京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第19394号 / 理博第4125号 / 新制||理||1593(附属図書館) / 32419 / 京都大学大学院理学研究科地球惑星科学専攻 / (主査)教授 平島 崇男, 教授 土`山 明, 准教授 河上 哲生 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

Analyses of fluid inclusions and microstructures within the Papoose Flat pluton were used to investigate the chemistry and temperatures of fluids circulating with the pluton during cooling. Based on previous microstructural analyses, the interior of this late Cretaceous granitic to granodioritic pluton has been divided into three domains: i) a central core characterized by magmatic microstructures, ii) a middle domain of high temperature (>500°C) solid-state deformation, and iii) an outermost domain characterized by relatively low temperature (<5000°C) solid-state deformation. According to previously published anisotropy of magnetic susceptibility analyses and pluton cooling models, plastic flow occurred in both the outer part of the pluton and within its aureole rocks while the core of the pluton was still molten. Solid-state deformation is proposed to have stopped when the pluton interior cooled through its solidus less than 100,000 years after magma emplacement. Microstructural analysis of samples from all three domains confirmed the transition from magmatic flow in the core of the pluton to solid-state deformation at the pluton margin. However, weakly developed solid-state microstructures overprint the dominant magmatic microstructures in samples from the core domain. The existence of solid-state microstructures in all three domains indicates that deformation continued during and after crystallization of the interior of the pluton. Two phase, low salinity (< 26 wt% NaCl equivalent), liquid-rich aqueous fluid inclusions predominate within both quartz and feldspar grains in all samples. Throughout the pluton, the majority of fluid inclusions are hosted by deformed grains. Feldspar-hosted primary inclusions are associated with sericitic alteration. Inclusions were also observed in feldspar as secondary or pseudosecondary inclusions along fractures. Inclusions in quartz are frequently found near lobate grain boundaries or near triple junctions; linear pseudosecondary inclusion assemblages are commonly truncated against lobate boundaries between adjacent quartz grains, indicating that discrete microcracking events occurred during plastic deformation. Homogenization temperatures overlap for all three microstructural domains. Coexisting andalusite and cordierite in the contact aureole, and the intersection of the Mus + Qtz dehydration reaction with the granite solidus, indicate trapping pressures between 3.8 and 4.2 kb. Ninety-eight percent of the calculated fluid inclusion trapping temperatures at 3.8 - 4.2 kb are below the granite solidus of 650°C. Seventy-six percent of the trapping temperature data fall within the more restricted range of 350-500°C; i.e. at temperatures which are lower than the commonly cited brittle-ductile transition temperatures for feldspar at natural strain rates, but above those for quartz. No correlation could be established between trapping temperatures and either host mineral or microstructural domain within the pluton. The similar, relatively low trapping temperatures indicate that the majority of inclusions preserved in all three domains were trapped during the late low strain magnitude stages of solid-state deformation. The most common fluid inclusion trapping temperatures (400-500°C) in all three microstructural domains are similar to the deformation temperatures indicated by microstructures and crystal fabrics in the outer part of the pluton; these trapping temperatures are obviously lower than temperatures associated with contemporaneous solid state and magmatic flow in the pluton interior. The similar trapping temperatures within the pluton core and margin must indicate that the inclusion-trapping event migrated from the margin to the core of the pluton as it cooled, because fluid inclusions would rapidly equilibrate to a density appropriate for the PT conditions of their host minerals. / Master of Science

No description available.

Ces vingt dernières années, le développement de réseaux haute résolution sismologiques et géodésiques denses a permis la découverte de nouveaux signaux géophysiques parmi lesquels on trouve les trémors non-volcaniques (Non-volcanic tremor, NVT, Obara 2002) et les glissements lents épisodiques (Slow Slip Event, SSE, Dragert et al., 2001). La combinaison de NVT et de SSE est communément observée le long des frontières de plaques, entre la zone sismogénique bloquée à faible profondeur et la zone en fluage ductile à plus grande profondeur (Dragert et al., 2004). Cette association définie des glissements et trémors épisodiques (Episodic Tremor and Slip, ETS), systématiquement associés à des surpressions de fluides et à des conditions proches de la rupture. Dans cette thèse, nous proposons de combiner une étude microstructurale de roches exhumées avec une approche par modélisation numérique afin de reproduire et de mieux comprendre la mécanique des glissements et trémors épisodiques.Nous nous sommes concentrés sur des roches continentales provenant de la Zone de Cisaillement Est du Tende (Corse, France), correspondant à une zone de cisaillement Alpine kilométrique ayant enregistré une déformation dans la zone de subduction (10 kb / 400-450°C, Gueydan et al., 2003). Ces conditions pression-température sont cohérentes avec la localisation des ETS dans les zones de subduction. Les analyses microstructurales et EBSD de ces roches mettent en évidence des localisations de la déformation le long de zones de cisaillement centimétriques contrôlées par une rhéologie dépendante de la taille des grains. La microfracturation de la phase dure (ici du feldspath) et le colmatage de ces microfractures correspondent, respectivement, à de processus de réduction et d’augmentation de la taille des grains.La plupart des récentes modélisations des ETS sont basées sur une loi frictionnelle dite rate-and-state, associant les SSE et les NVT à un cisaillement sur un plan. Contrairement à ces modèles, nous souhaitons modéliser l’ensemble de la roche (et non pas uniquement un plan) avec une rhéologie ductile dépendante de la taille des grains directement guidée par nos observations microstructurales (avec microfracturation et colmatage), Nous faisons l’hypothèse que les SSE peuvent résulter d’une localisation ductile de la déformation et non d’un glissement sur des fractures. Durant la localisation de la déformation, le pompage des fluides peut déclencher une fracturation de la roche par surpression de fluide, ce qui pourrait être la signature des NVT. Le modèle numérique 1D présenté ici nous permettra de valider ces hypothèses. En suivant la loi de Darcy, notre approche nous permet également de prédire les variations de la pression de pore en fonction des variations de la porosité/perméabilité et du pompage des fluides.Les résultats numériques montrent que l’évolution dynamique des microstructures, dépendante des fluides, définie des cycles de localisation ductile de la déformation liés aux augmentations de la pression de fluide. Notre modèle démontre que la disponibilité des fluides et l’efficacité du pompage des fluides contrôlent l’occurrence des ETS. Nous prédisons également les conditions pression-température nécessaires au déclenchement des ETS : 400-500°C et 30-50 km de profondeur en subduction, et ~500°C et 15-30 km de profondeur le long des zones de décrochement. Ces conditions PT sont cohérentes avec les exemples naturels.Aussi simple soit-elle, notre modèle mécanique s’appuyant sur des observations de terrain décrit correctement la relation entre surpressions de fluides, rhéologie dépendant de la taille des grains et le déclenchement des ETS. Des travaux restent à entreprendre comme par exemple la comparaison directe de nos résultats avec des données géophysiques (GPS) ou bien l’introduction d’un nouvelle assemble minéralogique, comme par exemple des roches mafiques pour prendre en compte des minéralogies océaniques. / These last twenty years, the development of dense and highly sensitive seismologic and geodetic networks permits the discovery of new geophysical signals named non-volcanic tremor (Obara 2002) and slow slip events (Dragert et al., 2001). The combination of non-volcanic tremor and transient slow slip is commonly observed at plate interface, between locked/seismogenic zone at low depths and stable/ductile creep zone at larger depths (Dragert et al., 2004). This association defines episodic tremor and slip, systematically highlighted by over-pressurized fluids and near failure shear stress conditions. In this thesis we propose to combine a microstructural analysis of exhumed rocks with a modeling approach in order to accurately reproduce and understand the physics of episodic tremor and slip.We focus on continental rocks from the East Tenda Shear Zone (Corsica, France), a kilometer-wide localized Alpine shear zone that record HP/LT deformation (10kb / 400-450°C, Gueydan et al., 2003). Such pressure-temperature conditions are consistent with the location of episodic tremor and slip in subduction zone. Microstructural and EBSD analyses on these rocks describe a pattern of strain localization in centimeter-scale shear zones guiding by a grain size-sensitive creep. Microfracturing of the strong phase (feldspar here) and the sealing of these microfractures act, respectively, as grain size decrease and grain size increase processes.Most of recent modeling approaches of episodic tremor and slip are based on the rate-and-state variable friction law, describing slow slip event and non-volcanic tremor as slow shear slip on a plane. In contrast with such models, we wish to model the entire rock volume, with a ductile grain size-sensitive rheology guided by our microstructural observations (e.g. microfracturing and sealing as grain size variation processes). We hypothesize that slow slip events may result from ductile strain localization and not transient slip on fractures. Fluid pumping during strain localization may trigger whole rock fracturing at near lithostatic conditions that can be the signature of non-volcanic tremor. The 1D numerical model presented here will allow us to validate these assumptions.We also can predict pore fluid pressure variation as a function of changes in porosity/permeability and strain rate-dependent fluid pumping following the Darcy’s flow law. The fluid-enhanced dynamic evolution of microstructure defines cycles of ductile strain localization related to the increase in pore fluid pressure. We show that slow slip events can be ductile processes related to transient strain localization, while non-volcanic tremor can correspond to fracturing of the whole rock at peak of pore fluid pressure. Our model shows that the availability of fluids and the efficiency of fluid pumping control the occurrence of episodic tremor and slip. We also well predict the temperature and depth ranges of episodic tremor and slip: 400-500°C and 30-50 km in subduction zones and ~500°C and 15-30 km in strike slip settings, consistent with natural examples.As simplistic as it is, our field-guided mechanical model well describe, at first order, the relation between high pore fluid pressure, grain size-sensitive rheology and episodic tremor and slip. Some efforts remain to be done like a real fit of geophysical data (GPS) or the introduction of the new mineralogical assemblage, such as mafic rocks to reproduce oceanic environment.

No description available.

Ce travail présente une caractérisation expérimentale de la localisation de la déformation et la rupture sur un matériau synthétique analogue de la roche dans différentes conditions de chargement axisymétrique. L’évolution de la localisation de la déformation a été caractérisée par la technique de corrélation d’images bidimensionnelle. Deux gammes d’essais en compression et en extension ont été réalisées à différentes pressions de confinement Pc. Profitant de la technique DIC et de la caractérisation détaillée des propriétés constitutives de GRAM1 par Mas et Chemenda (2014; 2015), nous montrons que l'initiation de la localisation des bandes de déformation est précédée par la déformation élastique uniforme puis la déformation élasto-plastique. La localisation de la déformation est initiée dans le régime de dilatance (la dilatance β est positive) et d’adoucissement (le module d’écrouissage h est négatif). Au cours de l'évolution du chargement, la déformation à l'intérieur de la bande de localisation de la déformation devient compactive (β <0) et est accompagnée par un durcissement du matériau (h> 0). Ceci provoque dans un premier temps l’élargissement de la bande puis l’apparition de nouvelles bandes qui se forment successivement. La formation de ces nouvelles bandes provoque un ralentissement de la déformation dans les bandes préexistantes, ce qui densifie progressivement le réseau. Dans les essais en extension axisymétrique, les fractures se forment perpendiculaires à la contrainte principale mineure σ3 sous un état de contrainte extensif à faible Pc puis compressif quand Pc augmente. Nos résultats présentent une transition continue de la fracture extensive à la fracture cisaillante avec une augmentation de la contrainte compressive. Ceci est en accord avec les résultats obtenus pour les essais en extension réalisées sur des roches naturelles (Ramsey et Chester, 2004 ; Bobich 2005) / This work presents an experimental characterization of the strain localization and the rupture on a granular rock analogue material under different conditions of axisymmetric loading. The evolution of the strain localization was characterized by the two-dimensional image correlation technique. Two series of triaxial compression and extension tests were carried out at different confining pressure Pc. Taking advantage of the DIC technique and detailed characterization of GRAM1’s constitutive properties by Mas and Chemenda (2014, 2015), we show that the initiation of deformation localization bands is preceded by the uniform elastic and then elasto-plastic deformation. The strain localization is initiated in the dilatant regime (positive dilatancy factor β) and strain-softening (negative hardening modulus h). During the band evolution, the deformation within it becomes compactive (β<0) and is accompanied by the material hardening (h>0), which causes the band to widen and new bands to form successively. The formation of new bands causes a slowing down the deformation within the prior bands, which makes the progressively densified band network to continuously evolve. In axisymmetric extension tests, the fractures are formed perpendicular to the minor principal stress σ3 in an extensive stress state at low Pc then compressive when Pc increases. Our results show a continuous transition from the extensive fracture to shear fracture with an increase of compressive stress. This is suitable of the results obtained for extension tests performed on natural rocks (Ramsey and Chester, 2004; Bobich, 2005)

High-entropy alloys (HEAs) represent a special group of solid solutions containing five or more principal elements. The new design strategy has attracted extensive attention from the materials science community. The design and development of HEAs with desired properties have become an important subject in materials science and technology. For understanding the basic properties of HEAs, here we investigate the magnetic properties, Curie temperatures, electronic structures, phase stabilities, and elastic properties of paramagnetic (PM) body-centered cubic (bcc) and face-centered cubic (fcc) AlxCrMnFeCoNi (0 ≤ x ≤ 5, in molar fraction) HEAs using the first-principles exact muffin-tin orbitals (EMTO) method in combination with the coherent potential approximation (CPA) for dealing with the chemical and magnetic disorder. Whenever possible, we compare the theoretical predictions to the available experimental data in order to verify our methodology. In addition, we make use of the previous theoretical investigations carried out on AlxCrFeCoNi HEAs to reveal and understand the role of Mn in the present HEAs. The theoretical lattice constants are found to increase with increasing x, which is in good agreement with the available experimental data. The magnetic transition temperature for the bcc structure strongly decreases with x, whereas that for the fcc structure shows a weak composition dependence. Within their own stability fields, both structures are predicted to be PM at ambient conditions. Upon Al addition, the crystal structure changes from fcc to bcc with a broad two-phase field region, in line with the observations. Bain path calculations suggest that within the duplex region both phases are dynamically stable. Comparison with available experimental data demonstrates that the employed approach describes accurately the elastic moduli of the present HEAs. The elastic parameters exhibit complex composition dependences, although the predicted lattice constants increase monotonously with Al addition. The elastic anisotropy is unusually high for both phases. The brittle/ductile transitions formulated in terms of Cauchy pressure and Pugh ratio become consistent only when the strong elastic anisotropy is accounted for. The negative Cauchy pressure of CrMnFeCoNi is found to be due to the relatively low bulk modulus and C12 elastic constant, which in turn are consistent with the relatively low cohesive energy. Our findings in combination with the experimental data suggest anomalous metallic character for the present HEAs system. The work and results presented in this thesis give a good background to go further and study the plasticity of AlxCrMnFeCoNi type of HEAs as a function of chemistry and temperature. This is a very challenging task and only a very careful pre-study concerning the phase stability, magnetism and elasticity can provide enough information to turn my plan regarding ab initio description of the thermo-plastic deformation mechanisms in AlxCrMnFeCoNi HEAs into a successful research.

L'objectif de cette étude est d'aboutir à une meilleure compréhension de la fragilisation par les métaux liquides (FML) à travers l'étude des systèmes T91/Pb-Bi et 316L/Pb-Bi et notamment d'en établir les mécanismes mis en jeu lors du contact entre ces aciers sous tension et le métal liquide. Ce travail s'inscrit dans le cadre du projet MEGAPIE-TEST mis en place pour étudier la faisabilité d'une cible de spallation au plomb-bismuth liquide. L'effet de l'eutectique plomb-bismuth liquide sur le T91 et le 316L a été étudié en fonction de la température et de la vitesse de déformation, en utilisant des éprouvettes CCT adaptées à l'étude de propagation de fissures. La présence de Pb-Bi modifie le mécanisme de rupture du T91 au détriment de la germination, croissance et coalescence des cavités. La rupture procède alors par décohésion des bandes de cisaillement. L'effet fragilisant du Pb-Bi est très marqué aux très faibles vitesses de déformation. Une transition fragile-ductile se produit aux grandes vitesses de déformation (~10-5 m.s-1 à 160°C). Les propriétés mécaniques du 316L ne sont pas autant affectées par la présence de Pb-Bi, toutefois une transition réelle est observée sur les faciès de rupture, où là également il y a compétition entre l'effet fragilisant du métal liquide et la rupture ductile. Le mécanisme suggéré dans ce travail est fondé sur la localisation de la déformation en pointe de la fissure combinée au phénomène de réduction d'énergie de surface induite par adsorption de métal liquide (effet Rebinder) et ne fait intervenir aucun processus diffusionnel en particulier aux joints de grains. / The aim of this work is to study liquid metal embrittlement (LME) on the T91/Pb-Bi and 316L/Pb-Bi systems. A particular attention is paid to obtain a better understanding of the mechanisms of fracture when steels are in contact with liquid metal. This work has been performed within the European projects MEGAPIE-TEST and EUROTRANS which aim to prove the feasibility of lead-bismuth nuclear systems such as spallation target and subcritical reactors. The effect of liquid Lead Bismuth Eutectic (LBE) on 316L and T91 steels has been studied in plane stress conditions as a function of temperature and strain rate, using a CCT geometry adapted for the study of crack propagation. The presence of LBE modifies the fracture mechanism of T91 and prevents fracture by growth and coalescence of cavities. Cracking proceeds by shear band decohesion. This embrittlement effect is very pronounced at low deformation rate whereas at the high strain rate range investigated, a brittle to ductile transition is observed. The temperature variation of the transition rules out LME mechanisms based on dissolution. A fracture mechanics analysis by the J-µa methodology allowed the quantification of the embrittlement degree which is estimated to 30% reduction in the energy required for crack propagation. The mechanical properties of the 316L steel are weakly affected by the presence of LBE, in spite of a change in the plastic deformation at the highest triaxiality point which strongly affecting fracture surfaces. The mechanism of this embrittlement seems to be based on the deformation localization at the crack tip combined with the phenomenon of surface energy reduction induced by the liquid metal adsorption. It does not involve any diffusion process. The deformation localization is confirmed by an electron microscopy study of the crack tip plasticity of 316L under the influence of a liquid metal.

Made available in DSpace on 2017-07-21T19:25:45Z (GMT). No. of bitstreams: 1 Ivan Mathias.pdf: 11996423 bytes, checksum: f3bdcfad9b494e72052f6a36c4a749d4 (MD5) Previous issue date: 2015-04-01 / Fundação Araucária de Apoio ao Desenvolvimento Científico e Tecnológico do Paraná / In this work two vitreous systems are studied, the lithium disilicate (LS2) and sodiumcalcium-silica with stoichiometry 2Na2O.CaO.3SiO2 (2N1C3S) and the glassceramics formed from these by heat treatment. Several properties were determined for the two systems as a function of crystallized volume fraction, from glass to fully crystallization (100%), highlighting the fracture toughness and the brittle-ductile transition, with the last two determined only for the LS2 glass-ceramic. Hardness and elastic modulus were obtained for the two glass-ceramics and their values increase with the crystallized volume fraction in the glass ceramic, with the exception of hardness of 2N1C3S, which has its maximum for the crystallized volume fraction of 9%. Thermal expansion coefficients were determined and are larger in the LS2 glassy phase and in the 2N1C3S crystalline phase, thereby generating mean residual stresses obtained by Selsing model of -76 MPa for the LS2 (compression in the crystal) and 232 MPa for the 2N1C3S (traction in the crystal). The indentation fracture toughness was also determined for the two systems using the Anstis' and Niihara's models. The results show an increase of indentation fracture toughness with the crystalline volume fraction for LS2 glass-ceramic and also a dependence with indentation load. As for the 2N1C3S glass-ceramic, indentation fracture toughness are reduced at intermediate crystalline fractions, which is attributed to residual stresses arising from the difference between the thermal expansion mismatch between the glass and the crystalline phases. LS2 glass-ceramic flexural strength increases with the crystalline fraction, from 103 ± 3 MPa for the glass to 260 ± 20 MPa for the fully crystallized sample. Without the removal of the crystallization surface layer, this value rises to 290 ± 20 MPa. The increase in flexural strength in the first 20% of the crystallized fraction is more pronounced. As the size of the precipitates was kept constant, this increase can be related only to the increase in the crystallized fraction. The residual stress in the matrix, the critical radius of spontaneous cracking of the crystals and the crack mean free path between the precipitates were considered in the analysis of the increase in flexural strength. The existence of pores in the samples was a factor that limited its resistance. The fracture toughness (KDTIC) a function of the crystallized fraction was determined for LS2 glassceramics using the double torsion technique. It was found that KDTIC increases with the crystallized fraction, from 0.75 MPa.m1/2 for the glass to about 3.50 ± 0.05 MPa.m1/2 for the fully crystallized sample, a significant increase of approximately five times. Several factors were analyzed as possible causes of the increase in KDTIC. The experimental data are better adjusted with a recently proposed model with one adjustable parameter that relates the ratio of the crystal and glass areas to the crystallized volume fraction. The brittle-ductile transition (BDT) of LS2 glass and glass-ceramic samples (39% crystallized volume fraction) were determined for three different strain rates. BDT temperatures were determined for each strain rate.Activation energies of BDT for the glass and glass-ceramic were obtained, which were 5.2 ± 0.2 eV and 7 ± 2 eV. It was found that BDT activation energy in glass resembles the activation energy of the LS2 viscous flow, thus concluding the BDT in LS2 is governed by viscous flow of the glass matrix. Finally, the fact of the activation energy of the glass ceramic be larger than the glass was attributed to the fact that the viscosity of the vitreous matrix is "hindered" by the presence of crystalline precipitates. A viscosity model of a rigid spheres composite was used as an analogy to explain this observation. / No presente trabalho são estudados dois sistemas vítreos, o dissilicato de lítio (LS2) e o soda-cal-sílica de estequiometria 2Na2O.CaO.3SiO2 (2N1C3S), bem como os vitrocerâmicos formados a partir destes através de tratamentos térmicos. Diversas propriedades foram determinadas para os dois sistemas em função da fração cristalizada, desde vidro até os 100%, com destaque para a tenacidade à fratura e a transição frágil-dúctil, sendo estas últimas determinadas somente para o LS2. Dureza e módulo de elasticidade foram obtidos para os dois sistemas e seus valores aumentam com a fração volumétrica cristalizada no vitrocerâmico, com exceção da dureza no 2N1C3S, que tem seu máximo para a fração cristalizada de 9%. Os coeficientes de expansão térmica foram determinados e são maiores na fase vítrea do LS2 e na fase cristalina do 2N1C3S, gerando assim tensões residuais médias obtidas pelo modelo de Selsing de -76 MPa para o LS2 (compressiva no cristal) e 232 MPa para o 2N1C3S (trativa no cristal). A tenacidade à fratura por indentação (KC) foi determinada também para os dois sistemas, sendo utilizados os modelos de Anstis e Niihara. Os resultados mostram um aumento com a fração cristalina para o LS2 e também uma dependência com a carga utilizada no teste. Já para o 2N1C3S, os valores de KC sofrem uma redução em frações cristalinas intermediárias, comportamento atribuído às tensões residuais oriundas da diferença entre os coeficientes de expansão térmica e anisotropias elásticas do material. Os ensaios de resistência à flexão mostraram que para o LS2 a resistência aumenta com a fração cristalina, passando de 103 ± 3 MPa para o vidro para 260 ± 20 MPa para a amostra totalmente cristalizada. Se não removermos a camada de cristalização superficial, este valor sobe para 290 ± 20 MPa. O aumento da resistência à flexão nos primeiros 20% da fração cristalizada é mais pronunciado. Como o tamanho dos precipitados foi mantido constante, esse aumento pode ser relacionado apenas ao aumento na fração cristalizada. A tensão residual na matriz, o raio crítico dos cristais para trincamento espontâneo e o livre caminho médio da trinca entre os precipitados foram considerados na análise do aumento da resistência à flexão. A existência de poros nas amostras foi um fator que limitou a sua resistência. Caso amostras sem poros fossem feitas, um aumento em torno de 20 a 30% da resistência seria obtido. A tenacidade à fratura (KDTIC) foi determinada para o LS2 pela técnica de torção dupla em função da fração cristalizada. Foi verificado que KDTIC aumenta com a fração cristalizada, passando de 0,75 MPa.m1/2 para o vidro para cerca de 3,50 ±0,05 MPa.m1/2 para a amostra totalmente cristalizada, um aumento significativo de aproximadamente cinco vezes. Diversos fatores foram apontados como possíveis causas do aumento da tenacidade e foi verificado que os fatores considerados de forma isolada não são suficientes para descrever completamente o aumento na tenacidade. Os dados experimentais são melhor ajustados com um modelo de um parâmetro de ajuste recentemente proposto que relaciona a razão entre as áreas dos cristais e do vidro na superfície de fratura com a fração cristalizada. A transição frágil-dúctil (TFD) de amostras vítreas e vitrocerâmica (39% fração cristalizada) de LS2 foram determinadas para três taxas de deformação. Foram determinadas as temperaturas de TFD para cada uma das taxas e foi verificada uma dependência com a taxa de deformação. Foram calculadas as energias de ativação para a TFD no vidro e vitrocerâmico, sendo elas de 5,2 ± 0,2 eV e 7 ± 2 eV. Verificou-se que a energia de ativação da TFD no vidro se assemelha a energia de ativação do escoamento viscoso do LS2, concluindo assim que a TFD no LS2 é governada pelo escoamento viscoso da matriz vítrea. Por fim, o fato da energia de ativação do vitrocerâmico ser maior que do vidro foi atribuída ao fato de que a viscosidade da matriz vítrea seria "dificultada" pela presença dos precipitados cristalinos. Um modelo de viscosidade de um compósito com esferas rígidas foi utilizado como analogia para explicar essa observação.

La transition d’un comportement séismique/instable à un comportement aséismique/stable est observée dans la partie en aval des zones sismogéniques (12-15 km de profondeur). Cette transition est supposée être contrôlée par l’activation de la plasticité de basse température du quartz à ~350°C. À cause de la grande profondeur à laquelle cette transition a lieu, le seul moyen pour étudier les processus physiques qui agissent en ces contestés, est l’étude des anciens prismes d’accrétion exhumés actuellement dans des chaines de montagnes. Le mélange tectonique de Hyuga et l’unité de Morotsuka appartiennent au prisme fossile de Shimanto et sont des unités metasédimentaires déformées à des températures peu inférieures ou égales à la limite fragile/ductile (~250 et ~340°C respectivement). Les résultats des observations de microstructures en microscopie optique et en microscopie électronique à balayage (diffraction des électrons rétrodiffusés) confirment que 1) la pression dissolution et une intense microfracturation sont les mécanismes de déformation principaux du quartz dans le mélange de Hyuga et localement l’activation de la plasticité du quartz est aussi observée; 2) dans l’unité de Morotsuka la recristallisation dynamique du quartz est pleinement active. Ces considérations indiquent que la température n’est pas le seul paramètre qui control l’activation de la plasticité du quartz, et laisse supposer la participation de l’effet adoucissant de l’eau. Avec le but de mieux comprendre le rôle de l’eau sur la rhéologie quartz, des expériences en Presse Griggs ont été menées, le matériel du départ étant de porphyroclasts de quartz (immergés dans une matrice sec) à la fois très riches en eau (provenant du mélange tectonique de Hyuga) et secs (quartz du Brésil). Ces expériences montrent l’effet très adoucissant de l’eau, qui à parité de conditions de déformation, favorise la migration de joint des grains dans le quartz de Hyuga tandis que le quartz du Brésil reste indéformé à exceptions de ses bordures extérieures. L’eau « en excès » est expulsée dans la matrice pour le quartz de Hyuga et stockée dans des bandes de cisaillement C’; l’eau incorporée par le quartz de Brésil n’est pas suffisantes pour favoriser la recristallisation dynamique. / The trasition from instable seismic to stable aseismic behaviour is observed in at the lower limit of the seismogenic zones in subduction zones (12-15 km). This transition is supposed to be controlled by the onset of quartz low grade plasticity at about 350°C. Due to inaccessibility of these geodynamic contests, the only way to study the physical processes acting at these depth are exhumed accretionary prisms exposed in mountain chains. The Hyuga tectonic mélange and the Foliated Morotsuka are metasedimentary units constituting the Shimanto accretionary prism (Japan). They were deformed at temperatures of ~250°C and ~340°C respectively, so slightly lower or equal to the temperature transition. Results by optical microscopy and EBSD reveal that 1) quartz deformation mechanisms active in Hyuga Tectonic Mélange are pressure solution and microfracturation accompanied by local quartz low grade plasticity; 2) dynamic recrystallization is totally active in quartz of the Foliated Morotsuka. These considerations allow to consider the role of water in triggering quartz plasticity especially in such water-rich contest as subduction zones. With the aim to better understand the role played by water on quartz rheology, we deformed high hydrated (from Hyuga unit) and dry (classic Brazil) quartz porphyroclasts within a quartz matrix, with the Griggs apparatus. These experiments show the weakening water effect on quartz strength. At the same deformation conditions, the high hydrated Hyuga quartz show recrystallization by grain boundary migration while the dry Brazil porphyroclasts are mostly undeformed, at exception of the outer recrystallized rims. The exceeding water expulsed from Hyuga quartz is stored in C’ shear bands in the matrix; water absorbed by dry Brazil porphyroclasts is not enough to promote dynamic recrystallization.