1,720,983 research outputs found
Strain dependent variation of microstructure and texture in naturally deformed Carrara marble
No full textThis study investigates the microstructure and texture variations across a mm-scale shear zone in Carrara marble of the Alpi Apuane (Italy).
The microstructures have been investigated for grain size, texture, and shape fabrics. Textures have been measured with Computer-Integrated
Polarization Microscopy (CIP) and Electron Back Scattered Diffraction (EBSD) separating porphyroclast and recrystallized grains. The deformation,
which post-dates an earlier deformation phase and subsequent annealing, is strongly localized. The microstructures and textures change
across the shear strain gradient and are interpreted to preserve a time sequence of progressive stages of deformation. The bulk shear strain rate is
estimated to be about 1011 sec1 at deformation temperatures of approximately 325 C 30 C.
The protomylonite is characterized by a core mantle structure with a bimodal grain size distribution which changes gradually to a completely
dynamically recrystallized microstructure with a unimodal grain size distribution in the mylonitic center of the shear zone. Core-mantlestructures
are produced by dominant rotation recrystallization accompanied by some grain boundary migration. The microstructural transition
from protomylonite to mylonite coincides with a change in texture. With increasing strain the single c-axis maximum of an earlier inherited
texture in the protomylonite is replaced by a similar texture in a different orientation (maximum normal to the shear plane) which is consistent
with dominant basal hai and r h2201i slip. The microstructural and textural variations depend on the proportion of recrystallized grains. As
dynamic recrystallization progresses with finite strain the texture development is finite strain-dependent. The comparison of the microstructures
and textures to other natural and to experimental examples explains the progressive change of the texture and demonstrates the texture evolution
produced by dynamic recrystallization
Analysis of End of Field Life Techniques and predicting Liquid Loading using Artificial Neural Network
No full textCivil Engineering and GeosciencesGeoscience & Engineerin
Petrophysical quantification of Utah reservoir sandstones and cap-rocks naturally exposed to CO2 fluid fluxes by use of high-resolution Micro CT
No full textCO2 emission is increasing globally, resulting in a global increase in temperature and as a result climate change. CO2 storage in the subsurface is an option/ transitional measure to reduce CO2 in the atmosphere and mitigate the climate change issue. Generally, for CO2 storage to be a viable option, it must be ensured that the stored gasses should be at least 10.000 years safe in the subsurface. Simulating long-term storage of CO2 in the laboratory for 10.000 years in storage experiments is not feasible and therefore a good option is to analyze rock samples that have naturally been exposed to CO2 for such a large time frame. Such a CO2 aquifer can be found in eolian sandstones in Utah, USA. Due to the occurrence of fractures in these sandstones, CO2 bearing water has flown through these sandstones resulting in possible reaction and carbonate precipitation in the pores. In this study, the main points of interest are the determination of volume fractions of various components like porosity, carbonate precipitation close to and far from the fractures, by the use of Micro Computed Tomography (Micro CT). Micro CT results are compared to conventional lab (macroscopic) technology. The ultimate goals of this research are gaining actual insight/ data on what the result is if CO2 storage is conducted and insight and recommendation regarding the use of Micro CT technology. The first conclusion on results regards calibration. Calibration difficulties in Micro CT technique generate noise. If calculating the fraction of components, noise has no influence on the fraction result. If connectivity analysis is conducted, noise will influence the result and should therefore be filtered. Sensitivity analysis for resolution shows that 2.5 ?m is a suitable resolution to image these eolian sandstones and the smaller the samples are the higher resolution can be acquired. Lab based methods for porosity (Pycnometer, Wet Test and Hg Shell Test) generally agree with each other. The image based methods (2D and 3D) also agree with each other. The lab based methods and image based methods do not exactly agree with each other. The correlations regarding permeability, porosity and particles per volume show exponential relations. After conducting 3D grain analysis, the effect of carbonate precipitation is quantified. Because of the precipitation between grains, the grains tend to ‘deform’, which creates higher surface area and volume. The exact influence of a fracture on carbonate precipitation can be summarized to be minimal. In the case of sample car614 the influence is clearly seen, but in all other samples the fracture seems to have had no influence or overruled by other (depositional) phenomena.Petroleum EngineeringGeoscience & EngineeringCivil Engineering and Geoscience
Seismic attenuation analyses of fracturing in reservoir rocks
No full textIn this study seismic wave velocity and seismic wave attenuation in dry Bentheimer sandstone, Indiana limestone, and Whitby shale are studied while continuously deforming and therefore increasingly fracturing the rock samples up until macroscopic failure. A pulse transmission technique at a central frequency of 1 MHz is used in combination with a high pressure system that is constantly applying uniaxial stress. At early deformation stages before yielding, increasing elastic strains correlate with decreasing seismic attenuation suggesting closure of existing fractures. The deviation from a linear stress-strain relationship indicates the yield point, i.e. the onset of plasticity or fracturing. After yielding, increasing plastic strains correspond to increasing seismic attenuation suggesting formation of new micro-fractures. Thereby, P-wave attenuation is more sensitive to the growth of fractures than S-wave attenuation. The existence of fractures is confirmed by postmortem micro-CT-images. Absolute attenuation values are highest for Indiana limestone and lowest for Whitby shale. Furthermore, S-wave attenuation is the most sensitive to fracturing in Indiana limestone, followed by Bentheimer sandstone; P-wave attenuation is the most sensitive to fracturing in Bentheimer sandstone, followed by Indiana limestone. Attenuation in Whitby shale is the least sensitive to fracturing for both P-waves and S-waves. This study confirms that attenuation is more sensitive to fracturing than the seismic velocity; hence attenuation is suggested to be a valuable seismic parameter to identify fractures. Finally, the study presents the possible usage of amplitude decay-over-time analysis of seismic waves for possible borehole applications and the prediction of induced seismicity.Petroleum Engineering and GeosciencesSection Applied Geophysics and PetrophysicsCivil Engineering and Geoscience
Groningen Gas Field: Heterogeneities characterization of the sediments,by log evaluation.
No full textThis research focuses on the analyze of the heterogeneity of the Groningen gas field, i.e. characterization of the reservoir rock and the sealing Zechstein salt on top of the reservoir, by modeling the rock properties of the formations. The aim of this study is twofold, namely to characterize the thickness, shale content and porosity of the reservoir rock and the sealing Zechstein salt. But also to set an geological framework of the Groningen gas field, which can serve as a basic input for subsidence studies and geomechanical modeling of the Groningen field. The Groningen gas field was first discovered in 1959 by well Slochteren 1 and has an horizontal extent of around 900 km2. The reservoir is situated at a depth of 2600-3200 m and the estimated recoverable volume is about 2700 billion m3. It lies within the Southern Permian Basin gas province, an extensive E-W stretched sedimentary basin. In order to analyze the heterogeneity of the sediments (this include the Ten Boer, Ameland and Slochteren member) and the salt rock, software named Quantumgis is used to obtain two dimensional interpolated plots of the three evaluated rock properties. The three rock properties evaluated are thicknesses, shale content and porosities. Well log data have been used to determine these properties. For the thicknesses, the penetration depth of the well bores are used. For the determination of the shale content, gamma-ray and spontaneous potential logs have been used. And eventually, the porosities are derived from density and neutron logs. The used data is obtained from NLOG, the Dutch gas and oil portal. The Zechstein member is a relative thick formation, and show a relative heterogeneous thickness pattern. The Ten Boer member has a relative homogeneous thickness pattern. It has a high shale content and low porosity. Both properties show a heterogeneous pattern. The Ameland member is relative thin and is only present in the northern half of the field, with a high shale content. The distribution of the shale content show a heterogeneous pattern. Finally, the Slochteren member, which is the main reservoir rock, has a heterogeneous thickness pattern. The porosities are high and the shale content is relative low. Both properties show a relative homogeneous pattern. In short, the evaluated formations show a relative heterogeneous pattern for the three rock properties. If one relates the obtained models in this research with the compaction models done by the NAM (NAM,2013), it is clear that production must be reduced or completely stopped in regions with high porosities, since at these regions subsidence values are the highest. This will lead to less compaction and a lower chance of seismic activities.Applied Geophysics and PetrophysicsGeoscience & EngineeringCivil Engineering and Geoscience
An Experimental Investigation on the Rock Mechanical Behavior of Synthetic Layered Systems and Load-Cycling of its Individual Constituents
No full textMany authors already made an attempt to understand the effect of load-cycling on material strength and the evolution of elastic parameters. However, until now there was no study on the effect of load-cycling on the evolution of elastic parameters over the complete stress strain curve, i.e. the linear elastic regime, the fracturing regime, and the fractured regime. In addition, none of the authors focussed on the effect of load-cycling on fracture network improvement. Although previous studies already showed that elastic moduli of layered systems may be determined from properties and volume fractions of its individual constituents, there is no study done on the relation between rock mechanical properties, i.e. strains and yield or failure stresses, of (synthetic) layered systems and its constituents. Furthermore, in the literature no description is found on fracture characteristics of a (synthetic) layered system. Hence, an experimental investigation is conducted on the rock mechanical behavior of synthetic layered systems subjected to increased-loading and the effect of load-cycling on its individual constituents’ rock strength, elastic parameter evolution, and fracture network improvement. The increased-loading and load-cycling rock mechanical experiments are unconfined compression tests performed at room temperature. When comparing two of the same rock materials with a maximum deviation of 1% in porosity, load-cycling leads to failure at much lower stress levels when compared to increased-loading. Within the linear elastic regime, load-cycling returns a stabilized Young’s modulus which is always larger than its envelope value, while the Poisson’s ratio of the last load cycle coincides with its envelope value. Load-cycling generates an improved fracture network when compared to increased-loading. Characteristics of the improved fracture network are the increased fracture densities and the more uniform distribution of the fractures over the volume of the material. For vertically stacked synthetic layered systems, the elastic moduli and strains can be well-predicted by the Reuss Average which uses the average rock mechanical properties and volume fractions of the individual constituents. In contrast, for synthetic layered systems the stress level at failure point is independent of the volume fractions of its constituents and is observed to be in the vicinity of its weakest constituent. Despite this, fractures are still observed in the strongest constituents. The fracture propagation through the strongest constituent is ascribed to be due to amplified stress concentrations at the tip of the propagating fracture. However, from 2-layered systems it is observed that fracture propagation through the strongest constituent depends on the thickness of the weakest constituent. At the constituents’ boundary in synthetic layered systems, there is no offset in fracture path when going from one constituent to another nor is there a sudden change in aperture. However, there is a change in fracture inclination in a way such that the fracture inclinations of the individual constituents are respected. In addition, fractures in synthetics include cataclastics over the high porous intervals (±10-25%), while the same fracture is clean over the low porous intervals (±0,5-5%).Civil Engineering and GeosciencesGeoscience & EngineeringPetroleum Engineerin
Finite Volume Method for Poroelasticity
No full textWhen modelling fluid fl ow in subsurface, the impact of solid deformation on fluid fl ow is often oversimplified/neglected in reservoir simulators. It is assumed that solid volume/s-tate of stress is a function of fluid pressure, while the opposite effect is not considered. This assumption is made mainly to reduce computational costs and complexity of fl uid flow models. Nevertheless, this simplification is not valid in case of unconsolidated rocks. This oversimplification results in wrong estimation in prediction of surface subsidence, earthquakes and fault activation, fluid production and rock permeability values, etc. Numerous reports suggest that neglecting the two-way coupling (i.e. both fluid-to-solid and solid-to- fluid coupling ) has led to disastrous events in many cases. This necessitates modified fl uid models and simulators which take into account the two-way coupled nature of solid deformation and fl uid flow in porous media. Efforts have been made to model this two-way coupled nature properly which can be categorized as follows. There have been attempts to connect commercial softwares to model this problem. They fail due to differences in data structure, different underlying assumptions embedded and due to the increased computational costs. Therefore, it is essential to integrate fluid modelling and solid deformation into a single simulator. This requires developing new numerical models. Presuming an elastic nature for unconsolidated rocks, Biot's consolidation equations are employed to numerically model the two-way coupled solid deformation and fluid fl ow in porous media, so called poroelasticity. During my master work, I developed 2D MATLAB codes based on two different finite volume discretization schemes (cell-centred and vertex-centred FVM). In the first stage, two cell-centred FVM 2D MATLAB codes were developed: One to model fluid flow, and one to model solid equations in poroelasticity. At the next stage, the two cell centred codes were iteratively coupled. Then, another 2D fully coupled model based on a vertex-centred finite volume discretization scheme was developed for poroelasticity. The fl uid and solid data structure in both developed models are the same; in other words, unknowns are collocated. To verify and conduct error analysis on the developed finite volume based simulators, 2D MATLAB codes for one classic benchmark problem in poroelasticity, namely the Mandel problem, as well as 2D MATLAB codes for synthetic test cases were developed. At the next stage, the performance of the two models is compared. Though both methods illustrate adequate performance, fully coupled finite volume method is preferred. Finally, the reaction of the fully coupled finite volume model to different systems, its performance under stress and displacement boundary condition configuration, sensitivity of the model to uncertainty of input parameters, robustness of the model, and applications of this method are being analysed. Furthermore, thorough discussion on enhancement of the model is provided. To our knowledge, it is for the first time that a cell-centred finite volume scheme is applied to poroelastic problems and is compared with the vertex-centred finite volume method. In addition, there is no prior investigation done of robustness for finite volume methods. The results showed that both cell-centred and vertex-centred FVM are computationally efficient in applications for poroelastic problems. However, the fully coupled vertex-centred finite volume model outperformed the sequentially coupled cell-centred model in terms of computational efficiency.Civil Engineering and GeosciencesGeoscience & EngineeringPetroleum Engineering and Geoscience
3D fracture analyses of various rock samples through x-ray micro-tomography
No full textHydrocarbon reservoirs are currently declining. Reservoirs are becoming depleted, and production needs to satisfy the demand and supply of hydrocarbons. For tight reservoirs such as shales, fracturing will become the main source of improving the permeability. It is therefore vital to conduct research into the behavior of fractures in the reservoir. The aim is to study how fractures develop in terms of fracture size, coalescence, patterns and in which dimensions they often occur. Often the importance of the behavior of these fractures and their contribution to flow in reservoirs also formed part of this research. In this investigation, the Posidonia shale core was examined using three different layering orientations. The three different layer orientated shale cores were subjected to axial loading in order to induce fracturing. These induced fractures were created in two ways. The first was by axial loading until the core reached maximum failure. The second involved loading in stages and scanning the core after each loading stage in the micro-CT scanner. This procedure was repeated until maximum failure was reached. An Indiana limestone core and sandstone core were also analyzed. The behavior, coalescence, frequently seen fracture patterns, angle with respect to the z-axis and apertures of the fractures were studied and compared. This was performed in both a quantitative and qualitative matter using the commercial Avizo® Fire software at TU Delft. A small part of this study touch on the modeling of the observed shear fracture pattern from a shale core preformed in Abaqus. The results from the fracture analyses shows that the largest fractures are seen in the shale cores, for which the fracture initiation was created by one loading stage until failure. Another observation shows that there is clear set of different fractures that is seen for each distinctive layered core. From the observed patterns a simple 2D simulation is built in Abaqus. From the simulation is seen that for cracks under a certain angle, the master fracture is intersected by a 300 smaller fracture, the results in opening of the master fracture. This is also the case in the 2D slice, which is a representation of the 2D-model fracture configuration.Civil Engineering and GeosciencesGeoscience & Engineerin
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
