The Intergovernmental Panel on Climate Change (2001) reports that increased anthropogenic greenhouse gas (GHG) emissions, of which carbon dioxide (CO2) is the main component, have caused the Earths temperature to rise. Therefore, it is necessary to find ways to reduce GHG emissions and to deal with the emissions that continue to be produced. Carbon capture and storage (CCS) is one method that is being considered to deal with GHG emissions, specifically CO2 emissions. The basic idea behind CCS is that CO2 is captured from a point source, such as a power plant, and is then transported to a storage site (e.g., an oil or gas reservoir), where it is subsequently stored. The International Energy Agency Greenhouse Gas Programme (IEA GHG) began a CO2 geological sequestration pilot project in 2000 in Weyburn, Saskatchewan as part of an enhanced oil recovery project operatedby Cenovus (formerly EnCana) in the Weyburn Field (White et al. 2004). The research presented in this thesis evaluates the sealing potential of the Lea Park Formation in the Weyburn Field by determining its permeability and CO2 breakthrough pressure. In this context, breakthrough pressure describes the differential pressure between a wetting phase (e.g., formation brine) and a non-wetting phase (e.g., CO2) that is sufficient to enable the non-wetting phase to form a connected flow system across a given volume of porous medium (e.g., a rock sample). A new system for measuring the permeability and CO2 breakthrough pressure of shales was developed in this research. The development effort included extensive trouble-shooting and, ultimately, the development of sample preparation and testing procedures. The new system was used to conduct permeability and CO2 breakthrough pressure tests on shale samples from the Lea Park Formation (i.e., Lea Park shale) and the Colorado Group (i.e., Colorado shale). Permeability results for samples from the Lea Park shale ranged from 14 to 35 nd (1410-21 to 3510-21 m2), and between eight and 46 nd (810-21 to 4610-21 m2) for the Colorado shale. A CO2 breakthrough pressure for the Lea Park shale was determined to be 0.02 MPa, while values of 0.02 and 2.7 MPa were measured for the Colorado shale. The CO2 breakthrough pressure test results indicate that the Lea Park shale will not withstand large pressures before allowing CO2 to flow through it. However, the permeabilities are extremely low; hence the rate of flow would be low. In other words, the low permeability of the Lea Park shale will be the controlling factor in terms of the rate of potential CO2 leakage through it. Calculations based on the properties measured in this research suggest that the time required for CO2 to flow from the base to the top of the Lea Park Formation would be on the order of ten thousand years. Based on diffusion coefficients published for other shales, calculations suggest that CO2 leakage via chemical diffusion would be several times slower leakage via hydraulically-driven flow.

The Intergovernmental Panel on Climate Change (2001) reports that increased anthropogenic greenhouse gas (GHG) emissions, of which carbon dioxide (CO2) is the main component, have caused the Earths temperature to rise. Therefore, it is necessary to find ways to reduce GHG emissions and to deal with the emissions that continue to be produced. Carbon capture and storage (CCS) is one method that is being considered to deal with GHG emissions, specifically CO2 emissions. The basic idea behind CCS is that CO2 is captured from a point source, such as a power plant, and is then transported to a storage site (e.g., an oil or gas reservoir), where it is subsequently stored. The International Energy Agency Greenhouse Gas Programme (IEA GHG) began a CO2 geological sequestration pilot project in 2000 in Weyburn, Saskatchewan as part of an enhanced oil recovery project operatedby Cenovus (formerly EnCana) in the Weyburn Field (White et al. 2004). The research presented in this thesis evaluates the sealing potential of the Lea Park Formation in the Weyburn Field by determining its permeability and CO2 breakthrough pressure. In this context, breakthrough pressure describes the differential pressure between a wetting phase (e.g., formation brine) and a non-wetting phase (e.g., CO2) that is sufficient to enable the non-wetting phase to form a connected flow system across a given volume of porous medium (e.g., a rock sample). A new system for measuring the permeability and CO2 breakthrough pressure of shales was developed in this research. The development effort included extensive trouble-shooting and, ultimately, the development of sample preparation and testing procedures. The new system was used to conduct permeability and CO2 breakthrough pressure tests on shale samples from the Lea Park Formation (i.e., Lea Park shale) and the Colorado Group (i.e., Colorado shale). Permeability results for samples from the Lea Park shale ranged from 14 to 35 nd (1410-21 to 3510-21 m2), and between eight and 46 nd (810-21 to 4610-21 m2) for the Colorado shale. A CO2 breakthrough pressure for the Lea Park shale was determined to be 0.02 MPa, while values of 0.02 and 2.7 MPa were measured for the Colorado shale. The CO2 breakthrough pressure test results indicate that the Lea Park shale will not withstand large pressures before allowing CO2 to flow through it. However, the permeabilities are extremely low; hence the rate of flow would be low. In other words, the low permeability of the Lea Park shale will be the controlling factor in terms of the rate of potential CO2 leakage through it. Calculations based on the properties measured in this research suggest that the time required for CO2 to flow from the base to the top of the Lea Park Formation would be on the order of ten thousand years. Based on diffusion coefficients published for other shales, calculations suggest that CO2 leakage via chemical diffusion would be several times slower leakage via hydraulically-driven flow.

Dans le contexte du stockage géologique des déchets radioactifs, ce travail contribue à la caractérisation de l’effet de l’endommagement diffus sur les propriétés de rétention d’eau et transfert de gaz (perméabilité et percée de gaz). Les matériaux considérés sont les bétons CEM I et CEM V sélectionnés par l’Andra, l’argilite du Callovo-Oxfordien (roche hôte) et les interfaces argilite/béton. Cette étude a fourni des informations sur la microstructure des bétons à partir de leurs propriétés de rétention d’eau mais également à partir de la porosimétrie au mercure. Chaque béton a une microstructure bien distincte, caractérisée par une proportion non négligeable de pores capillaires pour le CEM I et une grande proportion de pores des hydrates pour le CEM V. Plusieurs protocoles d’endommagement ont été développés. L’endommagement contribue à réduire la capacité de rétention d’eau du béton CEM I et à augmenter leur perméabilité au gaz. En revanche, tous les échantillons endommagés présentent une pression de percée au gaz significativement plus faible que celles des matériaux sains, et ceci quel que soit le type de béton. Pour l’argilite, on observe une prise d’eau progressive à HR=100%, qui engendre un endommagement du matériau. Ce dernier réduit sa capacité de rétention d’eau. Par ailleurs, ses propriétés de rétention d’eau et de transport de gaz dépendent fortement de son état hydrique initial ainsi que de son endommagement. Enfin, on observe un phénomène de colmatage au niveau des interfaces, d’abord mécanique, puis hydraulique (et surement chimique) suite à l’injection d’eau. Ceci a pour conséquence de réduire la pression de percée des échantillons d’interface / In the context of geological disposal of radioactive waste, this work contributes to the characterization of the effect of diffuse damage on the water retention and gas transfer properties of concrete (CEM I and CEM V) selected by Andra, Callovo-Oxfordian argillite (host rock) and argillite / concrete interfaces. This study provides information on the concrete microstructure from Mercury porosimetry intrusion and water retention curves: each concrete has a distinct microstructure, CEM I concrete is characterized by a significant proportion of capillary pores while CEM V concrete has a large proportion of C-S-H pores. Several protocols have been developed in order to damage concrete. The damage reduces water retention capacity of CEM I concrete and increases its gas permeability. Indeed, gas breakthrough pressure decreases significantly for damaged concrete, and this regardless of the type of concrete. For argillite, the sample mass increases gradually at RH = 100%, which creates and increases damage in the material. This reduces its ability to retain water. Otherwise, water retention and gas transport properties of argillite are highly dependent of its initial water saturation, which is linked to its damage. Finally, we observed a clogging phenomenon at the argillite/concrete interfaces, which is first mechanical and then hydraulic (and probably chemical) after water injection. This reduces the gas breakthrough pressure interfaces

En France, le système de stockage profond de déchets radioactifs envisagé est constitué d’une barrière naturelle (roche hôte argileuse, argilite) et de barrières artificielles, comprenant des bouchons d’argile gonflante (bentonite)-sable pour son scellement. L'objectif de cette thèse est d’étudier l’efficacité du gonflement et du scellement des bouchons placés dans l’argilite, sous l’effet, à la fois, d’une pression d’eau et d’une pression de gaz (tel que formé dans le tunnel). Pour évaluer la capacité de scellement du bouchon bentonite/sable partiellement saturé en eau, on a évalué sa perméabilité au gaz Kgaz sous pression de confinement variable (jusqu’à 12MPa). L'étanchéité au gaz (Kgaz < 10-20m2) est obtenue sous confinement Pc≥9MPa si la saturation est d’au moins 86-91%. Par ailleurs, nous avons évalué le gonflement et l'étanchéité du bouchon de bentonite-sable imbibé d’eau dans un tube d’argilite ou de Plexiglas-aluminium lisse ou rugueux. La présence de gaz diminue la pression effective de gonflement (et la pression de percée de gaz) à partir d’une pression Pgaz= 4 MPa. Après saturation complète en eau, l’écoulement continu de gaz au travers du bouchon seul se fait à Pgaz=7-8MPa s’il dispose d’une interface lisse avec un autre matériau (tube métallique), alors que celui au travers de l’ensemble bouchon/argilite a lieu à Pgaz=7-7,5MPa. Le passage à travers le bouchon gonflé au contact d’une interface rugueuse se fait à une pression de gaz bien supérieure à la pression de gonflement du bouchon. Les essais de percée de gaz montrent que l'interface et l'argilite sont deux voies possibles de migration de gaz lorsque l’ensemble bouchon/roche hôte est complètement saturé / In France, the deep underground nuclear waste repository consists of a natural barrier (in an argillaceous rock named argillite), associated to artificial barriers, including plugs of swelling clay (bentonite)-sand for tunnel sealing purposes. The main objective of this thesis is to assess the sealing efficiency of the bentonite-sand plug in contact with argillite, in presence of both water and gas pressures. To assess the sealing ability of partially water-saturated bentonite/sand plugs, their gas permeability is measured under varying confining pressure (up to 12MPa). It is observed that tightness to gas is achieved under confinement greater than 9MPafor saturation levels of at least 86-91%. We than assess the sealing efficiency of the bentonite-sand plug placed in a tube of argillite or of Plexiglas-aluminium (with a smooth or a rough interface). The presence of pressurized gas affects the effective swelling pressure at values Pgas from 4MPa. Continuous gas breakthrough of fully water-saturated bentonite-sand plugs is obtained for gas pressures on the order of full swelling pressure (7-8MPa), whenever the plug is applied along a smooth interface. Whenever a rough interface is used in contact with the bentonite-sand plug, a gas pressure significantly greater than its swelling pressure is needed for gas to pass continuously. Gas breakthrough tests show that the interface between plug/argillite or the argillite itself are two preferential pathways for gas migration, when the assembly is fully saturated