This work presents the validation and enhancement of existing correlations for estimating and predicting the permeability in low permeability gas sands. The "original" problem of predicting the corrected or "liquid equivalent" permeability has been under investigation since the early 1940s — in particular, using the application of "gas slippage" theory to petrophysics by Klinkenberg. In the first part of this work, the viability of the Jones-Owens and Sampath-Keighin correlations for estimating the Klinkenberg-corrected (absolute) permeability from single-point, steady-state measurements were investigated. We also provide an update to these correlations using modern petrophysical data. In the second part of this work we proposed and validated a new "microflow" model for the evaluation of an equivalent liquid permeability from gas flow measurements. This work was based on a more detailed application of similar concepts employed by Klinkenberg. In fact, we obtained the Klinkenberg result as an approximate form of this result. A theoretical "microflow" result was given as a rational polynomial (i.e., a polynomial divided by a polynomial) in terms of the Knudsen number (ratio of the mean free path of the gas molecules to the characteristic flow length (typically the radius of the capillary)), and this result can be applied as an explicit correlation device, or as an implicit prediction model (presuming the model is tuned to a particular data set). The following contributions are derived from this work: ● Validation and extension of the correlations proposed by Jones-Owens and Sampath-Keighin for low permeability samples. ● Development and validation of a new "microflow" model which correctly represents the flow of gases in low permeability core samples. This model is also applied as a correlation for prediction of the equivalent liquid permeability in much the same fashion as the Klinkenberg model, although the new model is substantially more theoretical (and robust) as compared to the Klinkenberg correction model.

Abnormal pressures have been measured in the deep boreholes at the Bruce site, southern Ontario, where a deep geologic repository for low and intermediate level radioactive waste disposal has been proposed. The pressure regime in the stratigraphic units exhibits either higher than hydrostatic pressure (over-pressured) or lower than hydrostatic pressure (under-pressured) are considered to be abnormal. At the Bruce site, the Ordovician sediments are under-pressured while the underlying Cambrian sandstone and the overlying Guelph carbonate are over-pressured. Hypotheses have been documented in literature to explain the phenomenon of abnormal pressures. These hypotheses include osmosis, glacial loading and deglaciation unloading, exhumation of overlying sediments, crustal flexure and the presence of an immiscible gas phase. Previous work on the Bruce site has shown that the under-pressures in the Ordovician limestone and shales could not be explained by glaciation and deglaciation or by saturated analyses. The presence of a gas phase in the Ordovician formations has been determined to be a reasonable cause of the under-pressure developed in the Ordovician shales and limestones at the Bruce site. Support for the presence of a gas phase includes solution concentrations of methane, concentrations of environmental isotopes related to methane and estimates of water and gas saturations from laboratory core analyses. The primary contribution of this thesis is the sensitivity analyses performed on the hydrogeologic parameters with respect to a one dimensional two-phase flow model. First, a one dimensional two-phase air and water flow model was adopted and reconstructed to simulate the long-term evolution of the groundwater regimes at the DGR site. Then the hydrogeologic parameters which impact the presence of under-pressure in the groundwater are investigated. Data required to quantify the properties of geologic media and groundwater are adopted directly from borehole testing and laboratory testing results. The permeable boundaries of the domain are assumed to be water saturated and pressure specified (using hydrostatic conditions in the Guelph Formation and hydrostatic with 120 m over-pressure condition in the Cambrian and Precambrian). Isothermal conditions were assumed, thus constant water density and viscosity values are estimated for the average total dissolved solids (TDS) concentration of the modelled stratigraphic column. A constant diffusion coefficient (a diffusivity of $0.25\times10^{-8}$ m$^2$/s) of air in water is assumed with a saturation-dependent tortuosity. The air generation rate is assumed to simulate the gas phase generated in the Ordovician formations. The numerical simulation of up to 4 million years provides a means to explore the behaviour of gas phase dissipation due to partitioning into the water phase and diffusive transport in the solute phase. Results confirmed that the presence of a gas phase would result in the under-pressure in water.

Abnormal pressures have been measured in the deep boreholes at the Bruce site, southern Ontario, where a deep geologic repository for low and intermediate level radioactive waste disposal has been proposed. The pressure regime in the stratigraphic units exhibits either higher than hydrostatic pressure (over-pressured) or lower than hydrostatic pressure (under-pressured) are considered to be abnormal. At the Bruce site, the Ordovician sediments are under-pressured while the underlying Cambrian sandstone and the overlying Guelph carbonate are over-pressured. Hypotheses have been documented in literature to explain the phenomenon of abnormal pressures. These hypotheses include osmosis, glacial loading and deglaciation unloading, exhumation of overlying sediments, crustal flexure and the presence of an immiscible gas phase. Previous work on the Bruce site has shown that the under-pressures in the Ordovician limestone and shales could not be explained by glaciation and deglaciation or by saturated analyses. The presence of a gas phase in the Ordovician formations has been determined to be a reasonable cause of the under-pressure developed in the Ordovician shales and limestones at the Bruce site. Support for the presence of a gas phase includes solution concentrations of methane, concentrations of environmental isotopes related to methane and estimates of water and gas saturations from laboratory core analyses. The primary contribution of this thesis is the sensitivity analyses performed on the hydrogeologic parameters with respect to a one dimensional two-phase flow model. First, a one dimensional two-phase air and water flow model was adopted and reconstructed to simulate the long-term evolution of the groundwater regimes at the DGR site. Then the hydrogeologic parameters which impact the presence of under-pressure in the groundwater are investigated. Data required to quantify the properties of geologic media and groundwater are adopted directly from borehole testing and laboratory testing results. The permeable boundaries of the domain are assumed to be water saturated and pressure specified (using hydrostatic conditions in the Guelph Formation and hydrostatic with 120 m over-pressure condition in the Cambrian and Precambrian). Isothermal conditions were assumed, thus constant water density and viscosity values are estimated for the average total dissolved solids (TDS) concentration of the modelled stratigraphic column. A constant diffusion coefficient (a diffusivity of $0.25\times10^{-8}$ m$^2$/s) of air in water is assumed with a saturation-dependent tortuosity. The air generation rate is assumed to simulate the gas phase generated in the Ordovician formations. The numerical simulation of up to 4 million years provides a means to explore the behaviour of gas phase dissipation due to partitioning into the water phase and diffusive transport in the solute phase. Results confirmed that the presence of a gas phase would result in the under-pressure in water.

In the coming decades, the world will require additional supplies of natural gas to meet the demand for energy. Tight gas reservoirs can be defined as reservoirs where the formation permeability is so low (< 0.1 md) that advanced stimulation technologies, such as large volume fracture treatments, are required before a reasonable profit can be made. Hydraulic fracturing is one of the best methods to stimulate a tight gas well. Most fracture treatments result in 3-6 fold increases in the productivity index. However, if one computes the effective fracture length of most wells, we usually find that the effective length is less than the designed propped fracture length. The “propped length” is the distance down the fracture from the wellbore where proppants have been placed at a high enough concentration to “prop open” the fracture. The “effective length” is the portion of the propped fracture that cleans up and allows gas flow from the reservoir into the fracture then down the fracture to the wellbore. Whenever the effective length is much shorter than the designed propped length, several reasons must be evaluated to determine what might have occurred. For example, the difference could be caused by one or more of the following issues: insufficient fracture fluid cleanup, proppant settling, proppant embedment, proppant crushing, or poor reservoir continuity. Although all these causes are possible, we believe that fracture fluid cleanup issues may be the most common reason the industry fails to achieve the designed propped fracture length in most cases. In this research, we have investigated fracture fluid cleanup problems and developed a better understanding of the issues involved which hopefully will lead to ways to improve cleanup. Fracture fluid cleanup is a complex problem, that can be influenced by many parameters such as the fluid system used, treatment design, flowback procedures, production strategy, and reservoir conditions. Residual polymer in the fracture can reduce the effective fracture permeability and porosity, reduce the effective fracture half-length, and limit the well productivity. Our ability to mathematically model the fundamental physical processes governing fluid recovery in hydraulic fractures in the past has been limited. In this research, fracture fluid damage mechanisms have been investigated, and mathematical models and computer codes have been developed to better characterize the cleanup process. The codes have been linked to a 3D, 3-phase simulator to model and quantify the fracture fluid cleanup process and its effect on long-term gas production performances. Then, a comprehensive systematic simulation study has been carried out by varying formation permeability, reservoir pressure, fracture length, fracture conductivity, yield stress, and pressure drawdown. On the basis of simulation results and analyses, new ways to improve fracture fluid cleanup have been provided. This new progress help engineers better understand fracture fluid cleanup, improve fracture treatment design, and increase gas recovery from tight sand reservoirs, which can be extremely important as more tight gas reservoirs are developed around the world.

The concrete is not one hundred percent impermeable since the water that remains inside it causes its corrosion, in the case of reinforced concrete, exposed in an area of marine atmosphere, the sea salt mostly present in large particles of the marine spray, produce the reduction of the alkalinity of the concrete causing a rapid corrosion of the steel. There are buildings built in this marine area that have been designed without durability criteria, in which the use of pozzolanic materials is considered, for example, to fill the pores of the cement matrix and thus guarantee its impermeability. In the present study, the effect of clay brick powder (PLA) as a replacement for cement in concrete manufacturing is addressed, evaluating different characteristics of its components. The results indicate that pozzolanic activity and compressive strength increase, slump, voids content and the coefficient of permeability to water decreases.

Different methods have been proposed for history matching production of shale gas/oil wells which are drilled horizontally and usually hydraulically fractured with multiple stages. These methods are simulation, analytical models, and empirical equations. It has been well known that among the methods listed above, analytical models are more favorable in application to field data for two reasons. First, analytical solutions are faster than simulation, and second, they are more rigorous than empirical equations. Production behavior of horizontally drilled shale gas/oil wells has never been completely matched with the models which are described in this thesis. For shale gas wells, correction due to adsorption is explained with derived equations. The algorithm which is used for history matching and forecasting is explained in detail with a computer program as an implementation of it that is written in Excel's VBA. As an objective of this research, robust method is presented with a computer program which is applied to field data. The method presented in this thesis is applied to analyze the production performance of gas wells from Barnett, Woodford, and Fayetteville shales. It is shown that the method works well to understand reservoir description and predict future performance of shale gas wells. Moreover, synthetic shale oil well also was used to validate application of the method to oil wells. Given the huge unconventional resource potential and increasing energy demand in the world, the method described in this thesis will be the "game changing" technology to understand the reservoir properties and make future predictions in short period of time.

The petroleum engineering literature clearly shows that large proppant volumes and concentrations are required to effectively stimulate low-permeability gas sands. To pump large proppant concentrations, one must use a viscous fluid. However, many operators believe that low-viscosity, low-proppant concentration fracture stimulation treatments known as ??waterfracs?? produce comparable stimulation results in low-permeability gas sands and are preferred because they are less expensive than gelled fracture treatments. This study evaluates fracture stimulation technology in tight gas sands by using case histories found in the petroleum engineering literature and by using a comparison of the performance of wells stimulated with different treatment sizes in the Cotton Valley sands of the East Texas basin. This study shows that large proppant volumes and viscous fluids are necessary to optimally stimulate tight gas sand reservoirs. When large proppant volumes and viscous fluids are not successful in stimulating tight sands, it is typically because the fracture fluids have not been optimal for the reservoir conditions. This study shows that waterfracs do produce comparable results to conventional large treatments in the Cotton Valley sands of the East Texas basin, but we believe it is because the conventional treatments have not been optimized. This is most likely because the fluids used in conventional treatments are not appropriate or have not been used appropriately for Cotton Valley conditions.

Volatile organic compound (VOC) contamination of low permeability geologic deposits due to Dense Non-Aqueous Phase Liquid (DNAPL) penetration through fractures is exceptionally difficult to remediate using in-situ methods as the low permeability of the sediments limits the delivery of reagents proximal to contaminant mass. This thesis examines in detail the extent of organic contaminant treatment away from hydraulically-induced fractures injected with particulate Zero Valent Iron as (1) ZVI and glycol (G-ZVI) and (2) an emulsified ZVI (EZVI) mixture within a contaminated glaciolacustrine clayey deposit. Continuous vertical cores were collected through the treatment zone at 2 and 2.5 years after substrate injections and soil sub-sample spacing was scaled to show the extent of the treatment zone adjacent to the ZVI in the fractures, expecting the treatment would be controlled by diffusion limited transport to the reaction zone. Analytical results show evidence of treatment in both the EZVI and the G-ZVI containing fractures with the presence of degradation by-products and reduced VOC concentrations in the fracture and surrounding clay matrix. / Natural Sciences and Engineering Research Council of Canada, University Consortium for Field-Focused Groundwater Contamination Research

Les réservoirs non-conventionnels présentent un milieu fracturé à multi-échelles, y compris des fractures stimulées et des fractures naturelles, augmentant l'hétérogénéité et la complexité de la simulation de réservoir. Ce travail propose un modèle unique et simple tout en tenant compte des paramètres clés d'un réservoir, tels que l'orientation des fractures, l'anisotropie et la faible perméabilité du réservoir. L'échange matrice-fracture n'est pas correctement modélisé en utilisation les modèles Discrete Fracture Model (DFM) standards en raison de la très faible perméabilité. Dans ce travail nous proposons l'extension de la méthode MINC (Multiple interagissant Continua) aux modèles DFM afin d'améliorer l'échange matrice-fracture. Notre DFM basé sur la méthode MINC, est un modèle triple porosité où les fractures de très grandes conductivités sont explicitement discrétisées et le reste est homogénéisé. Autrement aux modèles standards et afin d'améliorer l'échange de flux entre la matrice et la fracture, une maille matrice est subdivisé selon une fonction de proximité en tenant compte de la distribution des fractures. Notamment, notre approche est particulièrement utile pour les simulations multiphasique avec un changement de phase dans l'échange matrice/fracture, qui ne peut pas être simulé avec une approche standard. Enfin, nous avons appliqué notre approche pour un cas DFN synthétique dans un réservoir de gaz à condensat et un réservoir tight-oil. Un bon accord a été observé en comparant nos résultats à des solutions de référence obtenues avec des maillages très fins. / Unconventional low permeability reservoirs present a multi-scale fractured media, including stimulated fractures and natural fractures of various sizes, increasing the heterogeneity and the complexity of the reservoir simulation. This work proposes a methodology to address this challenge, taking into account reservoir key parameters such as fractures locations, orientation, anisotropy and low permeability matrix in a unique model as simple as possible. Using standard Discrete Fracture Models (DFMs), the matrix-fracture interaction is not properly handled due to the large grid cells and very low matrix permeability. In this work, we extended the MINC (Multiple INteracting Continua) method to the DFM in order to improve the matrix-fracture flow exchange. Our DFM based on a MINC proximity function is computed by taking into account all discrete fractures, within a triple-porosity model where the propped fractures are explicitly discretized and other fractures are homogenized. In order to improve the flow exchange between the matrix and fracture media, the matrix grid cell is subdivided according to the MINC proximity function based on the distance to all discrete fractures, by using randomly sampled points. Our approach is particularly useful for multi-phase flow simulations in matrix-fracture interaction with phase change, which cannot be handled by a standard approach. Finally, we applied our technique to synthetic DFM case in a retrograde gas and a tight-oil reservoirs. A good agreement is observed by comparing our results to a reference solution where very fine grid cells were used.

Les procédés d’élaboration de pièces composites par infusion de résine, malgré leurs nombreux avantages, peinent à s’imposer dans les phases de production industrielle en raison de difficultés pour les maitriser. Ainsi, en partenariat avec ESI Group, un modèle complet pour la simulation de ces procédés est développé à l’ENSM-SE depuis les travaux précurseurs de P. Celle.Nos travaux portent sur la généralisation de ce modèle afin de traiter des cas, ainsi que sur son extension à la simulation des écoulements « post-infusion ». L'approche repose sur un découpage du domaine en trois zones (drainant, préformes imprégnées, préformes sèches) consistant ainsi à coupler un écoulement de Stokes dans le drainant à un écoulement de Darcy dans les préformes. De plus, l'influence mutuelle de la résine sur le comportement des préformes et de la déformation des préformes sur la perméabilité est considérée, à travers la loi de Terzaghi et des lois exprimant la perméabilité en fonction de la fraction de fibres, paramètre accessible uniquement dans une approche 3D mécanique couplée. Enfin, le procédé est découpé en trois phases : compression initiale des préformes sèches, remplissage et « post-infusion ». Les méthodes numériques, développées dans ces travaux, s'appliquent à des cas réels d'infusion souvent mis de côté dans les publications récentes car inaccessibles, impliquant des perméabilités très faibles (~10-15 m²), un drainant fin (~1 mm) et des géométries complexes.Cette approche innovante a été implémentée dans un code de calcul industriel (ProFlotTM), validée analytiquement sur des cas tests et expérimentalement sur des cas industriel dans le cadre du projet européen INFUCOMP. / Composite manufacturing processes by resin infusion, despite their many benefits, struggle to establish themselves in the industrial production phases due to difficulties to control them. So, in partnership with ESI Group, a comprehensive model for the simulation of these processes is developed at the ENSM-SE since the pioneering work of P. Celle.Our work focuses on the generalization of this model to handle complex industrial cases in three dimensions, as well as its extension to “post-infusion” flow simulation. The approach is based on three domains decomposition of the field (Distribution medium, impregnated preforms, dry preforms) consisting in coupling a Stokes flow in the distribution medium with a Darcy flow in the preforms. In addition, the mutual influence of the resin on the preforms and of the preforms deformation on the permeability is considered, through Terzaghi’s law and models expressing the permeability as a function of the fibre fraction, data only accessible with a 3D coupled mechanical approach. Finally, the process is divided into three phases: initial compression of dry preforms, filling and “post-infusion”. The numerical methods developed in this work, apply to real infusion cases often discarded in recent publications, involving very low permeability (~10-15 m²), thin distribution medium (~ 1 mm) and complex geometries (3D curved).This innovative approach has been implemented in an industrial simulation code (ProFlotTM), validated analytically over test cases and experimentally over industrial cases in the European project INFUCOMP.