification processes is complicated by the multiple physical scales and phenomena involved. Practical modeling approaches are still subject to significant limitations due to remaining gaps in the systematic understanding of dendritic microstructure formation. The present work investigates some of these problems at the microscopic level of interfacial morphology using phase-field simulations. The employed phase-field models are implemented within a finite-element framework, allowing efficient and scalable computations on high-performance computing facilities. Particular emphasis is placed on the evolution and interaction of dendrite sidebranches in the broader context of dendrite fragmentation, varying and dynamical solidification conditions.

The phase diagram of the Mg-Pd system was redetermined in the range from 50 at. % Mg to 100 at. % Mg. It contains several intermediate phases with some of them being complex metallic alloy phases (CMAs), i.e., characterized by (i) giant unit cells with more than hundred atoms, (ii) inherent disorder, (iii) the presence of a cluster substructure. Phase fields and heterogeneous equilibria of the intermediate phases β- Mg6Pd, γ-Mg57Pd13, δ-Mg56.4Pd13.6, ε-Mg306Pd77, ζ-Mg64Pd17, η-Mg3Pd, θ-Mg5Pd2, ι- Mg2Pd and κ-MgPd were determined. The first five phases are CMAs with Mackay clusters as fundamental structural units. The crystal structure of the most magnesium-rich compound β-Mg6Pd was redetermined. It was found to be more complicated than previously thought due to correlated disorder of only two atom sites in the cubic unit cell. The γ-, ε- and ζ-phases form in a small window of temperature (50 oC) and composition (3 at. %) close to 80 at. % Mg. A new structure type was assigned to Mg3Pd (Cu3P, P63cm). The single phase field of the θ-phase is caused by constitutional vacancies. The new ι-phase crystallizes with NiTi2 structure-type. / Das Phasendiagramm des Systems Mg–Pd wurde im Bereich von 50 bis 100 At.-% Mg neu bestimmt. In diesem Phasendiagramm finden sich mehrere intermediäre Phasen, darunter auch komplexe intermetallische Verbindungen (engl.: complex intermetallic alloys, CMAs). CMAs sind charakterisiert durch (i) große Elementarzellen mit mehr als einhundert Atomen, (ii) intrinsischer Fehlordnung und (iii) dem Vorhandensein einer Cluster- Substruktur. Die Phasenfelder und heterogenen Gleichgewichte der intermediären Phasen β-Mg6Pd, γ- Mg57Pd13, δ-Mg56.4Pd13.6, ε-Mg306Pd77, ζ-Mg64Pd17, η-Mg3Pd, θ-Mg5Pd2, ι-Mg2Pd und κ- MgPd wurden bestimmt. Die ersten fünf der eben genannten Phasen sind CMAs mit Mackay Clustern als fundamentales strukturelles Einheit. Alle übrigen Phasen besitzen einen einfacheren kristallografischen Aufbau. Die Kristallstruktur der Mg-reichsten Verbindung β-Mg6Pd wurde neu bestimmt und ist weitaus sich als komplizierter als bisher angenommen. Die Ausdehnung des Einphasenfeldes von β-Mg6Pd lässt sich jedoch sehr einfach mit korrelierter Fehlordnung von lediglich zwei Atomen in der kubischen Elementarzelle verstehen. Die γ-, ε-, und ζ-Phasen bilden sich in einem schmalen Temperatur- (50 °C) und Zusammensetzungsbereich (3 at. %) nahe 80 at. % Mg. Der Verbindung Mg3Pd (Cu3P, P63cm) wurde ein neuer Strukturtyp zugewiesen. Die Ausdehnung des Einphasenfeldes der θ-Phase lässt sich mit dem Einbau konstitutioneller Leerstellen erklären. Die neue ι-Phase kristallisiert im NiTi2 Strukturtyp.

The glass-forming composition ranges in Al-La-Ni and Mg-TM (Cu, Zn)-Y alloys were predicted using Miedema’s model. Glass-forming abilities of Al-La-Ni alloys and Mg-Cu-RE alloys were studied in terms of reduced glass transition temperature (Trg), supercooled liquid region (∆Tx) and γ parameters. The glass-forming ability parameters of Mg-Cu-RE (RE: rare-earth) alloys were correlated with Mendeleev number. The Miedema model has been used to determine glass-forming composition range in binary Al-La, Al-Ni and La-Ni alloy systems and the ternary Al-La-Ni system by neglecting the ternary interactions. The glass-forming composition range for Al-La, Al-Ni and La-Ni alloy systems extends from 5 to 90 at% La, 30 to 80 at% Ni and 5 to 95 at% Ni, respectively. In these systems the predicted glass-forming composition range is wider than the experimentally observed range. Miedema model, restricting the difference of enthalpy of formation between the amorphous and solid solution phases to within –10000 J/mole to –55000 J/mole gives rise to better prediction of glass-forming composition range compared to the original models. The concept of mixing enthalpy and mismatch entropy has been used in order to quantify Inoue’s criteria of glass formation. The mixing enthalpy and normalised mismatch entropy of the ternary Al-La-Ni alloys, calculated by the extended regular solution model, vary between –12 to –40 kJ/mol and 0.16 to 0.65, respectively. The enthalpy contour plot has been constructed to distinguish the glass-forming compositions on the basis of the increasing negative enthalpy of the composition. Six Al rich Al-La-Ni alloys with nominal compositions Al89La6Ni5, Al85La10Ni5, Al85La5Ni10, Al82La8Ni10, Al80La10Ni10 and Al60La20Ni20 three La rich Al-La-Ni alloys with nominal compositions Al34La33Ni33, Al40La40Ni20 and Al25La50Ni25 have been chosen from the Al-La-Ni ternary phase diagram, to study the glass-forming ability of Al-La-Ni ternary alloy system and the correlation between La-based and Al-based glasses. All the alloys have been prepared using arc melting unit. All the alloy ribbons have been prepared using single-wheel vacuum melt-spinning unit. Two different wheel speeds of 20 m/s and 40 m/s were used for preparing ribbons of all the nine alloys. All the Al-La-Ni compositions, excluding equi-atomic composition (Al34La33Ni33) and Al60La20Ni20, give rise to amorphous phases. The supercooled liquid region and reduced glass transition temperature of this system increases with a decrease in Al content and an increase in La content. The glass-forming ability of the Al rich Al-La-Ni alloys is lower than that of the La-rich Al-La-Ni alloys. The glass-forming ability has been explained by taking into account the binary heat of mixing and the atomic radius mismatch of the constituent elements. Preferential crystallisation takes place during the heat treatment of glassy ribbons. The crystalline products are partially influenced by composition and binary heat of mixing between elements. Mg65Cu25Y10 alloy is a classical glass former of a family of Mg-based alloys. The partial or complete substitution of Y with other rare earth elements has been introduced to correlate the Mendeleev Number with the glass-forming ability parameters: reduced glass transition temperatures (Trg = Tg/Tl), supercooled liquid regions (∆Tx = Tx – Tg) and γ-criterion (TX/(Tg + Tm)). Mg-Cu-RE alloys with nominal compositions Mg65Cu25Y10, Mg65Cu25Y5Gd5, Mg65Cu25Y5Nd5, Mg65Cu25Gd10 and Mg65Cu25Nd10 were chosen for this work. The high reduced glass transition temperature, wider supercooled liquid region and higher γ value of Mg-Cu-Gd-Y amorphous alloy compared to Mg-Cu-Y and Mg-Cu-Nd-Y systems indicates that Mg-Cu-Gd-Y alloys possess higher glass-forming ability. The devitrification of all Mg-Cu-RE glassy alloys used for this work give rise to Mg2Cu (oF48) phase, which is known as anti-Laves phase. The glass-forming composition range for binary and ternary Mg-Cu-Y systems was calculated using Miedema’s model. The development of accurate methods of prediction of glass-forming ability in metallic systems is an important challenge. Pettifor has pioneered the Structure Mapping approach to binary intermetallics. The Pettifor approach can be adapted to the designing of bulk metallic glasses (BMGs). This method has been used to design Al-based and Mg-based BMG’s. Pettifor introduced an integer parameter to characterize the elements, which he called the Mendeleev Number. Essentially, Pettifor’s scheme orders the elements in a sequence of increasing electronegativity. With respect to Mendeleev Number, the Mg-Cu-RE system can be regarded as a binary system, because of the closeness of Mg and Cu (Mg:73, Cu:72, Y:25, Gd:27 and Nd:30). For this system, Mendeleev Number is a more effective parameter than atomic size (Mg: 1.60 Å, Cu: 1.27 Å), as a predictor of glass-forming ability. The effect of Y and rare earth elements on glass forming ability is similar. The atomic number of Y (39) is away from that of the rare earth elements and the Mendeleev Number of Y (25) comes in between those of the rare earth elements. Mg-Zn-Y system is an interesting system for researchers because of higher strength of these alloys. This system draws the crystallographers’ attention due to its composition-dependent structure variations. The Mg-rich RS/PM Mg-Zn-Y alloys yield superior mechanical properties. Therefore, the Mg-rich Mg-Zn-Y system has been chosen to study the microstructural evolution, even though the theoretical calculations for the glass-forming composition range for the Mg-Zn-Y system shows that this system is not a good glass former. Mg-Zn-Y system with nominal compositions Mg97Zn1Y2, Mg97Zn2Y1, Mg97−xZn1Y2Zrx and Mg92Zn6.5Y1.5 were chosen to study the microstructural evolution of these alloys. A small increase in Zn amount (above 2 at.%) in Mg-rich Mg-Y system results in quasicrystalline particles embedded in the matrix, whereas the addition of Zn up to 2 at.% leads to microstructural changes in the α-Mg solid solution.

Les alliages métalliques complexes sont des matériaux qui présentent un ordre à longue distance caractérisé par de grandes mailles comprenant plusieurs centaines d’atomes disposés en clusters. Une propriété caractéristique des CMAs est une conductivité thermique de réseau, dû aux phonons, très faible (~1.3 W/m.K), ce qui donne un intérêt pour leur utilisation comme thermoélectriques. Malgré de récentes avancées sur les connaissances de leurs structures, la nature des modes de vibrations des phonons dans ces réseaux restent une question ouverte : quel est le rôle des clusters ? Est-ce qu’il y a des modes critiques ? Pour répondre à cette problématique, mon projet de thèse a eu pour objectif de comprendre la nature des modes de vibrations à l’échelle atomique et la relation avec la conductivité thermique de réseau sur deux systèmes : la phase o-Al13Co4 qui est un approximant de la phase décagonale AlNiCo et le clathrate Ba8Ge40.3Au5.25, présentant des propriétés thermoélectriques. Mes investigations combinent des expériences de diffusion inélastiques des neutrons et des rayons-X et des simulations à l’échelle atomique.Une analyse détaillée des résultats expérimentaux obtenus par diffusion inélastique sur monocristaux pour les branches acoustiques a permis de mettre en évidence, pour la première fois, un temps de vie fini des phonons acoustiques lorsqu’ils interagissent avec les modes de basses énergies liés aux atomes dans les clusters. Pour les deux systèmes étudiés, nous observons que la branche acoustique n’est plus linéaire et le temps de vie des phonons acoustiques est réduit à quelques picosecondes. Ce faible temps de vie dépend peu de la température comme la conductivité thermique. Les simulations à l’échelle atomique, en utilisant des calculs DFT et des potentiels de pairs oscillants pour des simulations de dynamique moléculaire, ont permis de montrer que ce temps de vie est un effet anharmonique lié au désordre de structure. Les simulations confirment la faible dépendance en température de ce temps de vie. Dans o-Al13Co4, nous avons calculé la conductivité thermique avec la dynamique moléculaire et la méthode de Green-Kubo. Pour Ba8Ge40.3Au5.25 nous avons appliqué un modèle phénoménologique pour l’estimer en utilisant les résultats INS. En conclusion nous démontrons les effets de la complexité structurale sur la conductivité thermique en lien avec la dynamique de réseau. / Complex metallic alloys are long range ordered materials, characterized by large cells, comprising several hundreds of atoms and cluster building blocks. A key property of CMAs is the low lattice thermal conductivity (1.3 W/m. K), which suggests a potential application for CMAs for thermoelectricity. Despite recent advances structure determination, the nature of the phonons modes remains an open question: do the clusters playing a role? Are there critical modes? To tackle this problem, my PhD project aims to understand the vibrational modes at atomic scale and the relation to lattice thermal conductivity in o-Al13Co4 which is an approximant of the quasicrystal, decagonal phase AlNiCo and the clathrate Ba8Ge40.3Au5.25. In this worked we have used Inelastic Neutron and X-ray Scattering experiments and atomic scale simulations, based on density functional theory and empirical pair potentials.A detailed analysis of the results of inelastic scattering experiments on monocrystals for the acoustic branches have shown, for the first time, a finite lifetime for acoustic phonons when they interact with the low-lying dispersion-less excitations due to atoms in the cluster. In both systems, we observe that when an acoustic branch flattens near the zone boundary, the phonon lifetime is a few picoseconds. The phonon lifetime is approximately independent of temperature like the lattice thermal conductivity. Lattice and molecular dynamics simulations with DFT and empirical, oscillating pair potentials show that the finite phonon lifetime is an anharmonic effect, due to structural disorder, explaining the weak temperature of the phonon lifetime. For o-Al13Co4, we have calculated the thermal conductivity with the Green-Kubo method based on equilibrium MD simulations. For Ba8Ge40.3Au5.25 we have developed a phenomenological model based on individual phonon modes. In conclusion, we have demonstrated how structural complexity affects thermal conductivity through the lattice dynamics.

As tensões residuais influenciam o comportamento mecânico dos metais e ligas. O campo de tensões pode determinar o desempenho do material. Então, torna-se de grande importância o estudo e a determinação das tensões residuais. O método padrão da difração de raios X para medir tensões s residuais (Método sen2ψ ) apresenta limitações na determinação de tensões caracterizadas por alto gradiente. È por isso que o estudo do gradiente de tensão superficial é um dos mais importantes problemas teóricos e experimentais em mecânica, especialmente no caso da análise de tensões superficiais que surgem após vários tipos de tratamentos superficiais, como: processamento por laser e tecnologia de implantação iônica. A simulação computacional e os métodos numéricos nos dão a possibilidade de resolver os problemas da determinação dos parâmetros do gradiente de tensão. Neste trabalho foi desenvolvida e aplicada a modelação e simulação computacional para o estudo do comportamento de materiais na presença de distribuições de tensão caracterizadas por alto gradiente. Foi aplicada a modelagem para diferentes tipos de tensões com gradiente para metais puros e ligas metálicas. Foi determinada uma forma para usar a modelagem no caso de materiais compósitos e são apresentados resultados para o caso do silumínio, mostrando o comportamento similar deste material compósito aos materiais puros como aço e alumínio. Usando a análise de Fourier, foi determinada a função de distorção nos casos de perfis alargados pela existência de gradiente de tensão. Foi determinada também a relação entre as funções de atenuação e de tensão na função de distorção e são apresentados resultados para diferentes tipos de distribuições de tensão. Ademais, foi desenvolvida uma metodologia para a determinação dos parâmetros do gradiente de tensão usando a simulação computacional das linhas de difração desenvolvida neste trabalho. A metodologia desenvolvida é baseada na simulação dos perfis de difração distorcidos pelo gradiente de tensão superficial e na análise destas distorções. Das metodologias desenvolvidas para a determinação de distribuições de tensões para o caso de gradiente, a que usa transformada de Fourier é matematicamente correta, porem apresenta dificuldades na determinação das funções de gradiente para casos de distribuições muito complexas e na determinação dos relacionamentos das escalas das funções envolvidas no processo de deconvolução. A alta complexidade deste método e os problemas numéricos que acarreta nos levam a propor a metodologia desenvolvida usando as larguras integrais dos perfis de difração para a determinação aproximada da função de distribuição de tensões. Esta metodologia é de simples aplicação e a sua precisão dependerá da quantidade de dados determinados na simulação direta para a obtenção dos gráficos de calibração. / The residual stress influences the mechanical behavior of metals and alloys. The stress field can determine the material performance. Then, become of great importance the study and determination of residual stress field. The standard X-ray diffraction method for residual stress measurement ent ( sen2ψ Method) presents limitations in the strong stress gradient determination. For that reason, the superficial stress gradient determination is one of the most important theoretical and experimental problem in engineering, especially in the analysis of the stress gradients due to surfaces treatments as laser or ionic implantation. The computational simulation and numerical methods give us the possibility of solve this problems. In this work is developed and applied two methods of modeling and simulation for the study of the material behavior with strong stress gradient. Is applied this modeling technique for different stress distribution in pure metals and metallic alloys. Is established a methodology for the composite material cases and is presents the results for the case of siluminium, showing that the behavior is similar to the other metals testing as steel and aluminum. Also, using Fourier analysis is determined the distortion function for the broadened profiles in the presence of stress gradient. In this work is shown the relation between the attenuation and stress distribution functions in the distortion. Is shown, results for several stress distribution functions. Moreover, is developed a methodology for the determination of stress gradient parameters using the computational simulation of the diffraction lines also developed in this work. The methodology is based in the simulation of the profiles broadened by stress gradient and the analysis of these distortions. Of the developed methodologies for the determination of stress distribution functions for the gradient case, the one that uses Fourier analysis is mathematically correct, but presents difficulties in the determination of the gradient functions for cases of complex stress distributions and in the scales determination of the functions involved in the deconvolution process. The high complexity of this method and the numeric problems that it carries leaves as, to propose the use of the developed methodology using the integral breath of the diffraction profiles for the approximate determination of the stress distribution function. The methodology application is simple and its accuracy will depend on the amount of data determined in the direct simulation to obtain the calibration graphics.

As tensões residuais influenciam o comportamento mecânico dos metais e ligas. O campo de tensões pode determinar o desempenho do material. Então, torna-se de grande importância o estudo e a determinação das tensões residuais. O método padrão da difração de raios X para medir tensões s residuais (Método sen2ψ ) apresenta limitações na determinação de tensões caracterizadas por alto gradiente. È por isso que o estudo do gradiente de tensão superficial é um dos mais importantes problemas teóricos e experimentais em mecânica, especialmente no caso da análise de tensões superficiais que surgem após vários tipos de tratamentos superficiais, como: processamento por laser e tecnologia de implantação iônica. A simulação computacional e os métodos numéricos nos dão a possibilidade de resolver os problemas da determinação dos parâmetros do gradiente de tensão. Neste trabalho foi desenvolvida e aplicada a modelação e simulação computacional para o estudo do comportamento de materiais na presença de distribuições de tensão caracterizadas por alto gradiente. Foi aplicada a modelagem para diferentes tipos de tensões com gradiente para metais puros e ligas metálicas. Foi determinada uma forma para usar a modelagem no caso de materiais compósitos e são apresentados resultados para o caso do silumínio, mostrando o comportamento similar deste material compósito aos materiais puros como aço e alumínio. Usando a análise de Fourier, foi determinada a função de distorção nos casos de perfis alargados pela existência de gradiente de tensão. Foi determinada também a relação entre as funções de atenuação e de tensão na função de distorção e são apresentados resultados para diferentes tipos de distribuições de tensão. Ademais, foi desenvolvida uma metodologia para a determinação dos parâmetros do gradiente de tensão usando a simulação computacional das linhas de difração desenvolvida neste trabalho. A metodologia desenvolvida é baseada na simulação dos perfis de difração distorcidos pelo gradiente de tensão superficial e na análise destas distorções. Das metodologias desenvolvidas para a determinação de distribuições de tensões para o caso de gradiente, a que usa transformada de Fourier é matematicamente correta, porem apresenta dificuldades na determinação das funções de gradiente para casos de distribuições muito complexas e na determinação dos relacionamentos das escalas das funções envolvidas no processo de deconvolução. A alta complexidade deste método e os problemas numéricos que acarreta nos levam a propor a metodologia desenvolvida usando as larguras integrais dos perfis de difração para a determinação aproximada da função de distribuição de tensões. Esta metodologia é de simples aplicação e a sua precisão dependerá da quantidade de dados determinados na simulação direta para a obtenção dos gráficos de calibração. / The residual stress influences the mechanical behavior of metals and alloys. The stress field can determine the material performance. Then, become of great importance the study and determination of residual stress field. The standard X-ray diffraction method for residual stress measurement ent ( sen2ψ Method) presents limitations in the strong stress gradient determination. For that reason, the superficial stress gradient determination is one of the most important theoretical and experimental problem in engineering, especially in the analysis of the stress gradients due to surfaces treatments as laser or ionic implantation. The computational simulation and numerical methods give us the possibility of solve this problems. In this work is developed and applied two methods of modeling and simulation for the study of the material behavior with strong stress gradient. Is applied this modeling technique for different stress distribution in pure metals and metallic alloys. Is established a methodology for the composite material cases and is presents the results for the case of siluminium, showing that the behavior is similar to the other metals testing as steel and aluminum. Also, using Fourier analysis is determined the distortion function for the broadened profiles in the presence of stress gradient. In this work is shown the relation between the attenuation and stress distribution functions in the distortion. Is shown, results for several stress distribution functions. Moreover, is developed a methodology for the determination of stress gradient parameters using the computational simulation of the diffraction lines also developed in this work. The methodology is based in the simulation of the profiles broadened by stress gradient and the analysis of these distortions. Of the developed methodologies for the determination of stress distribution functions for the gradient case, the one that uses Fourier analysis is mathematically correct, but presents difficulties in the determination of the gradient functions for cases of complex stress distributions and in the scales determination of the functions involved in the deconvolution process. The high complexity of this method and the numeric problems that it carries leaves as, to propose the use of the developed methodology using the integral breath of the diffraction profiles for the approximate determination of the stress distribution function. The methodology application is simple and its accuracy will depend on the amount of data determined in the direct simulation to obtain the calibration graphics.

Cette thèse porte sur la fonctionnalisation de surface par microstructuration laser. L’étude expérimentale a consisté à texturer des surfaces d’acier inoxydable avec une source laser impulsionnelle à fibre dopée Yb (1030 nm, 300 fs), dans le but de contrôler leur mouillabilité et de les rendre superhydrophobes. Par une optimisation des conditions d’irradiation, il est possible de conférer à la surface une structuration à double échelle de rugosité. Des structures d’une dizaine de micromètres sont réalisées par ablations successives selon un motif de lignes croisées, sur lesquelles se forment des nanostructures auto-organisées. La simulation du couplage de l’énergie dans la cible a permis de déterminer les paramètres opératoires pour limiter l’accumulation thermique en surface. L’étude fait clairement apparaître le rôle de la texturation dans l’apparition du caractère superhydrophobe de la surface, tout en soulignant l’influence des propriétés physico-chimiques du matériau. / This PhD thesis is about surface functionalization by laser microstructuring. The experimental study consists in texturing stainless steel surfaces with a pulsed Yb fibre laser source (1030 nm, 300 fs), in order to control their wettability and confer to them superhydrophobic properties. With an optimization of the irradiating conditions on the target, it is possible to confer to the surface a dual-scale roughness. By successive ablations according to a pattern of crossed lines, microstructures in the range of tens of micrometres are realized, on which self-organized nanostructures are superimposed. Simulation of the energy coupling in the material allows to determine the process parameters to be used, in order to limit the thermal accumulation and avoid the melting of the surface. This study reveals the role of the laser texturing in the apparition of the superhydrophobic character and emphasizes the influence of the physicochemical properties of the material.