Granite | EPFL Graph Search

Granite (ˈɡrænɪt ) is a coarse-grained (phaneritic) intrusive igneous rock composed mostly of quartz, alkali feldspar, and plagioclase. It forms from magma with a high content of silica and alkali metal oxides that slowly cools and solidifies underground. It is common in the continental crust of Earth, where it is found in igneous intrusions. These range in size from dikes only a few centimeters across to batholiths exposed over hundreds of square kilometers. Granite is typical of a larger family of granitic rocks, or granitoids, that are composed mostly of coarse-grained quartz and feldspars in varying proportions. These rocks are classified by the relative percentages of quartz, alkali feldspar, and plagioclase (the QAPF classification), with true granite representing granitic rocks rich in quartz and alkali feldspar. Most granitic rocks also contain mica or amphibole minerals, though a few (known as leucogranites) contain almost no dark minerals. Granite is nearly always massive

Transmissivity of rock fractures: Normal loading and shear reactivation

Monia Woehner

Enhanced Geothermal Systems represent a major field of study in the context of renewable energy resources. To create extractable energy from those reservoirs, a high enough fluid flow rate for production needs to be achieved. This fluid flow rate is directly related to the permeability of the fracture system in the reservoir. Hence, by increasing the reservoir’s permeability the fluid flow increases as well. For this purpose, shear stimulation has proven itself effective but remains a challenging field of study. Many aspects of this mechanism are still unknown and have to be explored further. This study focuses on the fracture transmissivity evolution of low porosity rocks (i.e. granite and marble) with smooth fracture surfaces subjected to progressing shear displacement until several millimetres of offset. The aim of this study was to record fracture transmissivities continuously throughout the experiment without stopping the mechanical loading processes and thereby using the oscillatory pore pressure method. In this context, a new experimental procedure was created making a direct shear configuration using a standard triaxial apparatus possible. Both samples were exposed to a constant confining pressure of 25 MPa which can be directly reflected as normal stress on the fracture. The granite fracture transmissivities were in the range of 6.8317·10−20 to 3.4429·10−19 m3. They first followed a decreasing trend until reaching the minimum value. Then a strong peak could be observed after which transmissivities were increasing again until the end of the experiment. Numerous peaks and deviations from the linear trend could also be observed. The marble fracture transmissivities were in the range of 1.4604 · 10−20 to 2.1110 · 10−20 m3 and were generally lower than the granite’s. Also, the transmissivity evolution followed a different path, continiously and slowly decreasing until the final position. The curve was generally more flattened and did not show any strong peaks. These results showed that rock fractures behave in different manners according to their lithology and more precisely, to the material’s hardness. Rock fractures often form wear products which can either obstruct the flow paths or enhance them as a function of the wear mechanism and wear volume. The granite sample encountered the debris formation mechanism that would transition from mild to severe wear. This transition could be observed in its fracture transmissivity passing from a decreasing to an increasing trend. On the other hand, the marble sample was subjected to a continual asperity smoothing mechanism, thereby slowly closing up the fracture and decreasing its fracture transmissivity. Within the frame of EGS, the results of this work can have interesting implications for production rates. Results showed that not only loading conditions can affect the fluid flow through rock fractures. Rock lithologies and ranges of shear displacement also play an important role and should be explored further.

2020

Micromechanical Modelling of Elastic Wave Velocity Variations for Intact Brittle Rocks

Antoine Guggisberg

Until now, the micromechanical wing crack model of Ashby & Sammis 1990 describing the brittle field of rocks has only been used for failure predictions. However, it gives relevant information regarding the crack opening process, so this model is extended to compute elastic wave speeds according to the cracked solid theory of Kachanov. Laboratory experiments at various confining pressures on Westerly granite were realised to fit its parameters and test this extension. A decent correlation between modelled and measured velocities shows that this model might be used for seismic velocities and crack-induced anisotropy computation. Furthermore, the validity of the model is discussed through an energy budget and comparisons with in-situ measurements of velocity changes.

2021

ENHANCED GEOTHERMAL SYSTEMS (EGS) - NUMERICAL PREDICTION OF THE MODE AND LOCATION OF FRACTURE INITIATION

Mohamad Zaarour

Geo-energy is a comprehensive term used to describe any form of energy that comes from the Earth. This includes hydrocarbons such as gas, oil, and coal, but also geothermal energy (shallow and deep). The focus of this thesis is on Enhanced Geothermal Systems (EGS), which are renewable sources of energy. In EGS, hydraulic stimulation is the technique employed to enhance the permeability e.g. of the granite basements at great depths, which might exceed 5 kilometers. Rocks have natural fractures which affect the behavior of the rock mass. The mechanical goal of hydraulic stimulation is to open these pre-existing fractures. Due to the generally great depths, only indirect information such as in-situ stress, pumping records, and micro-seismicity, are available. Many issues in the field of reservoir geothermal (but also petroleum) production, are attributed to the lack of proper understanding of how these fractures interact with each other. The main goal of this study is to provide an insight into how the pre-existing fractures in rocks interact with each other. Specifically, this research answers the question of how pre-existing flaws of different geometrical parameters contained in rock specimens subjected to various loading configurations, affect the mode and the location of the mechanism of crack initiation. We develop a finite element (FE) linearly elastic MATLAB Partial Differential Equation (PDE) code. This numerical model predicts the mode and the location of the initiation of the local fractures at the inner tips of two interacting pre-existing flaws in Barre granite in the context of EGS. We investigate different combinations of double-flaw geometries (flaw inclination angle β, bridging angle α, and ligament length L) contained in specimens subjected to various loading scenarios (Combination of Vertical Load VL, Lateral Load LL, and Internal Flaw Pressure FP) likely to induce first local tensile and shear fractures, and we determine their corresponding predicted locations around the inner flaw tips. For a given double-flaw pair geometry, we subject the specimen to a certain loading configuration (Combination of VL, LL, and FP). We then retrieve the state of stress (maximum and minimum principal stresses 𝜎1 and 𝜎3, and maximum shear stresses 𝜏𝑚𝑎𝑥) at the Finite Element (FE) corner nodes on the boundary of the inner tips, from which we plot the corresponding Mohr’s circles. We study the interaction of these Mohr’s circles with the Griffith-Coulomb failure envelope. Since the main objective is to predict the first local failure’s mechanism and location, we iteratively vary the magnitude of the loading component until the first Mohr’s circle touches the failure envelope. Depending on where it does, the mechanism of the local fracture is inferred. If the maximum principal stress (𝜎1) of this Mohr’s circle is equal to the tensile strength of the granite (𝜎𝑇=7.5𝑀𝑃𝑎), a local tensile fracture is numerically predicted. If this Mohr’s circle touches the linear shear Coulomb portion of the envelope, it is a predicted local shear failure. The corresponding location of the FE node is then determined at the inner tip of the flaw, and this is where for the given geometry and loading scenarios, the first local fracture’s initiation is numerically expected. We follow this procedure for several double-flaw geometrical configurations. We compare the numerical predictions with the previously obtained experimental results in the MIT CEE Rock Mechanics Laboratory. This allows us to validate our set-up numerical model. 4 This study provides a solid ground to the understanding of the initiation and the interaction of pre-existing fractures in EGS. This thesis comes as an answer to a problem of major practical importance in the field of rock mechanics, and beyond. Typical geotechnical problems, such as rock slope stability and tunnel support, can also be better addressed when clearly understanding the process of crack initiation in rocks. This study also enables the correlation between numerical models and experimental findings, a challenge that has always existed in the research world. This thesis shows how our developed FE code can be used as a tool to predict the location and the mode of fractures in the laboratory, but also in the real world.

2020