Ce travail de thèse, porte sur le développement d’une technique originale de traitement de surface par jet d’azote cryogénique. Ce procédé a été initialement développé pour le décapage et le nettoyage des surfaces. Il est ici utilisé pour obtenir un durcissement superficiel sans altération ou endommagement de la surface du matériau traité. Sous certaines conditions, dans un premier temps, nous avons appliqué la technique jet d’azote en conditions statiques de traitement afin de cartographier les domaines d’utilisation du jet en fonction des paramètres de procédé (distance de tir et temps d’exposition). On montre, un durcissement superficiel sans endommagement de la surface du matériau cible (acier austénitique AISI 316L). L’influence de la distance à laquelle la surface est traitée (distance de tir) et le temps d’exposition du jet sur l’évolution de la microstructure, le durcissement et l’endommagement en surface de l’acier AISI 316L a été étudiée. Des analyses par microscopie électronique à balayage, des analyses d’images ainsi que des mesures de microdureté ont été effectuées sur les microstructures des surfaces traitées pour quantifier les effets de traitement par jet d’azote. Le durcissement en surface, du essentiellement à la transformation martensitique, est ainsi quantifié selon les conditions d’essai. Dans un second temps, un traitement en conditions cinématiques a été réalisé en vue d’obtenir un durcissement superficiel sans endommagement de matière. Les essais de traitement en conditions cinématiques ont été essentiellement réalisés sur les surfaces d’aciers austénitiques instables, l’AISI 316L et l’AISI 304L et, ponctuellement sur l’acier stable, l’AISI 310s. L’influence de la vitesse d’avance du jet et la pression de consigne sur l’évolution de la microstructure, les fractions de martensites formées et le niveau de durcissement en surface d’aciers AISI 316L et AISI 304L ont été étudiées. Les analyses EBSD, MEB ainsi que les mesures de microdureté réalisées sur les surfaces traitées ont permis de mettre en évidence le lien entre le niveau de durcissement et la quantité de martensite induite. Le durcissement de la surface de l’acier AISI 310s, qui reste très faible comparé à celui d’aciers instables, est le résultat de l’écrouissage de sa phase austénitique. Il a été également montré qu’un traitement avec un double passage du jet conduit à l’amélioration de la microdureté en surface des trois aciers traités / This work focuses on the study of an original surface treatment technique that uses supercritical cryogenic nitrogen jet. This process was initially designed for environmentally friendly surface cleaning, where indeed such gas recycles in the air after operation. In the present work, this technique is implemented for surface hardening use without damage of the surface to be treated. Two types of operation cases are studied: static jet tool impingement, cinematic using jet tool scanning on the top surface. In fact, these two static and cinematic treatment cases can be used in industrial operations. In the first stage, the treatment was performed under static conditions in order to map the domains of use of the process. Variation of the experimental parameters (standoff distance and dwell time - treatment time-) made possible to define several uses of the nitrogen jet. In particular the hardening without any damage of the surface of the material to be treated such as AISI 316L stainless steel. Thus, the influence of the standoff distance and the dwell time on the evolution of surface microstructure and damage and hardening was studied. To quantify the effects of nitrogen jet on the microstructure, SEM (Scaning Electrons Microscope) observations and micro hardness measurements were carried out on the treated surfaces. As a result, for different conditions of treatment, the relationship between hardness and martensite rate during surface transformation process, is shown and plotted. Secondly, we focus on hardening without surface damage. The treatments were essentially carried out on both AISI 316L and AISI 304L metastable stainless steels. The influence of both torch velocity and jet static pressure on the variation of microstructure, martensite fractions and hardening level, was also studied and discussed. Thanks to both SEM/EBSD analysis and micro hardness measurements, the relationship between martensite rate and increase of hardness, is highlighted. It is also established that the treatment using several passes allows to increase the surface micro hardness without damage. Finally, it is found that, for some particular working parameters, the nitrogen jet process can also be used for surface hardening without martensitic transformation

Made available in DSpace on 2015-05-08T14:59:54Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 2777493 bytes, checksum: eef0d78a961712e8d2857815f543d068 (MD5) Previous issue date: 2014-02-14 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Five different compositions of the Cu-Al-Mn shape memory alloys were cast under ambient atmosphere, and characterized thought optical microscopy, X-ray diffraction, and differential scanning calorimetry. Shape recovery and superelasticity was evaluated using mechanical tensile test, always comparing the results between them and observing the influence of manganese concentration on termomecanicals properties. Results were encountered regarding the shape recovery of approximately 5%, however, the compositions of the alloys Cu-Al-Mn analyzed have high brittleness, which should be kept to a minimum. Another favorable aspect in relation to these alloys is highly sensitive to the composition of its components, where 1 %peso increase in manganese content will reduce the Mi temperature around 60 K, which facilitates the handling of their transformation temperatures according to application needs. / Cinco composições diferentes da liga Cu-Al-Mn foram elaboradas em atmosfera ambiente, e caracterizadas por Microscopia Ótica, Microscopia Eletrônica de Varredura, Difratometria de Raios-X e Calorimetria Diferencial de varredura. Em seguida suas propriedades termomecânicas foram medidas através de ensaios de tração, superelasticidade e recuperação de forma, sempre observando a influência da concentração do manganês em suas propriedades. Foram encontrados resultados promissores em relação à recuperação de forma de aproximadamente 5%, em contrapartida, as composições das ligas Cu-Al-Mn analisadas, possuem elevada fragilidade, algo que deve ser reduzido ao máximo. Outro aspecto favorável em relação a estas ligas é a alta sensibilidade à composição de seus componentes, onde 1% em peso de aumento no teor de manganês irá reduzir a temperatura Mi em torno de 60 K, o que facilita a manipulação de suas temperaturas de transformação de acordo com a necessidade da aplicação.

FundaÃÃo de Amparo à Pesquisa do Estado do Cearà / Os arames eletrodos de aÃos inoxidÃveis martensÃticos tÃm seu emprego voltado nas soldagens de ligas ferrÃtica/martensÃticas por fornecer apropriada resistÃncia mecÃnica e à corrosÃo. Este trabalho apresenta um estudo do efeito da estrutura martensÃtica formada a baixa temperatura no metal de solda de juntas soldadas multipasse, sobre o nÃvel de tensÃes residuais, propriedades mecÃnicas e tenacidade, comparando com os resultados obtidos por soldagens convencionais. O arame eletrodo utilizado foi Fe-12%Cr-5%Ni, com temperatura de transformaÃÃo martensÃtica inferior a 200 ÂC, soldando secÃÃes de tubulaÃÃes de aÃo API 5L classe B utilizado em extraÃÃo e transporte de petrÃleo. As soldagens dos corpos de prova foram realizadas de modo automÃtico. Amostras de cada cordÃo do enchimento foram retiradas para a realizaÃÃo das anÃlises de composiÃÃo quÃmica, microdureza, caracterizaÃÃo microestrutural por microscopia Ãtica (MO) e microscopia eletrÃnica de varredura (MEV), difraÃÃo de raios-X e ensaio termo-magnÃtico. Realizando o mapeamento da textura cristalogrÃfica e determinando as variantes cristalogrÃficas da formaÃÃo da martensÃta por Electron Backscattering Diffraction (EBSD). Os resultados de tensÃes residuais mostraram que as juntas soldadas com eletrodo de baixa temperatura de transformaÃÃo martensÃtica apresentaram maiores intensidades de tensÃes residuais compressivas comparados com o eletrodo comum, e nÃo influenciaram no surgimento de textura cristalogrÃfica nos cordÃes de topo. A maior energia de soldagem influenciou significativamente no aumento da tenacidade do cordÃo e contribuiu para o surgimento de austenita revertida. / Os arames eletrodos de aÃos inoxidÃveis martensÃticos tÃm seu emprego voltado nas soldagens de ligas ferrÃtica/martensÃticas por fornecer apropriada resistÃncia mecÃnica e à corrosÃo. Este trabalho apresenta um estudo do efeito da estrutura martensÃtica formada a baixa temperatura no metal de solda de juntas soldadas multipasse, sobre o nÃvel de tensÃes residuais, propriedades mecÃnicas e tenacidade, comparando com os resultados obtidos por soldagens convencionais. O arame eletrodo utilizado foi Fe-12%Cr-5%Ni, com temperatura de transformaÃÃo martensÃtica inferior a 200 ÂC, soldando secÃÃes de tubulaÃÃes de aÃo API 5L classe B utilizado em extraÃÃo e transporte de petrÃleo. As soldagens dos corpos de prova foram realizadas de modo automÃtico. Amostras de cada cordÃo do enchimento foram retiradas para a realizaÃÃo das anÃlises de composiÃÃo quÃmica, microdureza, caracterizaÃÃo microestrutural por microscopia Ãtica (MO) e microscopia eletrÃnica de varredura (MEV), difraÃÃo de raios-X e ensaio termo-magnÃtico. Realizando o mapeamento da textura cristalogrÃfica e determinando as variantes cristalogrÃficas da formaÃÃo da martensÃta por Electron Backscattering Diffraction (EBSD). Os resultados de tensÃes residuais mostraram que as juntas soldadas com eletrodo de baixa temperatura de transformaÃÃo martensÃtica apresentaram maiores intensidades de tensÃes residuais compressivas comparados com o eletrodo comum, e nÃo influenciaram no surgimento de textura cristalogrÃfica nos cordÃes de topo. A maior energia de soldagem influenciou significativamente no aumento da tenacidade do cordÃo e contribuiu para o surgimento de austenita revertida. / The filler metal of martensitic stainless steels has its used welding of alloy ferritic / martensitic to provide suitable mechanical strength and corrosion. This research accomplishes a study of the mechanical and metallurgical properties correlate with the crystallography texture and variants selection during the martensitic transformation in the weld metal. The filler metal used was Fe-12% Cr-5% Ni, with martensitic transformation temperature less than 200ÂC, welding sections of steel pipeline API 5L grade B used in oil extraction and structure offshore. The welding of the specimens was performed automatically. Samples of each layer of filling were removed for analysis during chemical composition, microhardness, microstructural characterization by optical microscopy (OM) and scanning electron microscopy (SEM), X-ray diffraction and thermo-magnetic test. Itâs intend to accomplish a mapping of texture and variants selection of the formation of martensitic morphology by Electron Backscattering Diffraction (EBSD). The results have shown that residual stresses in welded joints filler metal of low-temperature martensitic transformation had higher compressive residual stress intensities compared to the common electrode, and did not influence the appearance of crystallographic texture on top of beads. The high heat input significantly influenced the increase toughness beads and contributed to the appearance of reverted austenite. / The filler metal of martensitic stainless steels has its used welding of alloy ferritic / martensitic to provide suitable mechanical strength and corrosion. This research accomplishes a study of the mechanical and metallurgical properties correlate with the crystallography texture and variants selection during the martensitic transformation in the weld metal. The filler metal used was Fe-12% Cr-5% Ni, with martensitic transformation temperature less than 200ÂC, welding sections of steel pipeline API 5L grade B used in oil extraction and structure offshore. The welding of the specimens was performed automatically. Samples of each layer of filling were removed for analysis during chemical composition, microhardness, microstructural characterization by optical microscopy (OM) and scanning electron microscopy (SEM), X-ray diffraction and thermo-magnetic test. Itâs intend to accomplish a mapping of texture and variants selection of the formation of martensitic morphology by Electron Backscattering Diffraction (EBSD). The results have shown that residual stresses in welded joints filler metal of low-temperature martensitic transformation had higher compressive residual stress intensities compared to the common electrode, and did not influence the appearance of crystallographic texture on top of beads. The high heat input significantly influenced the increase toughness beads and contributed to the appearance of reverted austenite.

L’alliage PuGa 1at.% stabilisé en phase δ n’existe que dans un état métastable et présente donc un caractère très sensible à l’environnement extérieur (transformation de phase sous sollicitations thermiques et mécaniques). L'originalité de ce travail consiste en une caractérisation quantitative In Situ de la transformation martensitique δ → α’ (nature et fraction des phases en présence, évolutions de paramètres de maille et microdéformations générées...) ainsi qu’en une étude microstructurale associées à différentes conditions de transformation (températures et sollicitations mécaniques). Le caractère isotherme pour la transformation de l’alliage PuGa 1at.% a été confirmé. L’analyse des cinétiques de transformation isotherme menée avec le modèle de Pati et Cohen a permis de discuter les différents phénomènes mis en jeu lors de la transformation (germination autocatalytique, interaction entre le nouveau variant formé et la matrice). Les modifications engendrées lorsque la matière est contrainte (augmentation de température Ms, orientation des produits de transformation) ont pu être calculées à partir du formalisme de Patel et Cohen. Des simulations numériques microstructurales ont également été réalisées afin d’appréhender l’effet des contraintes associées à la déformation libre de transformation sur la morphologie et l’arrangement des plaquettes de martensite, en regard aux observations expérimentales par microscopie optique et microscopie électronique à balayage. La confrontation directe de l’ensemble des résultats obtenus a permis de mettre en évidence l’influence majeure des contraintes accumulées dans la matière durant la transformation. En effet, alors qu’un effet auto-catalytique contrôle largement les premiers instants de la cinétique, une accumulation d’interactions mécaniques défavorables apparait progressivement expliquant le caractère partiel de cette transformation. L’étude de la transformation martensitique transformée à basses températures et sous contraintes a été complétée par l’étude de sa réversion en phase δ afin de comprendre les différents mécanismes pilotant cette réversion. Ce travail a montré l’existence d’une compétition entre deux modes de réversion direct et indirect, ce dernier étant étroitement lié à la mobilité du gallium et la stabilité thermodynamique des différentes phases de l’alliage en température / The δ-stabilized PuGa 1at.% is only in a metastable state and therefore is very sensitive to the external environment (phase transformation under thermal and mechanical loading). The originality of this work consists in a quantitative In Situ characterization of the δ → α' martensitic transformation (nature and amount of existing phases, evolution of lattice parameters and induced microstrains …) as well as a microstructural study conducted under different transformation conditions (temperature and mechanical loading). The isothermal character of the transformation kinetics in the PuGa 1at.% was confirmed. The analysis of the kinetics from the Pati and Cohen formalism gave the opportunity to investigate the mechanisms involved during the transformation (autocatalytic nucleation, interaction between the new variant formed and the matrix). Modifications induced in a stressed material (increase in temperature Ms, crystallographic orientation of transformation products) were calculated from the Patel and Cohen formalism. Microstructural numerical simulations were also performed in order to understand the effect of elastic interactions associated with transformation eigenstrain on the martensite plate morphology and plates arrangement in regard of the observations by SEM and OM. The direct confrontation of all results highlighted the large influence of accumulated stresses in the material during the transformation. Indeed, while an autocatalytic effect controls the first steps of the kinetic, an accumulation of unfavorable mechanical interactions occurs gradually explaining the partial nature of this transformation. The study of the martensitic transformation occurring at low temperature and under stresses was complemented by the study of its reversion into the δ phase in order to grasp the different mechanisms driving this reversion. This work shows the existence of competition between direct and indirect reversion modes, the latter being closely related to the mobility of gallium and to the thermodynamic stability of different phases of the alloy versus temperature

De nos jours, l’essor de la miniaturisation est un paramètre clé pour la réalisation de microsystèmesde plus en plus complexes. Les recherches sur l’élaboration de matériaux « intelligents », onttoujours suscité un grand intérêt. Dans ce cadre, on se propose d’étudier l’alliage à mémoirede forme magnétique Ni2MnGa sous la forme de couche mince. Ce matériau a la propriétéparticulière de répondre aux sollicitations mécaniques, thermiques et magnétiques. Le couplage deces effets permettrait l’élaboration de micro-actionneurs usuellement réalisés à partir d’assemblagescomplexes. Cette étude pluridisciplinaire traite de l’influence des paramètres d’élaboration sur lespropriétés fonctionnelles des couches minces. L’originalité de ce travail de thèse réside dansl’emploi de substrats en silicium dans la perspective d’une transposition technologique. Un procédéd’élaboration par PVD a été qualifié afin d’obtenir un film aux propriétés AMF magnétique / Nowadays, the miniaturization development is a key parameter in order to fabricate increasinglycomplex microsystems. Research on smart materials aroused a great interest. In this context, westudy the magnetic shape memory alloy Ni2MnGa as a thin layer. This material can be activatedby mechanical, thermal and magnetic stresses. The coupling of these effects would allow thedevelopment of micro-actuators usually made from complex assemblies. This multidisciplinary studyfocuses on the impact of the process on the functional properties of thin films. The originality of thiswork lies in the use of silicon substrates in the context of a technological implementation. A methoddeveloped by PVD was qualified to obtain a film with magnetic shape memory alloy properties.

In the past, it has been found that CuZr-based BMG composites containing B2 CuZr crystals in the glassy matrix display significant plasticity with obvious work hardening. In this work, it was tried to provide a strategy for pinpointing the formation of CuZr-based BMG composites, to modify the microstructures of these composites, and to clarify their yielding and deformation mechanisms. In order to pinpoint the formation of CuZr-based BMG composites, the phase formation and structural evolution of 11 kinds of CuZr-based alloy systems, altogether 36 different compositions, during heating and quenching processes were investigated. An endothermic event between the crystallization and melting peaks was found to be associated with a eutectoid transformation of the B2 CuZr phase. With the addition of elements to the CuZr-based alloys, this endothermic peak(s) shifts to lower or higher temperatures, implying that minor element additions can change the thermal stability of the B2 CuZr phase. By considering the thermal stability of the supercooled liquid, i.e. its resistance against crystallization, and the thermal stability of the B2 CuZr phase, a new strategy to select compositions, which form metastable CuZr-based composites consisting of an amorphous phase and B2 CuZr crystals, is proposed. It is characterized by a parameter, K = Tf /TL, where Tf and TL are the final temperature of the eutectoid transformation during heating and the liquidus temperature of the alloy, respectively. Based on this criterion, the present CuZr-based alloys are classified into three types. For Type I alloys with lower K values, it is difficult to obtain bulk metallic glass (BMG) composites. For Type III alloys with higher K values, BMG composites with larger dimensions are prone to be fabricated, whereas only moderate-sized BMG composites can be obtained for Type II possessing intermediate K values. Accordingly, CuZr-based BMG composites containing B2 CuZr phase in the glassy matrix for different alloy systems were successfully fabricated into different dimensions. For the sake of controlling the formation of the B2 CuZr phase in the glassy matrix and then changing the deformability of CuZr-based BMG composites, different methods were also used to fabricate these composites by: (1) introducing insoluable/high-melting particles; (2) appropriate re-melting treatments of master alloys; and (3) a new flash heating and quenching method. It was demonstrated that the volume fraction, size and distribution of the B2 phase in the glassy matrix can be controlled as well using the methods above. In order to clarify the excellent mechanical properties of CuZr-based BMG composites, the yielding and plastic deformation mechanisms of CuZr-based BMG composites were investigated based on SEM, XRD, and TEM observations. With the volume fraction of amorphous phase (famor) decreasing from 100 vol.% to 0 vol.%, a single-to-“double”-to-“triple”-double yielding transition was found. For the monolithic CuZr-based BMGs and their composites with the famor ³ 97.5 ± 0.5 vol.%, only one yielding at a strain of ~2% occurs, which is due to the formation of multiple shear bands in the glassy matrix, and the associative actions of the shear banding and the martensitic transformation (MT), respectively. When the famor is less than 97.5 ± 0.5 vol.%, a “yielding” occurs at a low strain of ~1%, which results from the yielding of B2 CuZr phase and the onset of the MT within B2 CuZr phase. When the famor is larger than 55 ± 3 vol.%, a “yielding” observed at strains >8% is ascribed from the operation of dislocations with a high density as well as partial de-twinning. It was also found that with the famor decreasing, the deformation mechanism gradually changes from a shear-banding dominated process, to a process being governed by the MT in the crystalline phase, resulting in different plastic strains. Owing to the importance of the MT and the shear banding to the deformation of CuZr-based BMG composites, the details of the MT and the shear banding process were investigated. On one hand, it was found that the MT temperatures of CuZr-based martensitic alloys have a clear relationship with the respective electronic structure and the lattice parameter of the equiatomic CuZr intermetallics. The MT temperatures of the studied alloys can be evaluated by the average concentration of valence electrons. Additional elements with larger atomic radius can affect the stacking fault energy and the electronic charge density redistribution, resulting in the difference of the electronic structures. On the other hand, the formation and multiplication of shear bands for CuZr-based BMG composites is associated with the storage and dissipation of the partial elastic energy during the plastic deformation. When microstructural inhomogeneities at different length scales are introduced into the glassy matrix, the elastic energy stored in the sample-machine system during the plastic deformation is redistributed, resulting in a transition of shear banding process from a chaotic behavior to a self-organized critical state. All in all, our studies and observations provide an understanding of the formation, deformation, and microstrcutural optimization of CuZr-based BMG composites and give guidance on how to improve the ductility/toughness of BMGs.:Contents Abstract V Kurzfassung IX 1 Theoretical background 1 1.1 Development of metallic glasses 1 1.2 Formation of metallic glasses 3 1.2.1 Thermodynamic considerations 5 1.2.2 Kinetic considerations 7 1.2.3 Structural considerations 10 1.3 Mechanical properties of metallic glasses 14 1.4 Deformation mechanisms of metallic glasses 18 1.4.1 Shear transformation zone theory 18 1.4.2 Free volume model 20 1.4.3 Potential energy landscape theory 21 1.4.4 Cooperative Shearing Model 22 1.5 Strategies to improve the ductility of metallic glasses 24 1.5.1 Nano-scaled microstructural inhomogeneities 25 1.5.2 Micro-scaled microstructural inhomogeneities 28 1.5.3 CuZr-based BMG composites 31 2 Experimental techniques 37 2.1 Sample preparation 37 2.1.1 Arc melting/suction casting 37 2.1.2 Centrifugal casting 38 2.1.3 High-frequency melting/injection casting 39 2.1.4 Melt spinning 39 2.1.5 Ball milling and powder consolidation 40 2.2 Structure characterizations 41 2.2.1 X-ray diffraction 41 2.2.2 Optical microscopy and scanning electron microscopy 41 2.2.3 Transmission electron microscopy 42 2.3 Thermal analysis 43 2.3.1 Differential scanning calorimetry 43 2.3.2 Dilatometry 44 2.4 Measurement of the elastic constants 44 2.5 Compression and tensile tests 44 3 Strategy for pinpointing the formation of CuZr-based BMG composites 46 3.1 Theoretical analysis for the formation of CuZr-based BMG composites 46 3.2 Nature of the eutectoid B2 CuZr transformation 49 3.2.1 Shift of endothermic peak(s) related to the eutectoid B2 transformation 49 3.2.2 Thermal stability of the B2 CuZr phase 52 3.3 Formation of the amorphous phase and the B2 CuZr phase 54 3.4 A new parameter for pinpointing the formation of CuZr-based BMG composites 57 3.5 Conclusions 59 4 Synthesis of CuZr-based BMG composites 60 4.1 Formation of Type I alloys 60 4.2 Formation of Type II alloys 62 4.2.1 Formation and microstructures of the Cu50Zr50 BMG composites 62 4.2.2 Formation and microstructures of the Cu-Zr-Ti BMG composites 67 4.2.3 Formation and microstructures of the Cu-Zr-Al and Cu-Zr-Ag BMG composites 70 4.3 Formation of Type III alloys 74 4.4 Conclusions 76 5 Processing routes for CuZr-based BMG composites 78 5.1 Influence of the melting current/time 78 5.2 Adjusting the cooling rate 81 5.3 Re-melting of the pre-alloy 82 5.4 Introduction of boron nitride particles 84 5.5 Effect of TaW inoculation 87 5.6 “Flash annealing” 93 5.7 Conclusions 100 6 Yielding and deformation mechanisms of CuZr-based BMG composites 101 6.1 Formation and microstructures of Cu47.5Zr47.5Al5 BMG composites 101 6.2 Deformation behavior of Cu47.5Zr47.5Al5 BMG composites 105 6.3 Yielding and plastic deformation mechanisms 110 6.3.1 Yielding and plastic deformation during stage I 110 6.3.2 Yielding and plastic deformation during stage II 113 6.3.3 Yielding and plastic deformation during stage III 114 6.3.4 Plastic deformation during stage IV 118 6.3.5 Fracture behavior 120 6.4 Modeling of the “yielding” behavior 121 6.5 Conclusions 124 7 Martensitic transformation behavior in CuZr-based alloys 126 7.1 Electronic structures and martensitic transformation 126 7.1.1 Electronic structures of the B2 CuZr phase 127 7.1.2 Electronic structures of CuZr martensites 129 7.2 Effect of minor additions on the martensitic transformation 130 7.2.1 Formation of Cu-Zr-Ti crystalline samples 130 7.2.2 Effect of Ti element on the martensitic transformation 133 7.2.3 Effect of minor elements on the martensitic transformation temperature 135 7.3 Martensitic transformation in rapidly solidified alloys 139 7.3.1 Martensitic transformation in the as-cast Cu50Zr50 alloys 140 7.3.2 Martensitic transformation in the as-cast Cu-Zr-Al alloys 142 7.4 Conclusions 145 8 Shear banding process of CuZr-based BMG composites 146 8.1 Serrated flow in CuZr-based BMG composites 146 8.2 Statistical analysis of the serrations for brittle and ductile BMGs 148 8.3 Different statistical results of the serration events for CuZr-based BMG composites during deformation 152 8.4 Energy criteria for serrations in CuZr-based BMG and their composites 155 8.5 Conclusions 158 9 Summary and Outlook 160 Publications 162 Acknowledgements 163 References 164 Schriftliche Erklärung 191 / In letzter Zeit zeigte sich, dass massive Cu-Zr-basierte metallische Glaskomposite, welche B2 CuZr-Kristallite in der amorphen Matrix enthalten, eine ausgeprägte Plastizität mit klarer Kaltverfestigung aufweisen. Im Rahmen dieser Arbeit wurde versucht, eine Strategie zur zielgenauen Einstellung der Phasenbildung und des dazugehörigen Gefüges von massiven CuZr-basierten Glas-Matrix-Kompositen bereitzustellen, sowie deren Fließ- und Verformungsmechanismen aufzuklären. Es wurden elf verschiedene CuZr-basierte Legierungssysteme, insgesamt 36 verschiedene Zusammensetzungen, während Heiz- und Abschreckprozessen untersucht, um die Phasenbildung samt Gefüge von massiven CuZr-basierten Glas-Matrix-Kompositen zielgenau einzustellen. Bei CuZr-basierten metallischen Gläsern kann eine endotherme Reaktion zwischen Kristallisation und Schmelzvorgang der eutektoiden Umwandlung von B2 CuZr zugeordnet werden. Mit Zugabe verschiedener Elemente zur CuZr-Basislegierung kann diese Umwandlung zu höheren bzw. niedrigeren Temperaturen verschoben werden. Bereits geringe Beimischungen beeinflussen die thermische Stabilität der B2 CuZr-Phase. Unter Berücksichtigung der thermischen Stabilität, sowie des Widerstands gegen Kristallisation der unterkühlten Schmelze und der B2 CuZr-Phase wurde eine neue Strategie zur Auswahl des Zusammensetzungsgebiets metastabiler CuZr-Legierungen verschiedener Durchmesser vorgeschlagen. Dieser Widerstand kann durch den Parameter K=Tf/TL beschrieben werden, wobei Tf die Endtemperatur der eutektoiden Umwandlung und TL die Liquidustemperatur sind. Basierend auf diesem Parameter können die untersuchten CuZr-basierten Legierungen in drei Klassen unterteilt werden. Für Legierungen vom Typ I mit niedrigeren K-Werten, ist es schwer massive metallische Glas-Komposite (BMG-Komposite) zu erhalten. Im Gegensatz dazu lassen sich für Legierungen vom Typ III, mit höheren K-Werten, BMG-Komposite mit größeren Probendurchmessern herstellen und Legierungen vom Typ II mit einem mittleren K-Wert mit moderaten Probendurchmessern erzeugt werden. Folglich wurden CuZr-basierte Glas-Matrix-Komposite verschiedener Legierungssysteme mit B2-Phase in der amorphen Matrix erfolgreich in unterschiedlichen Geometrien hergestellt. Zur Kontrolle der Ausbildung der B2-Phase in der amorphen Matrix wurden unterschiedliche Methoden verwendet, um duktile CuZr-basierte BMG-Komposite herzustellen: (1) Einbringen von unlöslichen, hochschmelzenden Partikeln; (2) geeignete Wiederaufschmelzbehandlungen der Vorlegierungen; (3) eine neue Schnellerhitzungs- und -Abschreckmethode. Es konnte gezeigt werden, dass der Volumenanteil, sowie die Größe und Verteilung der B2-Phase in der amorphen Matrix durch die oben genannten Methoden kontrolliert werden können. Um die mechanischen Eigenschaften hinsichtlich des Fließens und der plastischen Deformationsmechanismen von CuZr-basierten BMG-Kompositen aufzuklären, wurden diese näher mittels Rasterelektronenmikroskopie, Röntgenbeugung und Durchstrahlungs-elektronenmikroskopie untersucht. Mit sinkendem Volumenanteil der amorphen Phase (famor) von 100 vol.% auf 0 vol.% kann ein Übergang von einer über zwei zu drei Fließgrenzen beobachtet werden. Für monolithische CuZr-basierte BMGs und ihre Komposite mit einem Anteil famor ≥ 97.5 ± 0.5vol.% erfolgt das Fließen ab einer Stauchung von ~2% durch Ausbildung von mehreren Scherbänden in der amorphen Matrix bzw. dem Zusammenwirken des dazugehörigen Scherens und der Martensitumwandlung. Bei einem Anteil famor unter 97.5 ± 0.5 vol.% findet ein Fließen bei niedrigerer Stauchung von ~1% statt. Dies geschieht aufgrund des Fließens und der beginnenden martensitischen Umwandlungen der B2 CuZr-Phase. Bei einem Anteil famor größer als 55 ± 3 vol.% kann ein Fließen oberhalb einer Stauchung von 8% durch die Interaktion von Versetzungen bei hoher Versetzungsdichte sowie partiellem „Entzwillingen“, beobachtet werden. Es wurde herausgefunden, dass mit sinkendem famor der Verformungsmechanismus schrittweise von einem Scherband dominierten zu einem von der martensitischen Umwandlung dominierten Mechanismus übergeht. Dieser Übergang führt zu Unterschieden in der plastischen Verformung. Da für das Verformungsverhalten von CuZr-basierten BMG-Kompositen die deformationsinduzierte martensitische Umwandlung und die Entstehung sowie Ausbreitung von Scherbändern von herausragender Bedeutung sind, wurden sie näher untersucht. Einerseits wurde herausgefunden, dass die Umwandlungstemperatur der martensitischen Umwandlung von CuZr-basierten martensitischen Legierungen in klarer Beziehung zur entsprechenden Elektronenstruktur und der Gitterkonstanten der äquiatomaren intermetallischen CuZr-Phasen stehen. Die martensitischen Umwandlungstemperaturen der untersuchten Legierungen können über die mittlere Valenzelektronenkonzentration ausgewertet werden. Zusätzliche Elemente mit größerem Atomradius können die Stapelfehlerenergie und die Ladungsdichteverteilung ändern, was in unterschiedliche Elektronenstrukturen mündet. Andererseits ist die Entstehung und Vervielfachung von Scherbändern in CuZr-basierten BMG-Kompositen verbunden mit der Speicherung und Dissipation der partiellen elastischen Energie während der plastischen Verformung. Durch das Einbringen von Gefügeinhomogenitäten unterschiedlicher Größe in die Glasmatrix, wird die elastische Energie, die im System Probe-Maschine gespeichert ist, während der plastischen Deformation umverteilt. Dies führt zu einem Übergang des Schervorgangs von chaotischem Verhalten zu einem selbstorganisierten kritischen Zustand. Insgesamt stellen unsere Untersuchungen und Beobachtungen ein Verständnis der Ausbildung, Verfomung und Gefügeoptimierung von CuZr-basierten BMG-Kompositen bereit und sollen als Leitfaden zur Verbesserung der Duktilität bzw. Zähigkeit von BMGs dienen.:Contents Abstract V Kurzfassung IX 1 Theoretical background 1 1.1 Development of metallic glasses 1 1.2 Formation of metallic glasses 3 1.2.1 Thermodynamic considerations 5 1.2.2 Kinetic considerations 7 1.2.3 Structural considerations 10 1.3 Mechanical properties of metallic glasses 14 1.4 Deformation mechanisms of metallic glasses 18 1.4.1 Shear transformation zone theory 18 1.4.2 Free volume model 20 1.4.3 Potential energy landscape theory 21 1.4.4 Cooperative Shearing Model 22 1.5 Strategies to improve the ductility of metallic glasses 24 1.5.1 Nano-scaled microstructural inhomogeneities 25 1.5.2 Micro-scaled microstructural inhomogeneities 28 1.5.3 CuZr-based BMG composites 31 2 Experimental techniques 37 2.1 Sample preparation 37 2.1.1 Arc melting/suction casting 37 2.1.2 Centrifugal casting 38 2.1.3 High-frequency melting/injection casting 39 2.1.4 Melt spinning 39 2.1.5 Ball milling and powder consolidation 40 2.2 Structure characterizations 41 2.2.1 X-ray diffraction 41 2.2.2 Optical microscopy and scanning electron microscopy 41 2.2.3 Transmission electron microscopy 42 2.3 Thermal analysis 43 2.3.1 Differential scanning calorimetry 43 2.3.2 Dilatometry 44 2.4 Measurement of the elastic constants 44 2.5 Compression and tensile tests 44 3 Strategy for pinpointing the formation of CuZr-based BMG composites 46 3.1 Theoretical analysis for the formation of CuZr-based BMG composites 46 3.2 Nature of the eutectoid B2 CuZr transformation 49 3.2.1 Shift of endothermic peak(s) related to the eutectoid B2 transformation 49 3.2.2 Thermal stability of the B2 CuZr phase 52 3.3 Formation of the amorphous phase and the B2 CuZr phase 54 3.4 A new parameter for pinpointing the formation of CuZr-based BMG composites 57 3.5 Conclusions 59 4 Synthesis of CuZr-based BMG composites 60 4.1 Formation of Type I alloys 60 4.2 Formation of Type II alloys 62 4.2.1 Formation and microstructures of the Cu50Zr50 BMG composites 62 4.2.2 Formation and microstructures of the Cu-Zr-Ti BMG composites 67 4.2.3 Formation and microstructures of the Cu-Zr-Al and Cu-Zr-Ag BMG composites 70 4.3 Formation of Type III alloys 74 4.4 Conclusions 76 5 Processing routes for CuZr-based BMG composites 78 5.1 Influence of the melting current/time 78 5.2 Adjusting the cooling rate 81 5.3 Re-melting of the pre-alloy 82 5.4 Introduction of boron nitride particles 84 5.5 Effect of TaW inoculation 87 5.6 “Flash annealing” 93 5.7 Conclusions 100 6 Yielding and deformation mechanisms of CuZr-based BMG composites 101 6.1 Formation and microstructures of Cu47.5Zr47.5Al5 BMG composites 101 6.2 Deformation behavior of Cu47.5Zr47.5Al5 BMG composites 105 6.3 Yielding and plastic deformation mechanisms 110 6.3.1 Yielding and plastic deformation during stage I 110 6.3.2 Yielding and plastic deformation during stage II 113 6.3.3 Yielding and plastic deformation during stage III 114 6.3.4 Plastic deformation during stage IV 118 6.3.5 Fracture behavior 120 6.4 Modeling of the “yielding” behavior 121 6.5 Conclusions 124 7 Martensitic transformation behavior in CuZr-based alloys 126 7.1 Electronic structures and martensitic transformation 126 7.1.1 Electronic structures of the B2 CuZr phase 127 7.1.2 Electronic structures of CuZr martensites 129 7.2 Effect of minor additions on the martensitic transformation 130 7.2.1 Formation of Cu-Zr-Ti crystalline samples 130 7.2.2 Effect of Ti element on the martensitic transformation 133 7.2.3 Effect of minor elements on the martensitic transformation temperature 135 7.3 Martensitic transformation in rapidly solidified alloys 139 7.3.1 Martensitic transformation in the as-cast Cu50Zr50 alloys 140 7.3.2 Martensitic transformation in the as-cast Cu-Zr-Al alloys 142 7.4 Conclusions 145 8 Shear banding process of CuZr-based BMG composites 146 8.1 Serrated flow in CuZr-based BMG composites 146 8.2 Statistical analysis of the serrations for brittle and ductile BMGs 148 8.3 Different statistical results of the serration events for CuZr-based BMG composites during deformation 152 8.4 Energy criteria for serrations in CuZr-based BMG and their composites 155 8.5 Conclusions 158 9 Summary and Outlook 160 Publications 162 Acknowledgements 163 References 164 Schriftliche Erklärung 191

No description available.

No description available.

Die Motivation zur Untersuchung ternärer und quaternären CuZr-Legierungen bestand in der Annahme, dass die Zugabe von Kobalt den Stabilitätsbereich von B2 CuZr bis zur Raumtemperatur erweitert und Aluminium einen signifikanten Effekt auf die Glasbildungsfähigkeit des CuZr-Systems hat. Die vorliegende Dissertation befasst sich mit der Herstellung und Charakterisierung von Cu50-xCoxZr50 (0 ≤ x ≤ 20) und Cu50-xCoxZr45Al5- (x = 2, 5, 10 und 20) Legierungen. Hierbei wurden die Phasenbildung, die thermische Stabilität, die Mikrostruktur, die Martensitbildung und die mechanischen Eigenschaften der Legierungen untersucht. Die Abhängigkeit der Phasenbildung von der Erstarrungsrate und der thermodynamischen Stabilität von Cu-Co-Zr-Legierungen zeigte, dass die Zugabe von Kobalt die Glasbildungsfähigkeit von Cu-Co-Zr-Legierungen absenkt und die stabilen kristallinen Produkte des Systems von Cu10Zr7 + CuZr2 zu (Cu,Co)Zr Phase mit einer B2 Struktur verändert. Die Ergebnisse weisen darauf hin, dass bei den schmelzgesponnene Bänder mit wenigstens 5 Atom-% Co das Glas direkt in B2 (Cu,Co)Zr kristallisiert, während Massivproben mit Co-Gehalten zwischen 0 ≤ x < 5 die monokline (Cu,Co)Zr Phase und Cu10Zr7 sowie CuZr2 beinhalten, wobei für x ≥ 10 die B2 (Cu,Co)Zr Phase bei Raumtemperatur im Gleichgewicht ist. Des Weiteren werden mit steigendem Co-Gehalt die Martensitumwandlungstemperaturen zu niedrigeren Werten verschoben. Die Phasenbildung im ternären System wird im pseudo-binären (Cu,Co)Zr-Phasendiagramm zusammengefasst, welches die Entwicklung neuer Formgedächtnislegierungen sowie metallischer Glas-Komposite bei Zugabe des Glasbildungselementes Aluminium vereinfacht. In den Vierstofflegierungen erhöht Al die Glasübergangs- und Kristallisationstemperaturen und verbessert dadurch die Glasbildungsfähigkeit des Systems. Die röntgenographische Analyse zeigte, dass die Kristallisationsprodukte der schmelzgesponnenen Bänder variieren: von Cu10Zr7 + CuZr2 + AlCu2Zr zu (Cu,Co)Zr + AlCu2Zr, wenn Co ≤ 5 und Co ≥ 10. Die Herstellung von Massivproben mit unterschiedlichen Durchmessern führte zu einem vollständig amorphen Gefüge, einem metallischen Glas-Komposit oder einem vollständig kristallinen Gefüge. Für Co ≤ 5 tritt neben (Cu,Co)Zr und AlCu2Zr ebenfalls Cu10Zr7 auf. Mittels Rasterelektronen (REM)- und Transmissionselektronenmikroskopie (TEM) erfolgte die Analyse des Einflusses von Al- und Co-Zugaben auf die Mikrostruktur von CuZr-Legierungen. Für die Cu-Co-Zr-Al-Legierungen sowie Cu30Co20Zr45Al5 (ø = 4 mm) und Cu45Co5Zr45Al5 (ø = 2 mm) wurden mikrostrukturelle Untersuchungen mittels TEM durchgeführt. Nachfolgend wurde die Heterogenität der Mikrostruktur in der Cu40Co10Zr45Al5 (ø = 2 mm) untersucht. Der Einfluss von Co auf die mechanischen Eigenschaften von rascherstarrten Cu50-xCoxZr50 (x = 2, 5, 10 und 20 Atom-%) Legierungen zeigt, dass das Verformungsverhalten der Stäbe unter Druckbeanspruchung stark von der Mikrostruktur der (Cu,Co)Zr Phase und somit von der Legierungszusammensetzung abhängt. Kobalt beeinflusst die Bruchfestigkeit der Gussproben. Weiterhin zeigen Proben mit martensitischem Gefüge eine Kaltverfestigung. Neben der Kaltverfestigung zeigen die Legierungen mit hohem Co-Gehalt eine verformungsinduzierte Martensitumwandlung und weisen zwei Streckgrenzen auf. Für die Vierstofflegierungen wurde der Einfluss der Kühlrate und der chemischen Zusammensetzung auf die mechanischen Eigenschaften untersucht. Für Cu48Co2Zr45Al5 (ø = 1.5, 2, 3 und 4 mm) und Cu45Co5Zr45Al5 (ø = 3 mm) wurde der Einfluss der Kühlrate und der Heterogenität diskutiert. Die Ergebnisse zeigen, dass die mechanischen Eigenschaften der Cu50-xCoxZr45Al5-Legierungen stark von der Makrostruktur und dem Volumenanteil der amorphen und kristallinen Phase abhängen. Die verformungsinduzierte Martensitumwandlung in Cu40Co10Zr50- und Cu40Co10Zr45Al5-Gussstäben wurde mittels hochenergetischer Röntgenstrahlung durchgeführt. Die In-situ- Druckversuche erfolgten weg- und kraftkontrolliert. Das makroskopische und mikroskopische Spannung-Dehnungs-Verhalten sowie die Phasenumwandlungskinetik wurden dabei betrachtet. Die relativen Veränderungen der vollständig integrierten Intensität der ausgewählten B2- und Martensitreflexe, die auf die Veränderungen der Volumenanteile der entsprechenden Phasen unter Verformung hinweisen, wurden als Phasenumwandlungsvolumen M/M+B2 beschrieben. / The fact that the presence of Co extends the stability range of B2 CuZr to room temperature, together with the significant effect of Al on improving the glass forming ability of the CuZr system was the motivation to investigate the ternary and quaternary CuZr alloys with the aim of synthesizing BMG composites containing B2 (Cu,Co)Zr crystals. This PhD thesis deals with preparation and characterization of Cu50-xCoxZr50 (0 ≤ x ≤ 20) and Cu50-xCoxZr45Al5 (x = 2, 5, 10 and 20) alloys. The phase formation, thermal stability, microstructure, martensitic transformation and mechanical properties of these alloys were investigated. The dependence of phase formation on solidification rate and the thermodynamically stability of Cu-Co-Zr alloys reveals that the addition of Co decreases the glass forming ability (GFA) of the Cu-Co-Zr alloys and changes the stable crystalline products of the system from Cu10Zr7 + CuZr2 to (Cu,Co)Zr phase with a B2 structure. The results indicate that for the melt-spun ribbons with at least 5 % Co, the glass crystallizes directly into B2 (Cu,Co)Zr, while in the case of bulk specimens, compositions with 0 ≤ x < 5 of Co contain the monoclinic (Cu,Co)Zr phase and Cu10Zr7 and CuZr2, whereas, for x ≥ 10, the B2 (Cu,Co)Zr phase is the equilibrium phase at room temperature. Furthermore, increasing the cobalt content decreases the martensitic transformation temperatures to lower temperatures. The phase formation in the ternary system is summarized in a pseudo-binary (Cu,Co)Zr phase diagram, that helps for designing new shape memory alloys, as well as bulk metallic glass composites with the addition of glass former elements. In the quaternary alloys, Al increases the glass transition and crystallization temperatures and hence improves the GFA of the system. The X-ray analysis illustrates that for the melt-spun ribbons, the crystallization products vary from Cu10Zr7 + CuZr2 + AlCu2Zr to (Cu,Co)Zr + AlCu2Zr when Co ≤ 5 and Co ≥ 10, respectively. Depending on the cooling rates, the bulk samples represent a fully amorphous structure or BMG composites or a fully crystalline structure. For Co ≤ 5, beside (Cu,Co)Zr and AlCu2Zr, Cu10Zr7 exists as well. Scanning (SEM) and transmission (TEM) electron microscopy investigations were done to investigate the effect of Al and Co addition to the microstructure of CuZr alloys. In the case of Cu-Co-Zr-Al alloys, Cu30Co20Zr45Al5 (ɸ = 4 mm) and Cu45Co5Zr45Al5 (ɸ = 2 mm) compositions were selected for the microstructure verification using TEM. Later, the heterogeneity of the microstructure in Cu40Co10Zr45Al5 (ɸ = 2 mm) alloy was considered. The effect of Co on the mechanical properties of rapidly solidified Cu50-xCoxZr50 (x = 2, 5, 10 and 20 at.%) alloys depict that the deformation behavior of the rods under compressive loading strongly depends on the microstructure, and as a results, on the alloy composition. Cobalt affects the fracture strength of the as-cast samples; and deformation is accompanied with two yield stresses for high Co-content alloys, which undergo deformation-induced martensitic transformation. Instead samples with a martensitic structure show a work-hardening behavior. For quaternary alloys, the effects of cooling rate and chemical composition on mechanical properties of the alloys were investigated. Cu48Co2Zr45Al5 (ɸ= 1.5, 2, 3 and 4 mm) and Cu45Co5Zr45Al5 (ɸ = 3 mm) compositions were selected to discuss the effect of cooling rate and heterogeneity, respectively. The results depict that the mechanical properties of Cu50-xCoxZr45Al5 alloys strongly depend on the microstructure and the volume fraction of the amorphous and crystalline phases. The deformation-induced martensitic transformation of Cu40Co10Zr50 and Cu40Co10Zr45Al5 as-cast rods, was studied by means of high-energy X-rays. The in situ compression measurements were performed in track control and load control modes. The macroscopic and microscopic stress-strain behavior, as well as the phase transformation kinetics were considered. The relative changes in the fully integrated intensity of the selected B2 and martensite peaks, which indicate the changes in volume fraction of the corresponding phases under deformation, was described as phase transformation volume, M/M+B2.

Cette thèse vise à améliorer les propriétés d'alliage à mémoire de forme et de stockage d'hydrogène du TiNi par des substitutions chimiques sur le sous-réseau de Ni. L'effet de la substitution de Ni par Pd, Cu et Co sur les propriétés structurales, la transformation martensitique et les propriétés d'hydrogénation de TiNi, a été étudiée par des techniques structurales (diffraction des Rayons-X et des neutrons sur poudre), calorimétriques, par réaction solide-gaz et par mesure électrochimique. Des calculs de type DFT ont été effectués pour mettre en évidence les modifications de la structure électronique sur les propriétés d'hydrogénation. Les trois substitutions, TiNi1-zMz (M = Pd, Cu et Co; z ≤ 0,5), conduisent à la formation de composés pseudo-binaires. La substitution par Pd et par Cu augmentent le volume de la maille de TiNi, alors que celle par Co possède un effet inverse. Les températures de transformation martensitique suivent la même tendance que les changements de volume. Ils augmentent fortement pour M = Pd et légèrement pour M = Cu, tandis qu'elles diminuent pour M = Co. Les propriétés d'hydrogénation sont très sensibles à la nature chimique des substitutions. La capacité diminue fortement avec la substitution par Pd, modérément avec le Cu et reste stable pour le M = Co. Contrairement à l'effet attendu par des considérations géométrique, la substitution par Pd et par Cu diminuent la stabilité des hydrures. Les calculs DFT montrent qu'un effet électronique, et non géométrique, régit la stabilité des hydrures pour M = Pd. La substitution par le cobalt induit une formation de plusieurs hydrures par étapes successives, observée par mesure d'isothermes pression-composition en multiple plateaux. Pour les applications, la substitution de Ni par Cu avec une faible teneur (z = 0,2) augmente la capacité de décharge électrochimique de TiNi de 150 à 300 mAh/g en raison de la déstabilisation de l'hydrure. Cela ouvre de nouvelles perspectives pour l'utilisation d'alliages TiNi comme électrodes de batteries Ni-MH. En revanche, la substitution par Pd ou Cu (avec z = 0,5) est très efficace pour réduire la réactivité de TiNi avec l'hydrogène, et est donc intéressante pour des applications à mémoire de forme sous un environnement réducteur / The PhD thesis aims to improve shape memory and hydrogen storage properties of TiNi by chemical substitutions in the Ni sub-lattice. The effect of Pd, Cu and Co substitutions on crystal structure, martensitic transformation and hydrogenation properties of TiNi has been studied by structural (X-ray and neutron powder diffraction), calorimetric, solid-gas and electrochemical means. Ab initio DFT calculations were done to highlight electronic effects on hydrogenation properties. The three substitutions, TiNi1-zMz (M = Pd, Cu and Co; z ≤ 0.5), lead to the formation of pseudobinary compounds. Substitutions by Pd and Cu increase the unit-cell volume of TiNi, whereas the reverse effect occurs for Co. Martensitic transformation temperatures correlate with volume variations. They increase strongly for M = Pd and slightly for M = Cu, whereas M = Co decreases it. Hydrogenation properties are very sensitive to chemical elements substitution. The capacity decreases strongly for M = Pd, moderately for M = Cu and remains stable for M = Co. Contrary to expected effect by geometric model, both Pd and Cu substitutions decrease the stability of hydrides. DFT calculations show that electronic rather than geometric effects govern hydride stability for M = Pd. Co substitution induces step-wise formation of hydrides with a multi-plateau behaviour in pressure-composition isotherms. As concerns applications, low amount of Cu substitution (z =0.2) increases the electrochemical discharge capacity of TiNi from 150 to 300 mAh/g due to hydride destabilization. This opens new perspectives for using TiNi-based alloys in Ni-MH batteries. In contrast, Pd and high amount of Cu substitution (z = 0.5) are effective to decrease TiNi reactivity towards hydrogen, and therefore attractive for shape memory applications under reductive environment