Dissertations / Theses on the topic 'Al-Fe System'

This thesis presents the formation mechanism of intermetallics formed at Fe-Al film interfaces. Al thin films with different initial film thicknesses were coated on low carbon steel substrates by physical vapor deposition (PVD). By annealing the system at different temperatures and for different time intervals, several intermetallic phases were observed. X-Ray, SEM and EDS studies showed that intermetallic phases FeAl2 and Fe2Al5 are most dominant phases which were observed and they formed sequentially on the contrary of intermetallics which formed synchronous in bulk materials.

Les diagrammes d’équilibre sont le point de départ et la ligne directrice qui permet de prévoir et contrôler les phases pouvant se former au cours de différents processus industriels. Bien que l’étude expérimentale soit nécessaire pour les systèmes binaires et ternaires, elle est difficilement envisageable pour déterminer les diagrammes de phases des systèmes d’ordre supérieur sur de larges gammes de composition et de température. Afin de contourner ce problème, la méthode dite CALPHAD (CALculation of PHAse Diagram) a été développée. Son principe consiste à optimiser les paramètres des modèles thermodynamiques utilisés pour décrire l´énergie libre de Gibbs de chaque phase à partir d’informations expérimentales ou estimées (ab-initio). Le modèle appelé « Compound Energy Formalism » (CEF) est largement utilisé pour décrire les phases qui présentent plusieurs sous-réseaux. Ce modèle et ceux qui en dérivent permettent la modélisation d'une grande variété de composés. Les activités menées au cours de ce travail ont permis de développer une nouvelle approche du CEF (NACEF) basée sur une étude mathématique de ses paramètres thermodynamiques. Elle a conduit à une nouvelle formulation de la fonction d'énergie libre de Gibbs faisant intervenir de nouveaux paramètres indépendants. Cette nouvelle approche a été utilisée dans le cadre de ce travail afin de modéliser les phases intermétalliques binaires constituée de deux sous-réseaux présentant des défauts uniquement de type anti-sites (A,B)a(A,B)b. Le système Al-Fe-Nb sur lequel porte notre étude a été choisi en raison de son importance dans la fabrication de nombreuses familles d'alliages tels que les aciers, les alliages légers et plus récemment dans le développement de nouveaux matériaux réfractaires à base Nb pour des applications à hautes températures. Dans ce travail, de nouvelles modélisations des bordures binaires Al-Nb et Fe-Nb et pour la première fois du ternaire Al-Fe-Nb sont proposées en utilisant la NACEF et en s’appuyant sur les informations issues de la littérature ou obtenues dans cette étude
The equilibrium diagrams are the starting point and the guideline to predict and control the microstructure that will form during processing materials. Despite experiments being necessary in binaries and ternaries systems, it is difficult to experimentally determine phase diagrams of higher orders systems over wide ranges of compositions and temperature. The CALPHAD (CALculation of PHAse Diagrams) method was developed in order to solve this problem. The essence is to optimize the parameters of thermodynamic models that describe the Gibbs free energies of each phase aiming to reproduce the experimental and estimated (ab-initio) data. The compound energy formalism (CEF) is widely used in order to describe phases which present several sublattices. It allows the modeling of a large variety of phases and numerous methods have been developed to treat different situations. The activities in this work developed a new approach of the CEF (NACEF) based on a mathematic analysis of the parameters which leads to a new formulation of the Gibbs free energy function evolving new independent parameters in which new independent parameters are obtained to express the Gibbs free energy. This approach was used in this work to describe the intermetallic phases with two-sublattice in which the only defect type is anti-sites (A,B)a(A,B)b. The Al-Fe-Nb system was chosen due to its importance for the manufacturing process of several families of alloys currently used, e.g. steels, light alloys, and also for the development of new materials for high temperatures application. The binaries Al-Nb and Fe-Nb were reassessed and the Al-Fe-Nb system was assessed for the first time using literature information and new experimental data
Os diagramas de equilíbrio são o ponto de partida e a diretriz para prever e controlar a microestrutura ao final do processamento de um material. Apesar de experimentos serem necessários em sistemas binários e ternários, é muito difícil determinar experimentalmente diagramas de fase de sistemas de ordens superiores numa vasta amplitude de composições e temperatura. A fim de solucionar este problema, o método CALPHAD (CALculation of PHAse Diagrams) foi desenvolvido. A essência consiste em aperfeiçoar os parâmetros de modelos termodinâmicos que descrevem as energias livres de Gibbs de cada fase de modo a reproduzir as informações experimentais ou estimadas (ab-initio). O compound energy formalism (CEF) é amplamente utilizado para descrever fases que apresentam várias sub-redes. Ele permite a modelagem de uma grande variedade de fases e vários métodos têm sido desenvolvidos para o tratamento de diferentes situações. As atividades deste trabalho ajudaram a desenvolver uma nova abordagem para o CEF (NACEF) com base em um estudo matemático dos seus parâmetros termodinâmicos que levou a uma nova formulação para função da energia livre de Gibbs envolvendo novos parâmetros independentes. Esta nova abordagem tem sido utilizado como parte do presente trabalho para modelar fases intermetálicas binárias constituídas de sub-redes cujo único defeito é do tipo anti-sítio (A,B)a(A,B)b. O sistema Al-Fe-Nb foi escolhido devido a sua importância para o processo de fabricação de diversas famílias de ligas usadas atualmente, e.g. aços, ligas leves e, além disto, é um sistema importante para o desenvolvimento de materiais para aplicações em altas temperaturas. Neste trabalho os binários Al-Nb e Fe-Nb foram reavaliados e o sistema Al-Fe-Nb foi modelado pela primeira vez utilizando as informações da literatura e novos dados experimentais

Les revêtements galvanisés alliés sont produits par immersion à chaud d’une bande d'acier dans un bain de zinc fondu à environ 460 °C, saturé en fer et contenant de faibles quantités d'aluminium (de 0,1 à 0,135% poids), suivie d’un traitement thermique (jusqu'à des températures voisines de 500-530 °C pendant environ 10 s) afin de déclencher les réactions d'alliation entre le fer et le zinc. La microstructure finale de ce type de revêtement est composée d'une succession de couches stratifiées de phases Fe-Zn et ses propriétés d'usage sont directement liées à la distribution de ces phases dans le revêtement. Les paramètres process à appliquer sur ligne industrielle doivent donc être optimisés pour obtenir la microstructure de revêtement souhaitée avec des coûts minimaux. Le développement d'un tel revêtement passe par différentes réactions complexes : la formation de la couche d'inhibition, la rupture de cette couche, la consommation du zinc liquide et l'enrichissement en fer du revêtement solide. Les cinétiques de ces réactions doivent être étudiées et modélisées séparément afin de contrôler avec précision l'évolution du revêtement au cours du cycle thermique. Dans ce travail, les deux premières réactions ont été étudiées dans le cas des aciers IF Ti. La cinétique de formation de la couche d'inhibition est extrêmement rapide et n’a par conséquent pas été étudiée. L'attention a été portée sur la nature de cette couche et sur les mécanismes responsables de sa formation. Il a été démontré que la couche d'inhibition formée dans des bains classiques pour la production de ces revêtements est composée d'une première couche très mince de Fe2Al5Znx (20-30 nm) sur la surface de l’acier et d’une seconde couche plus épaisse de δ (FeZn7) (environ 200 nm) au-dessus. Lorsque l'acier est immergé dans le bain de zinc, la dissolution du premier dans le second conduit à une sursaturation en fer à l'interface solide / liquide. Une très fine couche de Fe2Al5Znx métastable germe alors sur la surface de l'acier favorisée par des relations préférentielles d’épitaxie avec la ferrite. Par la suite, une couche de δ germe sur la couche de Fe2Al5Znx ce qui permet à la microstructure finale de devenir thermodynamiquement stable. L'effet de la teneur en aluminium du bain sur la nature de la couche d'inhibition a également été étudié. Quand la teneur en aluminium du bain diminue, la couche de Fe2Al5Znx devient discontinue car cette phase devient plus métastable et sa germination sur la surface de l'acier moins probable. Cette étape d’inhibition n'est que transitoire et un traitement thermique prolongé conduira à la rupture de la couche d'inhibition et au développement des réactions Fe-Zn. Le mécanisme de rupture, contrôlé par la diffusion du zinc dans les joints de grains de l'acier, peut être expliqué à l'aide du diagramme de phase ternaire Al-Fe-Zn et résumé en deux étapes : la disparition de la couche de Fe2Al5Znx à l'interface couche d’inhibition / acier résultant de l’enrichissement de cette interface en zinc, et la germination de la phase Г (Fe3Zn10) aux joints de grains de l'acier lorsque la concentration en zinc y devient suffisante. C’est cette germination qui va provoquer localement la rupture de la couche d’inhibition. La cinétique de cette réaction dépend fortement de la composition chimique de l'acier IF Ti et de la teneur en aluminium du bain. D'une part, il apparaît que l'effet de la composition chimique de l'acier sur la cinétique de rupture d'inhibition est contrôlé par la compétition entre deux phénomènes opposés : la vitesse de diffusion du zinc dans les joints de grains de l'acier et la capacité de l'acier à y accumuler les atomes de zinc. D'autre part, la diminution de la teneur en aluminium du bain favorise la discontinuité de la couche de Fe2Al5Znx, ce qui accélère la rupture de la couche d'inhibition car le zinc est supposé diffuser plus rapidement dans δ que dans Fe2Al5Znx
Hot-Dip GalvAnnealed (HDGA) coatings are produced by the immersion of the steel strip into an iron-saturated liquid zinc bath at around 460 °C containing small amounts of aluminium (from 0.1 to 0.135 wt.%, normally) and its subsequent heating (up to temperatures around 500-530 °C for about 10 s, typically) in order to trigger the alloying reactions between iron and zinc. The final microstructure of this kind of coatings is composed of a sequence of stratified Fe-Zn phase layers and its in-use properties are directly related to the phase distribution within the coating. The process parameters to be performed in industrial lines must therefore be optimized in order to obtain a successful coating microstructure with the minimum costs. The development of such a coating passes through different and complex reactions: the inhibition layer formation, the inhibition layer breakdown, the liquid zinc consumption and the iron enrichment of the solid coating. The kinetics accounting for these reactions must be studied and modelled separately in order to accurately control the evolution of the coating along the heat treatment performed in the industrial line. In the present work, the two first reactions were investigated in the case of Ti IF steel grades. The kinetics of the inhibition layer formation is extremely fast and has therefore not been investigated in detail. Concerning this reaction, the focus was given to the nature of this inhibition layer and to the mechanisms accounting for its formation. It has been found that the inhibition layer formed in typical baths for galvannealed coatings production is composed of a very thin layer of the Fe2Al5Znx phase (20-30 nm) on the steel surface and a thicker layer of the δ (FeZn7) phase (around 200 nm) on its top. As the steel strip enters the zinc bath, iron dissolution from the former into the latter leads to an iron supersaturation at the solid / liquid interface. As a result, a very thin layer of metastable Fe2Al5Znx nucleates on the steel surface favoured by preferential epitaxial relationships with ferrite. Subsequently, δ nucleates on the Fe2Al5Znx layer allowing the final microstructure of the inhibition layer to become thermodynamically stable. The effect of the bath aluminium content on the nature of this inhibiting structure has also been studied. As the bath aluminium content is lowered, the Fe2Al5Znx layer becomes discontinuous: the lower the bath aluminium content is, the higher the metastability of Fe2Al5Znx is and the less probable its nucleation on the steel surface is. The inhibition state is only transient and continued heat treatment will lead to the inhibition layer breakdown and the development of the further Fe-Zn alloying reactions. The breakdown mechanism, controlled by the diffusion of zinc towards the steel grain boundaries, can be explained using the Al-Fe-Zn ternary phase diagram and summarized in two steps: the disappearance of the Fe2Al5Znx layer at the inhibition layer / steel interface as a result of the enrichment of this interface in zinc, and the local nucleation of the Г (Fe3Zn10) phase at the steel grain boundaries, breaking the inhibition layer off, when the zinc concentration at these locations becomes high enough. The kinetics accounting for this reaction strongly depends on the Ti IF steel chemical composition and the bath aluminium content. On the one hand, it has been found that the effect of the steel chemical composition on the inhibition layer breakdown kinetics would be ruled by the competition between two opposite phenomena: the rate of zinc diffusion at the steel grain boundaries and the ability of the steel to accumulate the zinc atoms at these locations On the other hand, decreasing the bath aluminium content favours the discontinuity of Fe2Al5Znx, which accelerates the inhibition layer breakdown as zinc is expected to diffuse faster through δ than through Fe2Al5Znx

In the current research, a framework based on classical molecular dynamics (MD) is developed for computational mechanical analyses of complex nanoscale materials. The material system of focus is a combination of fcc-Al and and #945;-Fe₂O₃. The framework includes the development of an interatomic potential, a scalable parallel MD code, nanocrystalline composite structures, and methodologies for the quasistatic and dynamic strength analyses. The interatomic potential includes an embedded atom method (EAM) cluster functional, a Morse type pair function, and a second order electrostatic interaction function. The framework is applied to analyze the nanoscale mechanical behavior of the Al+Fe₂O₃ material system in two different settings. First, quasistatic strength analyses of nanocrystalline composites with average grain sizes varying from 3.9 nm to 7.2 nm are carried out. Second, shock wave propagation analyses are carried out in single crystalline Al, Fe₂O₃, and one of their interfaces. The quasistatic strength analyses reveal that the deformation mechanisms in the analyzed nanocrystalline structures are affected by a combination of factors including high fraction of grain boundary atoms and electrostatic forces. The slopes as well as the direct or inverse nature of observed Hall-Petch (H-P) relationships are strongly dependent upon the volume fraction of the Fe₂O₃ phase in the composites. The compressive strengths of single phase nanocrystalline structures are two to three times the tensile strengths owing to the differences in the movement of atoms in grain boundaries during compressive and tensile deformations. Analyses of shock wave propagation in single crystalline systems reveal that the shock wave velocity (US) and the particle velocity (UP) relationships as well as the type and the extent of shock-induced deformation in single crystals are strongly correlated with the choice of crystallographic orientation for the shock wave propagation. Analyses of shock wave propagation through an interface between Al and Fe2O3 point to a possible threshold UP value beyond which a shock-induced structural transformation that is reactive in nature in a region surrounding the interface may be taking place. Overall, the framework and the analyses establish an important computational approach for investigating the mechanical behavior of complex nanostructures at the atomic length- and time-scales.

The knowledge of phase equilibria and thermodynamic properties of liquid and solid oxides can help us better understand metallurgical, ceramic or geological processes. The main aim of the present study is the critical evaluation and thermodynamic optimization of solid and liquid (MnO-Al2O3 based) oxides which are of interest to the steelmaking and ferro-Mn industries. These newly developed databases coupled with the previous databases can be used along with any software for Gibbs energy minimization to predict the phase relationships and the thermodynamic properties of any relevant system. Usually, thermodynamic databases can save both cost and time, which, otherwise would have been spent to optimize the existing process and develop any new process. The production of steels with higher amounts of manganese and aluminum has gained considerable importance in the recent past. Steels with high concentration of manganese and aluminum like TWIP steel and TRIP steel have exceptional properties which classify them as special steels; needless to say the various range of applications they can cater to. Ferromanganese, which contains a large amount of manganese, is also a very useful product required in the production of high manganese steels. The production of these alloys results in the generation of slags which are rich in MnO and Al2O3. Hence, knowledge of the phase relations between these two components is of utmost importance in order to maximize the efficiency of the production process. Only a very good knowledge of Gibbs energy of all the phases present in the binary system MnO-Al2O3 can allow us to predict the correct equilibrium conditions during the production process. The critical evaluation and thermodynamic optimization of all the available phase diagram data and thermodynamic properties of the system Mn-Al-O have been carried out in the first part of the present work. Thermodynamic modeling for different phases such as slag, spinel (cubic and tetragonal) and bixbyite has been performed using Modified Quasichemical Model, Compound Energy Formalism and random mixing model, respectively. The sublattice structure of solid solution phases were properly taken into account in the thermodynamic modeling and their thermodynamic properties and structural data were reproduced using the physically meaningful model parameters. All the reliable experimental data of the Mn-Al-O system were reproduced within error limits from room temperature to above the liquidus temperatures at all compositions and oxygen partial pressure ranging from metal saturation to air. The present MnAl2O4-Mn3O4 spinel solutions can be integrated with all the other spinel solutions developed earlier to obtain an extensive spinel solution database. This database along with the software for Gibbs energy minimization can be utilized to perform various calculations and predict the phase relations at any given condition. In the next part of the present work, the binary MnO-Al2O3 system was extended to the higher order systems like MnO-Al2O3-SiO2, CaO-MnO-Al2O3, FeO-MnO-Al2O3, MgO-MnO-Al2O3 and CaO-MnO-Al2O3-SiO2. Other calculations related to inclusion engineering in steelmaking were also carried out. This was done to check the accuracy of the database developed for the binary MnO-Al2O3 system. The database of model parameters can be used with thermodynamic software like Factsage for thermodynamic modeling of various industrial and natural processes. Calculations pertaining to prediction of thermodynamic properties of phases, cation distribution in spinel solutions, phase equilibria at any temperature, composition and oxygen partial pressure where no experimental data are available can also be performed.
La connaissance des équilibres de phase et des propriétés thermodynamiques des oxydes solides et liquides peut aider à mieux comprendre les processus métallurgiques, céramiques et géologiques. Le but de cette étude est l'évaluation critique et l'optimisation thermodynamique des oxydes solides et liquides impliquant MnO-Al2O3 qui sont utiles pour les industries de l'acier et du ferromanganèse. Les bases de données développées, couplées avec d'anciennes bases de données, peuvent être utilisées avec n'importe quel logiciel de minimisation de l'énergie de Gibbs pour prédire les équilibres de phase et les propriétés thermodynamiques de tout système. Souvent, les bases de données permettent de sauver temps et argent qui, autrement, auraient pu être utilisés pour optimiser des processus existant ou en développer de nouveaux. La production d'aciers à teneur élevé en Mn et Al a acquis une importance considérable. Les aciers à teneur élevé en Mn et Al, comme les aciers TWIP et TRIP, ont des propriétés exceptionnelles qui les classifient comme aciers spéciaux; inutile de mentionner toutes les applications auxquels ils peuvent répondre. Le ferromanganèse, qui contient de grandes quantités de Mn, est aussi un produit très utile dans la production d'aciers à haute teneur en Mn. La production de tels aciers génère des scories riches en MnO et Al2O3. Par conséquent, la connaissance des relations de phases entre ces deux composés est d'une importance capitale pour maximiser l'efficacité de la production. Seule une bonne connaissance de l'énergie de Gibbs de toutes les phases du système MnO-Al2O3 peut nous permettre de prédire les conditions d'équilibre lors de la production. L'évaluation critique et l'optimisation de toutes les données disponibles de diagrammes de phase et de propriétés thermodynamiques du système Mn-Al-O ont été réalisées dans la première partie de ce travail. La modélisation thermodynamique des différentes phases telles que le laitier, le spinelle (cubique et tétragonal) et la bixbyite a été effectuée, respectivement, à l'aide du Modèle Quasichimique Modifié, du Formalisme de l'Énergie des Composés et du modèle de mélange aléatoire. La structure du sous-réseau des solutions solides fut correctement prise en compte dans la modélisation et les propriétés thermodynamiques et données structurales furent reproduites en utilisant des paramètres ayant une signification physique. Toutes les données expérimentales fiables du système Mn-Al-O ont été reproduites à l'intérieur des limites d'erreur de la température ambiante jusqu'au-dessus du liquidus pour toutes les compositions et à des pressions partielles d'oxygène allant de la saturation en métal jusqu'à l'air. Les solutions de spinelle MnAl2O4-Mn3O4 peuvent être intégrées à toutes les autres solutions de spinelle développées antérieurement pour obtenir une base de données étendue pour le spinelle. Celle-ci, combinée à un logiciel de minimisation de l'énergie de Gibbs, peut être utilisée pour effectuer divers calculs et prédire les relations de phase dans n'importe quelles conditions données. Dans la seconde partie de ce travail, le système MnO-Al2O3 a été ajouté aux systèmes d'ordre supérieur tels que MnO-Al2O3-SiO2, CaO-MnO-Al2O3, FeO-MnO-Al2O3, MgO-MnO-Al2O3 et CaO-MnO-Al2O3-SiO2. Des calculs liés à l'ingénierie des inclusions impliquées dans la fabrication de l'acier ont également été réalisées. Ceci a été fait pour vérifier l'exactitude de la base de données du système MnO-Al2O3. Les paramètres du modèle peuvent être utilisés avec un logiciel comme FactSage pour la modélisation de divers procédés industriels et naturels. Les calculs relatifs à la prédiction des propriétés thermodynamiques des phases, la distribution des cations dans les solutions spinelle et les équilibres entre phases à n'importe quelle température, composition et pression partielle d'oxygène où aucune donné expérimentale n'existe, peuvent également être effectuées.

Two different aluminium alloys, AA6111 (Al-Mg-Si) and AA7055 (Al-Mg-Zn), were chosen as the aluminium alloys to be welded with DC04, and two welding methods (USW and FSSW) were selected to prepare the welds. Selected pre-welded joints were then annealed at T=400 - 570oC for different times. Kinetics growth data was collected from the microstructure results, and the growth behaviour of the IMC layer was found to fit the parabolic growth law. A grain growth model was built to predict the grain size as a function of annealing time. A double-IMC phase diffusion model was applied, together with grain growth model, to predict the thickness of each phase as a function of annealing time in the diffusion process during heat treatment. In both material combinations and with both welding processes a similar sequence of IMC phase formation was observed during the solid state welding. η-Fe2Al5 was found to be the first IMC phase to nucleate. The IMC islands then spread to form a continuous layer in both material combinations. With longer welding times a second IMC phase, θ-FeAl3, was seen to develop on the aluminium side of the joints. Higher fracture energy was received in the DC04-AA6111 joints than in the DC04-AA7055 joints. Two reasons were claimed according to the microstructure in the two joints. The thicker IMC layers were observed in the DC04-AA7055 joints either before or after heat treatment, due to the faster growth rate of the θ phase. In addition, pores were left in the aluminium side near the interface as a result of the low melting point of AA7055.The modelling results for both the diffusion model and grain growth model fitted very well with the data from the static heat treatment. Grain growth occurred in both phases in the heat treatment significantly, and was found to affect the calculated activation energy by the grain boundary diffusion. At lower temperatures in the phases with a smaller grain size, the grain boundary diffusion had a more significant influence on the growth rate of the IMC phases. The activation energies for the grain boundary diffusion and lattice diffusion were calculated as 240 kJ/mol and 120 kJ/mol for the η phase, and 220 kJ/mol and 110 kJ/mol for the θ phase, respectively. The model was invalid for the growth of the discontinuous IMC layers in USW process. The diffusion model only worked for 1-Dimensional growth of a continuous layer, which was the growth behaviour of the IMC layer during heat treatment. However, due to the highly transient conditions in USW process, the IMC phases were not continuous and uniform even after a welding time of 2 seconds. Therefore, the growth of the island shaped IMC particles in USW was difficult to be predicted, unless the nucleation stage was taken into consideration.

Le développement d’une troisième génération d’aciers Fe-Mn-Al-C à structure duplex, pour des teneurs en Mn et Al inférieures à 8 %mass, pourrait être une réponse prometteuse aux objectifs d’allègement de 20% des véhicules automobiles, tout en garantissant des propriétés de haute résistance mécanique et haute formabilité.Le choix des nuances et l’optimisation des conditions d’élaboration nécessitent de prévoir en particulier les compositions et proportions des phases existantes en fonction de la route métallurgique. Une base de données thermodynamiques fiable et précise est donc requise. Cependant les données de la littérature sur le système quaternaire Fe-Mn-Al-C, dans les domaines de composition envisagés, sont limitées.Ce mémoire est consacré à l’établissement des équilibres de phases ferrite-α, austénite-γ et carbure-κ (Fe,Mn)3AlC entre 700 et 1000°C par une approche couplée d’expériences ciblées et de modélisation thermodynamique. Pour appuyer l’évolution expérimentale des fractions de phases et des compositions, une modélisation cinétique (DICTRA) est proposée. La cinétique de formation de l’austénite en fonction de la composition de l’alliage et de la température de maintien dans le domaine intercritique a été caractérisée. Les phases en équilibre, caractérisées par DRX, MEB, microsonde, sont représentées sous forme de conodes α/γ, γ/κ, α/γ/κ, ce qui permet de définir les domaines de stabilité de l’austénite et du carbure κ. Ces données expérimentales sont utilisées pour affiner la description thermodynamique du système quaternaire mais il est nécessaire de réviser la modélisation du carbure κ
A third generation of Fe-Mn-Al-C steels with a duplex structure, for Mn and Al contents less than 8%mass, could be a promising response to the 20% weight lightening of automotive vehicles, by keeping a high strength and a high formability.The knowledge of the corresponding quaternary phase diagram serves as a roadmap for the choice of compositions and the optimization of elaboration conditions. A reliable and precise thermodynamic database is therefore required. However, the literature data on the Fe-Mn-Al-C quaternary system in the targeted domains are limited.This study is devoted to the establishment of phase equilibria involving ferrite-α, austenite-γ and carbide-κ (Fe,Mn)3AlC between 700 and 1000°C by a coupled approach of experiments and thermodynamic modeling. A kinetic model (DICTRA) is proposed to support the experimental evolution of phase fraction and composition. The kinetics of austenite formation as a function of the alloy composition and of the maintaining temperature in the intercritical domain have been calculated. The phases in equilibrium, characterized by XRD, SEM, EPMA, are represented as α/γ, γ/κ, α/γ/κ tie-lines in order to specify the stability fields of γ and κ. These data are used to refine the thermodynamic description of the quaternary system but it is necessary to revise the modeling of κ carbide

The corrosion properties of magnesium alloys strongly depend on the alloy composition and impurities. Heavy elements like nickel, and iron have low solubility in solid magnesium. The dissolved elements in molten magnesium precipitate out on solidification and form intermetallic particles that are the cause of corrosion. Iron content should be kept below the standards specified by ASTM B94/94 using aluminium and manganese. Manganese forms intermetallic particles with iron and aluminium thereby lowering the solubility of iron, and these particles are cathodic compared to magnesium matrix. No method for the removal of nickel has been known previously. Dissolution was the only method to lower the nickel content. Published solubility data for nickel in pure magnesium is inconsistent and not available for magnesium alloys. Therefore various systems are studied to determine the behaviour of nickel in Mg-Al alloys. Methods for removal of nickel from Mg-Al alloys are also discussed.

The interaction of phosphate with Fe(III) and Al(III) is important in soils, wastes and other systems of environmental significance. The goal of this research was to characterize phosphate sorption in single- and mixed Fe- and Al-oxide systems using XANES (X-ray absorption near edge structure spectroscopy). The specific objectives of this research were: 1) To determine the quantitative distribution of phosphate between Fe-and Al-oxide minerals in mixtures containing these minerals; 2) To assign XANES spectral features for phosphate associated with Fe(III) or Al(III) to specific electronic transitions; and 3) To characterize adsorption versus surface precipitation in single- and binary mixtures of Fe- and Al-oxide minerals. Phosphate was sorbed in single-mineral aqueous suspensions of ferrihydrite (ferric hydroxide), boehmite (aluminum oxyhydroxide), goethite (iron oxyhydroxide), or non-crystalline (non-xl) Al-hydroxide, and mixtures of ferrihydrite/boehmite, goethite/boehmite, and ferrihydrite/non-xl Al-hydroxide at pH 6. Samples were reacted at 22 degrees Celsius for 42 h. Phosphate sorption isotherm trends for mixed-mineral systems were L-curves and were intermediate to those of the respective minerals in the mixture. Phosphorus K-XANES spectra for phosphate on Fe- vs. Al-oxide minerals differed in that a weak doublet peak was observed for Fe-oxides on the low-energy side of the P K-edge, i.e., in the pre-edge region. The quantitative distribution of phosphate between ferrihydrite and boehmite in mixtures of these minerals was determined using linear combination fitting (LCF) analysis of the XANES pre-edge region. Results showed that phosphate essentially distributed itself in proportion to the maximum phosphate sorption capacity of each of these minerals. Using a XANES fitting procedure, phosphate was found to show a greater apparent preference for boehmite and non-xl Al-hydroxide minerals in goethite/boehmite and ferrihydrite/non-xl Al-hydroxide mixtures, respectively. To interpret XANES spectra based on molecular bonding configuration, spectral features were assigned to specific electronic transitions using bonding arguments supported by extended Huckel (EH) model computations of molecular orbital energies (projected density of states-PDOS). Experimental evidence (both XANES and UV-visible spectroscopy) was given for the white-line peak in Fe(III)/phosphate systems being caused by a dipole allowed transition of a P 1s electron to a P(3p)-O(2p) antibonding molecular orbital. Similarly, the white-line peak in Al-phosphate systems was assigned to a dipole allowed transition into a Al(3p)-O(2p)-P(3p) antibonding molecular orbital. The pre-edge feature in XANES spectra was assigned to a dipole allowed transition into a Fe(4p)-O(2p) antibonding molecular orbital. Using these XANES spectral assignments, the increase in FWHM (full width at half maximum height) of the white-line peak in XANES spectra indicated precipitation. Based on a linear increase in FWHM with increasing sorbed phosphate concentration, Al-phosphate surface precipitation occurred in boehmite and non-xl Al-hydroxide systems. On the contrary, no evidence was found for Fe-phosphate precipitation in single-mineral systems of goethite and ferrihydrite. Surface precipitation occurred in goethite/boehmite mixtures following similar trends as in boehmite, but no evidence for surface precipitation was found in ferrihydrite/non-xl Al-hydroxide mixtures over the range of phosphate studied (up to 1230 mmol/ kg). In these mixtures, mineral interactive effects apparently inhibited Al-phosphate precipitation as occurred when phosphate was reacted with non-xl Al-hydroxide alone. Furthermore, phosphate showed a trend of affinity preference for non-xl Al-hydroxide with increasing adsorbed P concentrations in the ferrihydrite/non-xl Al-hydroxide mixtures.

Le but de ce travail est la determination des enthalpies de formation a 25\c des composes ternaires du systeme al-fe-si et d'une phase ternaire du systeme al-mn-si, en vue de pouvoir calculer le systeme quaternaire a l'aide de programmes informatises. Ce travail comporte trois parties. La premiere partie a consiste en l'elaboration des phases ternaires, ceci a ete realise au centre d'etude de chimie metallurgie du cnrs a vitry (94). La deuxieme partie a consiste a verifier que les alliages etaient reellement monophases et d'adapter le traitement thermique approprie, pour cela les techniques de metallographie et de radiocristallographie ont ete utilisees. Ensuite une mesure precise de la composition des alliages a ete realisee a l'aide d'une appareillage d'absorption atomique. Des mesures d'analyses thermiques (dsc) ont ete realisees sur tous les alliages, afin de determiner les temperatures des invariants et des liquidus. La troisieme partie consiste a determiner les enthalpies de formation a 25\c des alliages al-fe-si par calorimetrie de chute dans un bain d'aluminium liquide a 800\c, sous atmosphere d'argon et de verifier la consistance des mesures. L'alliage almnsi ayant ete elabore et controle par une societe (norvegienne), nous avons determine l'enthalpie de dissolution a dilution infini de manganese dans l'aluminium a 800\c puis l'enthalpie de formation de l'alliage almnsi a 25\c.

The aim of this work was to improve the heat capacity estimation of a material for usage within a CALPHAD-type assessment. An algorithm is derived that estimates the trend of heat capacity with temperature based on zero Kelvin properties and the thermal expansion coefficient at the Debye temperature. The algorithm predicts not only the trend of heat capacity but also the temperature trend of the volume and the bulk modulus, which can be also included in new thermodynamic databases. The algorithm is used to assess thermophysical properties of the intermetallic phases eta (Fe2Al5), epsilon~(Fe5Al8) and tau4 (FeAl3Si2). The heat capacity of the intermetallic phases zeta, eta, theta and epsilon of the Al-Fe system and of tau4 of the Al-Fe-Si system was measured using DSC. For the phases zeta, eta, and theta, a non-linearly increasing heat capacity approaching the melting temperature was observed. In addition, the heat capacity of three bcc-based Al-Fe samples including the B2-->A2 transition were determined. The Al-rich section of the Al-Fe phase diagram was studied using DTA and quenching experiments. The homogeneity ranges of the intermetallic phases were determined using SEM/WDS measurements. Based on own and literature values, a thermodynamic description of the Al-Fe system was assessed including the modelling of A2/B2 ordering and the homogeneity range of all intermetallic phases. In addition, thermodynamic parameters of the Al-Fe-Si, Al-Fe-Mg, and the Fe-Mg-Si system were assessed to obtain a thermodynamic description of the Al-rich side of the Al-Fe-Si-Mg system, which can be used to study phase transitions of typical A356-aluminium alloys.

碩士
國立清華大學
工程與系統科學系
95
A single-walled carbon nanotube (SWNT) is a graphene layer rolled up into a cylinder, with its end cap structure similar to the half of C60. Depending on their (m, n) indices, SWNTs have different electrical properties. SWNTs can be metallic, semiconducting or small-gap semiconducting, depending on their structure and diameters. So control of the diameter of single-walled carbon nanotubes becomes the challenge of developing SWNTs-based nanoelectronic devices. A reliable method of controlling the diameter distribution of single-walled carbon nanotubes is needed. In this thesis, we presented a study of synthesizing high quality SWNTs with narrow diameter distribution by optimizing the Mo/Fe/Al/SiO2 catalytic system in thermal CVD. It was found that thickness of Al layer had significant effects on the quality of SWNTs. A high quality, almost bundle-free, SWNTs with G/D area ratio of 45 was obtained by optimizing the catalytic system with Mo(0.5nm)/Fe(1nm)/Al(5nm)/SiO2(100nm) multi-layer catalyst and the CVD growth process parameters. A narrow diameter distribution of 0.8~1.4 nm (mostly ~1.2 nm) was achieved. Finally, we summarized these results by proposing a SWNTs growth mechanism with multi-layer catalytic system to explain the effects of varying the thickness of each layer.

Superalloys offer high temperature strength, excellent creep, corrosion and oxidation resistances, microstructural stability and good fatigue life at elevated temperatures. The composition of the superalloys has been modified continuously to improve the properties. The addition of Pt improves oxidation resistance without compromising the mechanical properties of the superalloys. To further enhance the performance of the superalloy components, various coatings are applied on them. The-(NiPt)Al intermetallic compound bond coats, which are presently utilized, have certain drawbacks. Diffusion of Al from the bond coat to superalloy during service leads to accumulation of stress near the bond coat. The refractory elements present in superalloy precipitate as topological close packed (TCP) phases in the interdiffusion zone. Consequently, a Pt enriched γ(Ni) + γ’(Ni3Al) phase mixture has been proposed as a possible alternative since TCP phases do not form in the interdiffusion zone. In this thesis, diffusion studies are performed on several binary and ternary systems with the primary purpose of understanding the effect of Pt in Ni based superalloys and also in γ + γ’ phase mixture bond coats. Further, a detailed interdiffusion study is conducted in Mo- and W- based binary and ternary systems to understand the growth of the TCP phases. By performing bulk and multifoil diffusion couple experiments, different diffusion parameters like, inter, intrinsic, tracer, impurity diffusion coefficients and activation energy that are necessary to understand the diffusion mechanism are determined. Additionally using the nanoindentation technique on diffusion couples, variation of mechanical properties such as, hardness and modulus with composition is studied. First, interdiffusion in Ni-Pt, Co-Pt, Co-Ni, Ni-Fe and Co-Fe binary systems is examined. In Ni-Pt and Co-Pt, experimental results show that Pt is the slower diffusing species at all compositions. In both the systems, driving force is found to be the reason for higher values of intrinsic diffusion coefficients observed in the range of 40-60 at. % Pt. Contribution of vacancy wind effect on diffusion parameters is found to be negligible. It is found from the multifoil diffusion couple experiments that Ni is the faster diffusing species in the Co-Ni system. Bulk diffusion couple experiments are conducted in the Co-Ni-Pt and Co-Ni-Fe systems, by coupling binary alloys with the third element. Uphill diffusion is observed for Co and Ni in Pt rich corner of the Co-Ni-Pt system. Main and cross interdiffusion coefficients are calculated at the compositions where two diffusion profiles intersect. In both the systems, the main interdiffusion coefficients are positive over the whole composition range and the cross diffusion coefficients show both positive and negative values at different regions. Hardness measured by performing the nanoindentations on diffusion couples of both the systems, shows the higher values at intermediate compositions. The effect of Pt in and’ phases of Ni-Al system are examined by conducting interdiffusion experiments between Ni(xPt) alloys and (NixPt)40Al alloy of β phase, so that both and’ phases grow in the interdiffusion zone. The interdiffusion coefficients in Ni-Al binary system increases with the Al content in the -phase, and they do not vary significantly with composition in the ’ phase. The average effective interdiffusion coefficients of Ni and Al in the and ’ phases increase with the addition of Pt. Nanoindentation studies on diffusion couples show that the hardness of both and ’ phase increases with the addition of Pt. In the +’ phase mixture bond coats, effect of Pt on interdiffusion of major alloying elements of CMSX4 superalloys are discussed. A phase mixture of and ’ with increasing Pt content is coupled with CMSX4 superalloy. The addition of Pt to the +’ phase mixture increases the diffusion rate of Ni, while the diffusion rate of Al, decreases with the addition of 5% Pt, and increases with further addition of Pt. No significant change in the diffusion rates of Co or Cr is observed. The growth kinetics and diffusion in systems (both binary and ternary) with TCP phases are examined. Interdiffusion studies performed in Co-Mo system show significant volume change because of the growth of the phase. Intrinsic diffusion coefficient of Mo is found to be higher than that of Co. Diffusion studies conducted in Ni-Mo system show reasonably low activation energy in the phase, indicating the grain boundary controlled diffusion process. The Co-Ni-Mo and Co-Ni-W ternary phase diagrams are revisited and the phase boundary composition of the TCP phases are found to be different from those reported earlier. Following, the average effective interdiffusion coefficients are calculated and compared with the data calculated in the binary systems to examine the role of the third element. It is noticed that the average effective interdiffusion coefficients in the Co(Ni,Mo) and Co(Ni,W) solid solution increases with the addition of Ni. On the other hand, these diffusion coefficients decrease with the addition of Ni in thephase in both the systems. The role of the driving force for diffusion and possible change in defect concentrations on different sublattices are discussed.