La géophysique procure une vision d'ensemble du sous-sol sous forme de cartes et de coupes et apporte des informations précieuses sur la géologie et l'organisation du sous-sol. Bureau d'étude spécialise dans les diagnostics de sols et la recherche d'objets enterrés, Geoscope, partenaire de la thèse CIFRE, désire développer sa thématique géophysique appliquée aux sites et sols pollués. L'objectif est de localiser, de limiter et caractériser les zones polluées aux hydrocarbures. L’étude porte sur l’impact des pollutions hydrocarbonées sur la réponse géoélectrique, et plus spécifiquement sur la réponse en polarisation provoquée spectrale. Les mécanismes de polarisation, qui peuvent être modifiés par la présence d’hydrocarbures, interviennent sur des gammes de fréquences caractéristiques. La polarisation provoquée spectrale, qui mesure la réponse d’un milieu dans le domaine fréquentiel, apporte des informations plus précises que les autres méthodes géoélectriques qui fonctionnent dans le domaine temporel. Les effets de la présence d’hydrocarbures sont étudiés à différentes échelles : (i) à l’échelle du laboratoire, sur des milieux synthétiques afin d’appréhender les mécanismes de réponse (ii) à l’échelle de deux sites pollués afin de confronter les modèles développés en, laboratoire aux données réelles, d’apprécier les limites des méthodes et d’adapter les protocoles utilisés. / Geophysics provides an overview of the basement in the form of maps and sections and gives valuable informations on the geology and the basement organization. Geoscope, partner of the project, is an engineering office specialized in soil diagnosis and buried objects searching and wants to develop environmental geophysics. The aim is to locate, limit and characterize hydrocarbon contaminated areas.The study focuses on the hydrocarbon contamination impact on the geoelectrical response, and more precisely on the spectral induced polarization response. Polarization mechanisms, which can be modified by hydrocarbons presence, occur on typical frequency ranges. Spectral induced polarization, which measures the response of a medium in the frequency domain, gives more precises informations than others methods which work in temporal domain.The effects of hydrocarbon presence are studied at different scales: (i) at the laboratory scale on synthetic media to understand response mechanisms (ii) across two contaminated sites to compare models developed in laboratory to real data, to evaluate the limits of the methods and to adapt protocols.

Les méthodes géophysiques non intrusives sont requises pour caractériser la zone vadose, les réservoirs d’hydrocarbures, ou les sites de stockages de CO₂. Nous analysons l’impact d’une injection de CO₂ gazeux sur la conductivité électrique et les propriétés électrocinétiques des roches carbonatées partiellement ou totalement saturées grâce aux méthodes de polarisations provoquée et spontanée. Les données sont analysées au regard au regard de mesures effectuées sur une roche chimiquement neutre, vis-à-vis du CO₂, c’est-à-dire , un grès de Fontainebleau. Nous comparons également nos résultats aux données de la littérature. Le coefficient de couplage électrocinétique est un paramètre clé, dont nous allons étudier la dépendance vis-à-vis de la saturation, sujet hautement débattu de nos jours. En utilisant la méthode de polarisation spontanée, nous étudions les relations entre le coefficient de couplage électrocinétique, la pression capillaire ainsi que la perméabilité relative au sein des roches carbonatées. Un système expérimental a été élaboré pour mesurer simultanément la perméabilité relative, l’indice de resistivité et le coefficient de couplage électrocinétique en écoulement diphasique de type eau-gaz, en fonction de la saturation en azote ou en dioxyde de carbone. Les résultats sont comparés à des modèles théoriques basés sur les approches de Brooks et Corey et van Genuchten, dont les exposants nous permettent d’ajuster les données avec les modèles de pression capillaire avec succès. Les échantillons sont saturés avec des saumures de différentes compositions, dont les ions sont mono- ou divalents et pour lesquelles la force ionique est comprise entre 10⁻⁴ et 10⁰ Mol L⁻¹. La valeur absolue du coefficient de couplage électrocinétique augmente lorsque la force ionique de la solution diminue, ce qui a déjà été observé dans les roches gréseuses. Le potentiel zêta a été calculé en utilisant une version modifiée de l’équation d’Helmholtz-Smoluchowski, qui prend en compte les effets liés à la conductivité de surface. Comme pour le coefficient de couplage, la valeur absolue du potentiel zêta chute lorsque la force ionique augmente. Nous nous intéressons également aux effets liés à une injection de CO₂ et à la dissolution de la calcite sur la valeur de ce potentiel zêta. Enfin, nous utilisons la méthode de polarisation provoquée pour déterminer l’influence de conductivité de l’eau porale sur la conductivité électrique complexe, la chargeabilité normalisée ainsi que le temps de relaxation en milieu carbonaté non saturé. Nous montrons que ces paramètres peuvent être considérés comme des paramètres de polarisation de la double couche électrique lorsque la conductivité de l’eau porale est comprise entre 10⁻³ et 10⁰ S m⁻¹. / Minimally intrusive geophysical methods are required to characterize both the vadose zone of the Earth, hydrocarbon reservoirs and CO₂ sequestration. We investigate the impact of gaseous CO₂ on both electrical conductivity and electrokinetic properties of limestones under saturated and unsaturated conditions, using the spectral induced polarization and the self-potential methods. These data are contrasted with measurements performed on a Fontainebleau sandstone and data from the literature. That said, the dependence of a key parameter, the streaming coupling coefficient, with the saturation remains highly debated. Using the self-potential method, we explore the relationship between the streaming potential coupling coefficient, the capillary pressure curves and the permeability in carbonate rocks characterized by distinct textures. A new core flooding system is used to measure simultaneously both the relative permeability, the resistivity index and the streaming potential coupling coefficient in steady-state two-phase flow conditions as a function of the saturation with CO₂ or N₂. The results are compared with a recently developed theoretical model, which can accommodate either the Brooks and Corey or the van Genuchten models for the capillary pressure curves. Saturation was achieved with monovalent and divalent brines with ionic strength ranging from 1x10⁻³ Mol L⁻¹ to 1x10⁰ Mol L⁻¹. The magnitude of the coupling coefficient increases with decreasing ionic strength similarly to the trend observed for sandstones. The zeta potential has been calculated at full saturation using a modified version of the Helmholtz-Smoluchowski equation that accounts for surface electrical conductivity. Under atmospheric conditions, the magnitude of the zeta potential is decreasing with the increase of the ionic strength. We also investigate the effects of a CO₂ release and the calcite dissolution on the magnitude of the zeta potential. Finally, we use the spectral induced polarization method to determine the pore water conductivity effects on the complex electrical conductivity, the normalized chargeability and the main relaxation time during drainage in a clay free limestone. We also show evidences that these parameters could be considered as polarization parameters of the electrical double layer in the pore water conductivity range 10⁻³ - 10⁰ S m⁻¹.

No description available.

Geophysical methods are essential for exploration and monitoring of subsurface formations, e.g. in carbon dioxide sequestration or enhanced geothermal energy. One of the keys to their successful application is the knowledge of how the measured physical quantities are related to the desired reservoir parameters. The work presented in this thesis shows that the presence of carbon dioxide (CO2) in pore space gives rise to multiple processes all of which contribute to the electrical rock conductivity variation. Basically, three mechanisms take place: (1) CO2 partially replaces the pore water, which is equivalent to a decrease in water saturation. (2) CO2 chemically interacts with the pore water by dissolution and dissociation. These processes change both the chemical composition and the pH of the pore filling fluid. (3) The low-pH environment can give rise to mineral dissolution and/or precipitation processes and changes the properties of the grain-water interface. Investigations on the pore water phase show that the reactive nature of CO2 in all physical states significantly acts on the electrical conductivity of saline pore waters. The physico-chemical interaction appears in different manifestations depending mainly on the pore water composition (salinity, ion types) but also on both temperature and pressure. The complex behaviour includes a low- and a high-salinity regime originating from the conductivity increasing effect of CO2 dissociation, which is opposed by the conductivity decreasing effect of reduced ion activity caused by the enhanced mutual impediment of all solutes. These results are fundamental since the properties of the water phase significantly act on all conduction mechanisms in porous media. In order to predict the variation of pore water conductivity, both a semi-analytical formulation and an empirical relationship for correcting the pore water conductivity, which depends on salinity, pressure and temperature, are derived. The central part of the laboratory experiments covers the spectral complex conductivity of water-bearing sand during exposure to and flow-through by CO2 at pressures up to 30MPa and temperatures up to 80°C. It is shown that the impact of CO2 on the real part of conductivity of a clean quartz sand is dominated by the low- and high-salinity regime of the pore water. The obtained data further show that chemical interaction causes a reduction of interface conductivity, which could be related to the low pH in the acidic environment. This effect is described by a correction term, which is a constant value as a first approximation. When the impact of CO2 is taken into account, a correct reconstruction of fluid saturation from electrical measurements is possible. In addition, changes of the inner surface area, which are related to mineral dissolution or precipitation processes, can be quantified. Both the knowledge gained from the laboratory experiments and a new workflow for the description and incorporation of geological geometry models enable realistic finite element simulations. Those were conducted for three different electromagnetic methods applied in the geological scenario of a fictitious carbon dioxide sequestration site. The results show that electromagnetic methods can play an important role in monitoring CO2 sequestration. Compared to other geophysical methods, electromagnetic techniques are generally very sensitive to pore fluids. The proper configuration of sources and receivers for a suitable electromagnetic method that generates the appropriate current systems is essential. Its reactive nature causes CO2 to interact with a water-bearing porous rock in a much more complex manner than non-reactive gases. Without knowledge of the specific interactions between CO2 and rock, a determination of saturation and, consequently, a successful monitoring are possible only to a limited extend. The presented work provides fundamental laboratory investigations for the understanding of the electrical properties of rocks when the reactive gas CO2 enters the rock-water system. All laboratory results are put in the context of potential monitoring applications. The transfer from petrophysical investigations to the planning of an operational monitoring design by means of close-to-reality 3D FE simulations is accomplished.

Geophysical methods are essential for exploration and monitoring of subsurface formations, e.g. in carbon dioxide sequestration or enhanced geothermal energy. One of the keys to their successful application is the knowledge of how the measured physical quantities are related to the desired reservoir parameters. The work presented in this thesis shows that the presence of carbon dioxide (CO2) in pore space gives rise to multiple processes all of which contribute to the electrical rock conductivity variation. Basically, three mechanisms take place: (1) CO2 partially replaces the pore water, which is equivalent to a decrease in water saturation. (2) CO2 chemically interacts with the pore water by dissolution and dissociation. These processes change both the chemical composition and the pH of the pore filling fluid. (3) The low-pH environment can give rise to mineral dissolution and/or precipitation processes and changes the properties of the grain-water interface. Investigations on the pore water phase show that the reactive nature of CO2 in all physical states significantly acts on the electrical conductivity of saline pore waters. The physico-chemical interaction appears in different manifestations depending mainly on the pore water composition (salinity, ion types) but also on both temperature and pressure. The complex behaviour includes a low- and a high-salinity regime originating from the conductivity increasing effect of CO2 dissociation, which is opposed by the conductivity decreasing effect of reduced ion activity caused by the enhanced mutual impediment of all solutes. These results are fundamental since the properties of the water phase significantly act on all conduction mechanisms in porous media. In order to predict the variation of pore water conductivity, both a semi-analytical formulation and an empirical relationship for correcting the pore water conductivity, which depends on salinity, pressure and temperature, are derived. The central part of the laboratory experiments covers the spectral complex conductivity of water-bearing sand during exposure to and flow-through by CO2 at pressures up to 30MPa and temperatures up to 80°C. It is shown that the impact of CO2 on the real part of conductivity of a clean quartz sand is dominated by the low- and high-salinity regime of the pore water. The obtained data further show that chemical interaction causes a reduction of interface conductivity, which could be related to the low pH in the acidic environment. This effect is described by a correction term, which is a constant value as a first approximation. When the impact of CO2 is taken into account, a correct reconstruction of fluid saturation from electrical measurements is possible. In addition, changes of the inner surface area, which are related to mineral dissolution or precipitation processes, can be quantified. Both the knowledge gained from the laboratory experiments and a new workflow for the description and incorporation of geological geometry models enable realistic finite element simulations. Those were conducted for three different electromagnetic methods applied in the geological scenario of a fictitious carbon dioxide sequestration site. The results show that electromagnetic methods can play an important role in monitoring CO2 sequestration. Compared to other geophysical methods, electromagnetic techniques are generally very sensitive to pore fluids. The proper configuration of sources and receivers for a suitable electromagnetic method that generates the appropriate current systems is essential. Its reactive nature causes CO2 to interact with a water-bearing porous rock in a much more complex manner than non-reactive gases. Without knowledge of the specific interactions between CO2 and rock, a determination of saturation and, consequently, a successful monitoring are possible only to a limited extend. The presented work provides fundamental laboratory investigations for the understanding of the electrical properties of rocks when the reactive gas CO2 enters the rock-water system. All laboratory results are put in the context of potential monitoring applications. The transfer from petrophysical investigations to the planning of an operational monitoring design by means of close-to-reality 3D FE simulations is accomplished.