Cette étude porte sur l’analyse des séries jurassiques du Shaqra Group (Toarcian to Kimmeridgian) qui affleurent de manière continue en Arabie centrale le long d’un transect de plus de 1000 km de long. Ces séries se sont accumulées sur une vaste plate-forme épicontinentale peu profonde, en contexte tropical. Ces affleurements permettent ainsi d’observer la partie occidentale des séries renfermant des systèmes pétroliers prolifiques exploités en subsurface en Arabie Saoudite. L’analyse sédimentologique de nombreuses coupes et la réalisation de corrélations stratigraphiques de haute résolution sur un transect de 600 km au sud de Riyad, complété par des corrélations avec les données de forage plus à l’est (entre Riyadh et le Rimthan Arch), permettent de distinguer une organisation séquentielle à différents ordres de fréquence et de reconstituer l’évolution de la plate-forme au sein de ces séquences. La plate-forme jurassique évolue d’une plate-forme horizontale caractérisée par des systèmes mixtes à la transition continental-marin du Toarcien au Callovien moyen (formations Marrat et Dhruma) vers un système de type rampe - bassin intrashelf du Callovien au Kimméridgien inférieur (formations Tuwaiq et Hanifa) pour finir par une plate-forme aggradante carbonatée et silico-clastique (Fm. Jubaila) puis carbonatée et évaporitique en contexte aride (Fm. Arab) au Kimméridgien. Les cycles tectono-stratigraphiques de 2ème ordre du Jurassique inférieur et moyen sont limités à la base et au sommet par des discontinuités régionales. Ils occupent un dépôt-centre stationnaire et décrivent un onlap côtier de grande ampleur avec un maximum transgressif au Callovien moyen (Upper Tuwaiq Mb.). Durant le Jurassique supérieur, les dépôts de rampe carbonatée de la Formation Hanifa passent progressivement vers l’ouest à des dépôts plus profonds de bassin intrashelf relativement riches en matière organique (Khurais - Rimthan Arch). La séquence Jubaila – Arab-D montre des variations d’épaisseur qui indiquent une déformation de grande longueur d’onde de la plate-forme arabe à cette période. Les faciès récifaux du membre Arab D sont interprétés comme représentant le maximum d’inondation de ce cycle qui se termine par le développement de systèmes carbonatés – évaporitiques à la fin du Jurassique. / They serve as westernmost reference for adjacent prolific reservoirs and source-rock bearing intrashelf basins. Several hierarchical sequences (second to fourth order) have been recognized in outcrops sections (600 km long south of Riyadh) and correlated with gamma-ray logs of subsurface wells (550 km long from Riyadh to Rimthan Arch). The Jurassic platform evolved from very-flat continental-to-nearshore mixed carbonate-siliciclastic platform (Marrat-Dhruma; Toarcian to Middle Callovian) to differentiated ramp with deep intrashelf basins (Tuwaiq-Hanifa; Callovian to Early Kimmeridgian) to a lowstand followed by flat aggraded platform (Jubaila-Arab-D; Kimmeridgian). Tectonic related siliciclastic influx took place in arid condition during the Kimmeridgian (Jubaila Fm.). The Jurassic platform ends with the mixed carbonate-evaporite systems of the Arab Fm. A first second-order tectono-eustatic cycle (Marrat to Tuwaiq) is bounded at the base and top by regional unconformities. It has a stationary depocenter, and show long-term coastal onlap and marine transgression that reached its maximum extent during the upper Tuwaiq (Middle Callovian). The Hanifa Fm. consists of four 3rd-order sequences aggraded flat-toped platform (outcrops to Khurais) marked at the base by argillaceous limestone and top by pure high-energy carbonates with localized reef buildups. The Jubaila Arab-D is two 3rd-order sequences begin with low-stand deposits followed by long-term transgression. These formed flat successions with lateral thickness variations controlled by differential subsidence increased in the Arabian Basin. The transgression is marked by storm-influenced inner-platform with sandstone quartz, grainstones and restricted lime-mudstone. The Maximum marine transgression is placed in the Arab-D with reef buildups in the westernmost inner-platform. During highstand, the reefs are gently prograding out into Rimthan Arch leaving behind restricted lagoon and sabkhah/salina anhydrite.

L’arc des Petites Antilles résulte de la lente subduction vers l'Ouest des plaques Nord et Sud-Américaines sous la plaque Caraïbes (2cm/an). A la latitude de l’archipel guadeloupéen et à ~150 km à l’Ouest du front de déformation, le bassin d'avant-arc de Marie-Galante forme un bassin perché, incliné vers la fosse et limité vers l’Est par un haut-fond, l’Eperon Karukéra. À cette latitude, le bassin de Marie-Galante domine le prisme d’accrétion de la Barbade et fait face à la ride de Tiburon qui balaye la zone du Nord au Sud depuis la fin du Miocène supérieur. Le remplissage sédimentaire du Bassin de Marie-Galante montre des déformations actives au moins depuis ~30 millions d’années. L’objectif du travail est de reconstituer l’évolution tectono-sédimentaire de ce bassin pour apporter de nouvelles contraintes sur la compréhension globale de la zone de subduction frontale des Petites Antilles. Ce travail s'appuie sur les données de bathymétrie multifaisceaux et de sismique réflexion multi-traces haute résolution acquises lors des campagnes du programme KaShallow. Cette base de données, complétée de profils sismiques plus basse résolution de campagnes antérieures, permet d’avoir une couverture pseudo 3D et à quatre échelles de résolution de l'ensemble du bassin. Un échantillonnage par ROV et carottage ciblé a fourni 40 prélèvements dans les principales unités sismiques. Les analyses pétrologiques et les datations biostratigraphiques autorisent des reconstitutions paléo-environnementales depuis le Paléogène supérieur jusqu’à Actuel. L’interprétation sismique multi-échelle montre un bassin sédimentaire atteignant ~4,5s temps double (~4500 à 5625 m) sur un substratum magmatique pré-structuré. Ce bassin est composé de 5 grands ensembles sédimentaires (E-1, E1, E2, E3 et E4) subdivisés en 13 unités limitées par 14 surfaces de discontinuités. L’organisation séquentielle des unités sismiques permet de mettre en évidence 10 séquences de dépôts de troisièmes ordres (S-1 à S9). Le calage biostratigraphique de l’ensemble des séquences permet de proposer une évolution tectono-sédimentaire du bassin de l’Éocène à l’Actuel. Ainsi, nous distinguons quatre systèmes de failles normales associées à trois phases d’extensions qui contrôlent l’évolution architecturale et sédimentaire du bassin. 1/ Un système N050±10°E hérité, actif dès le Paléogène supérieur, qui contrôle le basculement général du bassin vers le SSE. Il est responsable de la formation de l'escarpement de Désirade d’environ 4500 m de dénivelé. Cette première extension est interprétée comme résultant de la fragmentation de l'avant-arc en réponse à l'augmentation du rayon de courbure de la zone de subduction. 2/ Un système N130°-N150°E, structurant à l’échelle de l’Éperon Karukéra, qui contrôle la sédimentation dès le Miocène inférieur et marque une première phase d'extension transverse à l’arc. 3/ Un système N160°-N180°E qui segmente le Bassin de Marie-Galante en un sous-bassin à l'Ouest et l'Éperon Karukéra à l'Est. Cette seconde extension, globalement perpendiculaire à la marge, s'accompagne d’une subsidence et d'une inversion de la polarité du bassin en réponse à son basculement vers la fosse qui débute au cours du Miocène moyen et se poursuit actuellement à l'Est du bassin. Cette évolution à long terme de l'avant-arc, concomitante avec le recul de l'arc volcanique vers l’Ouest, est considérée comme résultant d’une érosion basale de la plaque supérieure. 4/ Un système N090±10°E plus tardif est localisé au centre du bassin et qui contrôle le développement de plates-formes carbonatées néritiques sur certaines têtes de blocs, comme par exemple à Marie-Galante. Cette dernière extension, parallèle à l’arc, se manifeste dans le bassin à partir du Pliocène inférieur. Elle se superpose au régime d'extension perpendiculaire à l'avant-arc et est interprétée comme l'accommodation du partitionnement de la déformation en réponse à l’obliquité croissante du front subduction vers le Nord. / The Lesser Antilles result of the slow westward subduction of the North and South American plate under the Caribbean plate (2 cm / year). At the latitude of the Guadeloupe archipelago and ~ 150 km to the west of the deformation front, the fore-arc basin of Marie-Galante forms a perched basin tilted to the pit and limited to the East by a shoal, the Spur Karukéra. At this latitude, Marie-Galante basin dominates the accretionary prism of Barbados and faces wrinkle Tiburon sweeping the area from North to South from the late Miocene. The sedimentary fill Basin Marie-Galante shows active deformation since at least ~ 30 million years. The aim of the work is to reconstruct the tectono-sedimentary evolution of the basin to provide new constraints on the overall understanding of the frontal subduction zone Lesser Antilles. This work relies on multibeam bathymetry data and high-resolution seismic reflection multi-traces acquired during campaigns KaShallow program. This database, supplemented by lower resolution of previous campaigns seismic profiles, provides a pseudo-3D coverage and four scales of resolution of the entire basin. ROV sampling and targeted core provided 40 samples in the main seismic units. Petrological analysis and biostratigraphic dating allow paleoenvironmental reconstructions from the upper Paleogene up Actuel. Seismic interpretation multiscale shows a sedimentary basin reaching ~ 4,5s double (~ 4500-5625 m) on a substrate pre-structured magma. This basin consists of 5 main sedimentary units (E-1, E1, E2, E3 and E4) divided into 13 units bounded by discontinuities 14 surfaces. The sequential organization of seismic units allows to highlight sequences 10 deposits of third order (S-1 to S9). The biostratigraphic calibration of all sequences able to offer a tectono-sedimentary evolution of the Eocene basin to Present. Thus, we distinguish four normal fault systems associated with three phases of extensions that control the architectural and sedimentary evolution of the basin. 1 / A system N050 ± 10 ° E inherited assets from the upper Paleogene, which controls the overall pelvic tilt towards the SSE. He is responsible for the formation of the escarpment Désirade about 4500 m elevation. The first extension is interpreted as resulting from the fragmentation of the fore-arc in response to the increase in the radius of curvature of subduction. 2 / A system N130 ° -N150 ° E, structuring across the Spur Karukéra, which controls sediment from the Miocene and marks the first phase of transverse extension arc. 3 / A system N160 ° E ° -N180 which segments Basin Marie-Galante in a sub-basin to the west and the Spur Karukéra in the East. This second extension, generally perpendicular to the margin, is accompanied by subsidence and reversing the polarity of the basin in response to his switch to the pit, beginning during the Middle Miocene and is ongoing in the East the basin. This long-term evolution of the forearc, concurrent with the decline in volcanic arc to the west, is considered as resulting from a basal erosion of the top plate. 4 / A system N090 ± 10 ° later E is located in the center of the basin and controlling the development of neritic carbonate platforms on certain blocks heads, such as Marie-Galante. This latest extension, parallel to the arc occurs in the basin from the lower Pliocene. It is superimposed on the expansion plan perpendicular to the fore-arc and is interpreted as the accommodation of the partitioning of deformation in response to the increasing obliquity front subduction north.

De nouveaux polyuréthanes ont été synthétisés sans isocyanate et en intégrant des synthons biosourcés afin de développer des matériaux innovants et performants pour des applications dans le domaine de l’industrie automobile et du bâtiment. À l’issue d’un état de l’art, divers synthons et différentes stratégies ont été retenus afin d’obtenir de nouvelles architectures (macro)moléculaires et conférer aux matériaux finaux les propriétés physico-chimiques recherchées. Dans un premier temps, la réaction entre un biscyclocarbonate issu d’une molécule modèle, l’acide sébacique, et des diamines dimériques ainsi que plusieurs paramètres réactionnels ont été étudiés. Cette première étude a permis de montrer l’intérêt du ratio entre les monomères ainsi que l’influence de la fonctionnalité des diamines sur les propriétés chimiques, thermiques, rhéologiques et mécaniques des matériaux polyuréthanes sans isocyanate. Cette approche a permis dans un second temps de synthétiser des matériaux pouvant être entièrement biosourcés et de comparer leurs propriétés à celles des polyuréthanes conventionnels. Les performances de certains des matériaux thermodurcissables obtenus ont confirmé l’intérêt de cette stratégie pour les applications visées. Enfin, une troisième approche a été consacrée à la synthèse de polyuréthanes sans isocyanate thermoplastiques via la synthèse de prépolymères. Les différentes études ont montré l’importance de la structure des divers synthons sur l’organisation macromoléculaire, et ainsi sur les propriétés des matériaux finaux. / Novative and performant polyurethane materials without isocyanate were synthesized using biobased molecules for applications in automotive and building industries. After a state of the art, different building blocks were selected and various strategies were established to develop new (macro)molecular architectures. First, reactions between a biscyclocarbonate from a model structure and dimer diamines were studied as well as the impact of various reaction parameters. Secondly, this approach was adapted to fully biobased materials. The properties of these nonisocyanate polyurethanes were compared with those of conventional ones. The influence of the monomer structures on the final materials was also demonstrated. The performance of some thermoset materials have confirmed the importance of the chosen strategies for the desired applications. Finally, a third approach was dedicated to the synthesis of thermoplastic nonisocyanate polyurethanes via a prepolymer synthesis. The various studies have revealed the specific impact of the building block structure on the macromolecular organization, and thus on the properties of the final materials.

Chapter 2: This chapter introduces the five study sites (Ras Al‐Qasabah; Al Wajh; Yanbu; Farasan Banks; and Western Farasan Islands) along with the fieldwork and detailed benthic mapping and bathymetry mapping conducted in the Saudi Arabian Red Sea. In the Western Farasan Islands two candidate mapping technologies were compared. Firstly, the QuickBird multispectral satellite sensor and secondly the CASI‐550 airborne hyperspectral sensor. In processing the CASI imagery, it was necessary to customize processing to correct for an unusual across‐track artifact caused by lens condensation. On the basis of cost, logistical constraints, spectral reliability, and project needs, multispectral imagery was found to be the most appropriate technology for regional‐scale mapping. Over 20,000 sq. km of high quality QuickBird imagery were amassed across the five study sites. This represents approximately half the shallow water (<20 m) environment of the Saudi Arabian Red Sea. The work presented in this chapter provides a blueprint for processing such large image data sets. Maps with a minimum mapping unit (MMU) of 7.5 sq. m, and a thematic resolution of fifteen habitat classes were produced at an overall accuracy (Tau statistic) of 70%. The five study sites were found to differ substantially in terms of the type, quantity and spatial arrangement of habitats present. The study illustrates the power of remote sensing for delivering regional‐scale audits of coral reef environments. Chapter 3: Coral reefs and their associated accumulations of carbonate sediment adopt particularly complex planform geometries atop the coastal shelf of the Saudi Arabian Red Sea. By assembling 95,000 sq. km of remote sensing data into a GIS, this study aims to relate the morphology of these shallow‐water depositional environments to processes that sculpt the coastal zone. A typology is developed that sorts carbonate systems into end‐members on the basis of their morphology and relationship to the coastline. The resulting GIS was interrogated for spatial patterns in the distribution and abundance of the end‐members. While several depositional morphologies are present throughout the length of the Saudi Arabian Red Sea, the occurrence of others is restricted to narrow regions of latitude. Such differences in distribution can be explained in process‐terms by the rift tectonics of the Red Sea basin, spatial variability in the presence of sub‐seafloor evaporites, and the input of siliciclastic detritus onto the coastal shelf via wadis. This chapter provides a foundation for understanding the morphological diversity of shallow‐water carbonate systems in both the modern ocean and rock record. Chapter 4: In this chapter a framework is proposed for spatially estimating a proxy for coral reef resilience using remote sensing. Data spanning over 20,000 sq. km of coral reef habitat were obtained using the commercial QuickBird satellite, and freely available imagery (NASA, Google Earth). Principles of coral reef ecology, field observation, and remote observations, were combined to devise mapped indices. These indices capture important and accessible components of coral reef resilience. Indices are divided between factors known to stress corals, and factors incorporating properties of the reef landscape that resist stress or promote coral growth. The first‐basis for a remote sensed resilience index (RSRI), an estimate of expected reef resilience, is proposed. Developed for the Red Sea, the framework of the analysis is flexible and with minimal adaptation, could be extended to other reef regions. The chapter illustrates how remote sensing can be used to deliver more than simply habitat maps of coral reefs. Chapter 5: In this chapter, a fundamental measure of coral reef health, coral cover, is assessed in relation to two physical parameters, water depth and wave height. Light availability declines rapidly with depth, which influences the photosynthetic productivity of coral. Where waves break, they produce a severe increase in marine turbulence, and generate currents that may extend beyond the surf zone. The study is focused on the Farasan Banks where some 4000 sq. km of reef habitat are spread across 12,000 sq. km of the Saudi Arabian coastal shelf. The size of the system creates logistical challenge for standard field‐based monitoring methodologies, such as SCUBA surveys. Here, rapid video assessments were employed to deliver measures of coral health across eight percentage cover classes at 472 locations. Whilst water depth can be reliably derived from satellite, assessing wave height is problematic since the parameter is both spatially and temporally variable. Using daily, satellite derived meteorology, a spatially explicit wave model was developed spanning the nine year period from 1999 to 2008. For the majority of the video sites in the Farasan Banks, coral cover was found to be <11%. This statistic hides the counter trend, however that there are robust patterns in higher coral abundance that can be characterized by water depth and wave height. In the inshore, wave height had little bearing on coral cover, instead video sites with a high coral cover were found with a greater probability in shallow (<9m, reef environments. In the offshore, wave exposure exercises stronger control on coral cover than in the inshore, such that video sites with a coral cover greater than 50% were exclusively found in areas where significant wave height exceeds 2 m. The water depth at which the highest coral cover occurs is also deeper offshore than inshore. Once quantified, the conservative behavior of coral cover with respect to water depth and hydrodynamic exposure offers relevant insight to the management of coral reef environments at regional extent. Chapter 6: Carbonate sequence stratigraphy is founded on the principle that changes in relative sea level are recorded in the rock record by the accumulation of sediment with relative‐water‐depth dependent attributes. While at the scale of a shelf‐to‐basin transect, facies clearly stratify by water depth, the relationship blurs for depths <40 m, the most vigorous zone of carbonate production, where the intrinsic processes of storm and wave reworking influence the seabed through submarine erosion and sediment redistribution. Remote sensing imagery, field observations, and hydrodynamic models for two reef‐rimmed shore‐attached carbonate platforms in the Red Sea show neither water depth nor energy regime to be reliable indicators of facies type when considered in isolation. Considered simultaneously, however, the predictive power of the two variables rises significantly. The results demonstrate it to be an oversimplification to assume a direct link between palaeo‐water depth and depositional lithofacies diversity, while highlighting the importance of hydrodynamics in directing the accumulation of carbonate sediments in the shallow photic zone. While the size distributions of facies in the two focus areas, Al Wajh and Ras Al‐Qasabah, follow power laws, no direct relationship between the lateral continuity of the facies belts and water depth or wave height are reported. The work is relevant for the interpretation of meter‐scale subtidal carbonate cycles throughout the geologic record.

Many of the habitats occupied by fish are highly dynamic, naturally demonstrating substantial abiotic fluctuations over diurnal, tidal or seasonal cycles. It is also the case that throughout their 545 million year evolutionary history, fish have existed in aquatic environments very different to those of the present day. However, the past several decades have seen unprecedented rates of environmental change, at local and global scales, arising from human activities. The two major themes of the present thesis are: 1) Respiratory responses of fish to changes in environmental oxygen and temperature in the context of exploring intra- and inter-specific trait variation and its ecological implications 2) The effects of environmental factors (oxygen, carbon dioxide, temperature and seawater chemistry) on the intestinal precipitation and excretion of calcium carbonate by marine teleosts. In the first study (chapter two) a comprehensive database of fish critical PO2 (Pcrit) data compiled from the published literature is presented. The systematic review of this literature provided the opportunity to critically examine methodologies for determining Pcrit as well as its usefulness as an indicator of hypoxia tolerance in fish. The second study (chapter three) examines whether inter- and intra-specific variation in thermal and hypoxia tolerance in two reef snapper species (Lutjanus carponotatus and Lutjanus adetii) reflects their distributions across the contrasting biophysical environments of the reef flat and reef slope surrounding Heron Island on the Great Barrier Reef. L. carponotatus was clearly the most thermally and hypoxia tolerant of the two species, demonstrating a ~3.5 °C wider thermal tolerance zone (higher CTmax, lower CTmin) and ~26% lower Pcrit than L. adetii. These results suggest that the contrasting distribution of these species between flat and slope reef zones is reflected in their physiological tolerances. However, there was no evidence of intra-species variation in tolerance between flat and slope caught L. carponotatus individuals, indicating that this species does not form physiologically distinct subpopulations between these reef zones. The third study (chapter four) experimentally quantified the effect of hypercarbia (3000 μatm) and hypoxia (50% air saturation) on gut carbonate production by the European flounder (Platichthys flesus). Both hypercarbia and hypoxia resulted in a significant increase in carbonate excretion rate (1.5-fold and 2.4-fold, respectively) and acted synergistically when combined. In the final study (chapter five), gut carbonate production was measured in the European flounder undergoing conditions simulating the ‘calcite seas’ of the Cretaceous. The results of this study support the hypothesis that ocean conditions prevalent during the Cretaceous period resulted in piscine carbonate production rates substantially higher (~14-fold) than the present day. Ultimately, this thesis directly links the environmental physiology of fish at the individual level to wider scale implications (past, present and future), ranging from local ecological patterns all the way up to global carbon cycles.

Le carbonate de diméthyle (DMC) est utilisé comme intermédiaire en chimie des polymères. Actuellement, la production industrielle du DMC met en jeu un procédé polluant et dangereux. En revanche, la synthèse du DMC à partir de méthanol et de CO2 est un procédé « vert ». Cependant, la thermodynamique de la réaction est très défavorable, se traduisant par une conversion inférieure à 1%. L'objectif de cette thèse est de développer et d'étudier des catalyseurs très actifs qui, couplés à un réacteur membranaire, permettraient d'augmenter le rendement et l'activité. Ce manuscrit couvre plusieurs aspects : i) une étude de criblage de catalyseurs, ii) une étude de type structure-activité sur des séries de catalyseurs, basée sur des caractérisations structurales, texturales et de surface, iii) une étude mécanistique et iv) une étude cinétique. Un protocole de mesure d'activité adapté aux faibles conversions est développé. Le criblage catalytique met en évidence l'activité des solutions solides de cérine-zircone (CZ). Les CZ préparées par pyrolyse de flamme sont un ordre de grandeur plus actives que les CZ préparées par coprécipitation. De façon surprenante, il n'est pas possible a priori de distinguer des catalyseurs très actifs et peu actifs à partir de leurs caractéristiques structurales et texturales. Cependant, on observe que l'activité est corrélée à la densité et à la nature des espèces de surfaces méthoxides et carbonates. Le mécanisme réactionnel de l'état de l'art est affiné grâce à des mesures d'échanges isotopiques et de DRIFT. L'étude cinétique intègre des corrections thermodynamiques de l'équilibre de phases du binaire MeOH-CO2 qui constitue un liquide expansé / Dimethylcarbonate (DMC) is used in polymer synthesis as well as a fuel additive and solvent. The current industrial DMC production is a polluting and hazardous process. On the other hand, the direct carboxylation of methanol with CO2 is a green route to DMC. However, this reaction is highly limited by thermodynamics, limiting the conversion to less than 1%. The integration of a catalyst in a catalytic membrane reactor for water and DMC removal, would shift the equilibrium conversion thereby improving the DMC yield. The aim of this thesis is to develop and study highly active catalysts for DMC synthesis. This manuscript covers: (i) catalyst screening (ii) a structure-activity relationship study (iii) a mechanistic approach and (iv) a kinetic study. A protocol to measure the activity at low conversion has been developed. Catalyst screening evidenced solid solutions of ceria/zirconia (CZ) as the most active and selective. Flame sprayed pyrolysis ceria/zirconia are one order of magnitude more active than coprecipitated CZ. Interestingly, structural and textural features like crystalline and porous structure were similar regardless of the activity. Nevertheless, the activity could be correlated to the nature and the density of the methoxides and carbonates surface species. The mechanism was refined by isotopic exchange and DRIFT experiments. A kinetic study was performed in a batch reactor that integrated the physical equilibria of the gas-expanded reacting mixture

L'objectif de cette thèse est d'étudier les transformations physico-chimiques lors de l'élaboration d'un verre à partir d'un mélange granulaire. Les mélanges vitrifiables industriel formant des verres de vitrage ou de conditionnement sont approchés par le système à trois oxydes SiO2-CaO-Na2O, apportés sous forme de sable, de calcaire et de carbonate de sodium. Cette simplification permet d'étudier le lien entre microstructrure de l'empilement granulaire et transformations pendant le chauffage. La variété des chemins réactionnels entre les grains de matières premières transmet en effet l'hétérogénéité du mélange au verre. Le système ternaire est décomposé en deux sous-systèmes. D'une part, les réactions entre le disilicate de sodium et 4 porteurs de calcium (CaCO3, CaO, CaSiO3, Ca2SiO4) révèlent la formation de couches d'espèces cristallines entre le grain calcique en cours de dissolution et le silicate fondu. D'autre part, la synthèse ex situ de carbonate mixte de calcium et de sodium est réalisée. Ce produit est analysé et caractérisé ; son apparition in situ dans un mélange ternaire est observée, et sa réaction avec la silice est étudiée. Dans un mélange vitrifiable silico-sodocalcique, la tomographie X in situ et post mortem montre l'importance de phénomènes transitoires de mouillage et de dégagement gazeux qui orientent les réactions chimiques. En particulier, un mécanisme local et de courte durée convertit une profondeur limitée des grains de calcaire en carbonate mixte. Enfin, l'homogénéité du verre obtenu à 1300°C est étudiée à l'aide d'une méthode de déconvolution de spectres Raman innovante, pour plusieurs granulométries et espèces porteuses de calcium. / This thesis focuses on the physical and chemical transformations undergone by a granula glass batch during the glass making process. Industrial window or container glass batches are complex mixtures, that we approached by the ternary oxide system SiO2-CaO-Na2O. Industrial raw materials, namely silica sand, limestone and sodium carbonate, are used. This simplification makes it possible to study the coupling between the granular microstructure and the physical and chemical transformations during the heating. Indeed, the multiple possible reaction paths propagate the heterogeneity of the granular batch to the glass. To support our study of the ternary system, two sub-system are investigated. On one hand, reactions between sodium disilicate and 4 calcium-bearing species (CaCO3, CaO, CaSiO3, Ca2SiO4) show the formation of successive layers of crystalline intermediate species between the dissolving grains and the molten silicate. On the other hand, sodium calcium mixed carbonate is synthesized ex situ and characterized. Its formation in a ternary glass batch is observed in situ, and its reactivity with silica is studied. In a ternary soda-lime glass batch, X-ray in situ and ex situ microtomography reveals the significance of wetting and degassing phenomena in orienting the grains towards a reaction path. Conversion of a limited depth of limestone grains into mixed carbonate for a short time is notably witnessed. This mechanism is strongly dependent on the grains surrounding limestone particles. Finally, we use an innovative deconvolution procedure to measure the impact of granulometry and the nature of calcium-bearing raw materials on the final glass homogeneity.

Intrusion-related, carbonate rock–hosted replacement deposits are an important source of global base metal production that includes: Cu-Zn skarn, Zn-Pb-Ag carbonate replacement and Ag-base metal deposits. These deposits are located in multiple geological settings and are commonly associated with low-grade Cu-Mo calc-alkaline porphyry districts. Visible alteration halos to these deposits range from ten to hundreds of metres around high temperature skarn deposits, being small to imperceptible around the distal relatively low temperature members of this clan. Patterns of visible and cryptic alteration are described and constrained in this study particularly around paleo-fluid flow zones at different distances and elevations from the ore centre. This was done in order to identify the large-scale zonation, mechanisms, and effects of fluid infiltration especially into the distal portions of these deposits. The main alteration tracers employed included mineralogy, major and trace element geochemistry, oxygen and carbon isotopes, and the fluorescent signature of calcite veins associated with the fluids conduits. Two areas were selected for this study: the large Cu-Zn Antamina skarn and the Ag-base metal Uchucchacua vein deposits in the Peruvian Central Andes as these deposits represent the end-members of the polymetallic carbonate rock-hosted deposits and hence, provide an excellent opportunity to examine the margins and upper sections of these hydrothermal systems. Geochronological analyses of intrusive dikes were used to establish the magmatic and hydrothermal evolution associated with mineralization, as well as the genetic linkage between proximal and distal portions of the mineralizing system at Antamina. Oxygen, carbon and strontium isotope data of vein minerals is also used to constrain the temperature of the fluids proximal and distal to the magmatic centre. Chemical data of fluorescent and non-fluorescent veins were used to determine the main activators of fluorescence and to constrain its relationship with mineralizing fluids. In summary, results of this research identify a large-scale zonation of visible and cryptic alteration around paleo-fluid flow zones demonstrating the linkage between Ag-base metal veins, replacement and skarn deposits. Anomalous halos determined throughout Antamina and Uchucchacua deposits may contribute to the development of a more systematic exploration methodology for these types of deposits. / Science, Faculty of / Earth, Ocean and Atmospheric Sciences, Department of / Graduate

Deep-water strata of the Neoproterozoic Kaza Group and Isaac Formation (Cariboo Group) in the southern Canadian Cordillera (B.C.) were deposited in a passive-margin basin during the break-up of supercontinent Rodinia. At the Castle Creek and Mount Quanstrom study areas, a remarkably continuous stratigraphic interval throughout these units preserves a record of basin-floor overlain by strata deposited in the lowermost part of the slope. Although similar stratal intervals have been described from ancient and modern deep-marine settings, they still remain poorly understood. Three main stratal units are recognized within the study areas. The lower unit consists of three channel-lobe systems formed in the basin floor to slope transition. Uniquely, siliciclastic-dominated strata here consist of a variety of small- and few large-scale scour elements, indicating transport bypass along the channel-lobe transition zone, in addition to detached or attached depositional lobes composed mostly of distributary channels, fine-grained deposits, and uncommon splays, and a rare slope leveed channel complex. The middle unit is a siliciclastic-dominated succession of stacked, km-scale mass-transport deposits (i.e. debrites and slides), which indicates the more frequent emplacement of increasingly larger mass failures on a prograding slope, and are overlain by fine-grained, splay deposits that are successively overlain by channel, ponded and fine-grained deposits. In contrast, the upper unit is a mixed siliciclastic-carbonate slope succession of the first Isaac carbonate, a regional marker horizon that comprises mostly carbonate-rich and siliciclastic-rich fine-grained strata intercalated with channel and gully complexes that are mostly filled with coarser-grained strata. Abrupt changes in facies trends, stratal stacking patterns and depositional styles throughout these units are largely linked to long-term changes in relative sea level and its control on sediment supply, namely sediment caliber, volume and mineralogy. Notably, in the upper unit, small-scale changes in sediment source and supply are related to shorter sea-level variations superimposed on the long-term eustatic change.

L'étude des milieux hétérogènes carbonatés et karstifiés trouve aujourd'hui de nombreuses applications dans les problématiques environnementales, comme la description fine de la structure d'un réservoir côtier, support d'un aquifère affecté par une intrusion d'eau salée. Au sud de l'ile de Majorque (Baléares, Espagne), la plate-forme carbonatée de Llucmajor représente un site privilégié pour ce type d'étude.La caractérisation des hétérogénéités ne peut être réalisée sans une description préalable des propriétés des réservoirs carbonatés et des facteurs modifiant la structure du réservoir. Un des paramètres clés définissant la structure hétérogène est la porosité. Au regard de la large gamme de tailles d'hétérogénéités allant du micromètre à la centaine de mètre, l'utilisation de méthodes multi-échelles est nécessaire.Ce travail de thèse a porté sur la détermination multi-échelles de la porosité, allant de la microporosité à la porosité karstique, et l'établissement de relations entre les changements d'échelles d'observation de ces propriétés. Dans un premier temps, l'application de la microtomographie à rayons X extrait la microporosité et la connectivité des réseaux microporeux. Puis, la tomographie a été utilisée avec une résolution plus faible afin d'étudier le réseau micro- et macroporeux à l'échelle de la carotte. A l'échelle du puits, les images de paroi de puits permettent de déterminer la porosité allant des macropores centimétriques aux karsts. Enfin, la géophysique de surface a permis d'observer l'extension des karsts et hétérogénéités dans une grande partie du site.Ainsi, la structure poreuse du site étudié a été intégrée dans la géologie régionale de la plate-forme. L'observatoire installé sur le site pour surveiller la propagation de l'eau salée dans l'aquifère peut être replacé dans son contexte géologique et pétrophysique, afin d'affiner l'interprétation de ses mesures de variations de salinité. / The study of heterogeneous and karstified carbonates is important in many environmental issues, like the detailed description of a coastal reservoir structure containing an aquifer affected by saline intrusion. At the south of the island of Mallorca (Spain), the Llucmajor carbonate platform represents a typical site for such researches. Characterization of heterogeneities requires a better knowledge of the carbonate reservoir properties and the various factors affecting the structure. One of the key parameters defining the heterogeneous structure is the porosity. According to the wide range of heterogeneity sizes from micrometer to hundreds of meters, the use of multiscale methods is required.This thesis focused on the determination of multiscale porosities from the microporosity to karstic porosity, and of relationships between observation scale changes of these properties. Firstly, the application of X-ray microtomography allows to extract the microporosity and the microporous network connectivity. Then, X-ray tomography was used with a lower resolution to study the micro- and macro-porous network at the core scale. At the borehole scale, the borehole wall images are used to determinate macrostructures from centimeter pores to karsts. Finally, surface geophysics allows to observe the extension of karst and heterogeneities at the site scale.Finally, the porous structure of the study site is integrated into the regional geology of the platform. A hydrogeophysical observatory was installed on site to monitor the spreading of salt water in the aquifer. The interpretation of its salinity measurements can take into account the geological and petrophysical context.