Frash, L. P. (Los Alamos National Laboratory) | Arora, K. (Colorado School of Mines) | Gan, Y. (Colorado School of Mines) | Lu, M. (Colorado School of Mines) | Gutierrez, M. (Colorado School of Mines) | Fu, P. (Lawrence Livermore National Laboratory) | Morris, J. (Lawrence Livermore National Laboratory) | Hampton, J. (New England Research)
ABSTRACT: It is possible to engineer and control the extents of the stimulation rock volume for hydraulic fracturing. Currently, available tools and methods intended to accomplish this task focus on optimizing injection fluid properties, utilizing existing rock stress boundaries, controlling stimulation intervals in the injection well, and manipulating injection pressures and rates. What if it were possible to control hydraulic fracture extents more directly than these methods do and to have confirmation of these extents in the subsurface? For this, we propose a ‘fracture caging’ concept where an array of injection wells and production wells are drilled prior to stimulation as a means to identify and control the extent of a stimulated zone. Positive identification of stimulation extents occurs by monitoring production well pressures and flow rates. Control of fracture extents occurs by control of the production well pressures and arrangement of production wells so as to contain an intended stimulated zone. In this study, we present the fracture caging concept and validate it with laboratory experiments. Numerical modelling with LLNL’s GEOS code is used to predict the effectiveness of the fracture caging concept as it applies to the SIGMA-V (EGS Collab) geothermal energy research field site.
Control of hydraulic fracture propagation and flow by current methods relies on careful and tailored selection of injection fluid properties to be compatible with the rock properties at a given location. The specific methods used in practice vary significantly, including tools such as frac- ability or brittleness indices (Jin et al., 2004; Herwanger et al., 2015; Govindarajan et al., 2017), non-dimensional analysis of viscosity vs toughness regimes (dePater et al., 1994; Bunger et al., 2005), and models ranging from simple 2D analytical (Perkins and Kern, 1961; Geertsma and de Klerk, 1969; Nordgren, 1972) to full 3D with heterogeneity, pre-existing fracture networks, and more included (Fu et al., 2013). To complement these methods, the in-situ geologic boundaries are used because high- stress strata have been shown reliably effective for limiting hydraulic fracture height growth (Warpinski et al., 1982). The effectiveness of these design choices for controlling hydraulic stimulation can be verified by methods including proppant/tracer logging (Smith et al., 2013), distributed tiltmeter surveys (Wolhart et al., 2007), distributed microseismic monitoring (Rutledge et al., 2004; Warpinski, 2009). However, each new stimulation is subject to significant uncertainty due to the geologic structures that cannot all be identified and characterized a-priori with enough fidelity to predict precisely where fractures will go (Warpinski and Teufel, 1987). Methods that are less subject to geologic uncertainty could be instrumental for improving the effectiveness of hydraulic stimulation treatments.
ABSTRACT: In order to constrain the contribution of mechanical deformation to injectivity decline in an existing Gulf of Mexico (GoM) field, the hydromechanical behavior of the Castlegate sandstone was studied at close to in situ conditions of pressure and temperature with the goal of devising methods and expectations regarding further work on a waterflooded sandstone formation. In addition to the use of high temperature and pore pressure, the stress conditions for brittle failure were probed using pore pressure buildup as the driving force as opposed to conventional triaxial loading. Through a methodical analysis of the effects of thermal and mechanical loading on flow properties, we observe a very substantial loss of permeability during the attainment of in situ conditions before the initiation of the pore pressure buildup, with the potential of increasing permeability hindrance due to fines migration. Brittle failure during pore pressure buildup is associated with the formation of a thin dilatant shear fracture and moderate permeability reduction. Follow-up work will be aimed at realizing the integration between geomechanical modeling and reservoir core analysis in order to provide a comprehensive understanding of the physical mechanisms involved in waterflooding operations and formulate operational guidelines.
Improvements in recovery factor due to waterflooding of conventional reservoirs depend for a large part on displacement and sweep efficiency, but can also be critically affected by flow hindrance near the injectors. Among the factors that can potentially contribute to injectivity decline, the ones associated with the geomechanical response of the formation to injection are still poorly understood. Depending on the prevailing stress and strain boundary conditions, pore pressure fluctuations may locally drive the formation into failure in a number of ways, which include the generation of localized shear bands that can profoundly impact fluid flow. Other potential contributors are the high strain rates that are experienced during shut-ins and resumption of waterflooding operations, as well as the temperature contrast between the injected water and the formation which can cyclically induce further strain and microscopic damage. Finally, mechanical damage is inevitably accompanied by the generation of fines through grain detachment, microcracking and comminution. These fines are expected to cause some amount of flow hindrance, either by locally reducing the effective flow radius at the microscopic scale or by forming flow barriers through transport and accumulation.
In this work, we conducted laboratory experiments on samples of Castlegate sandstone, which was identified as a petrophysical and mechanical analogue for a hydrocarbon reservoir presently under study. The objectives were to devise a method, workflow and expectations regarding the ultimate application to a field undergoing water injection. Steady state liquid permeability was monitored throughout the tests, and the experiments also included the measurement of P-wave and S-wave velocities.
ABSTRACT: This paper represents an effort to outline some issues and potential guidelines associated with the integration of rock images with laboratory data. It is entirely based on the use of X-ray computed tomography (CT) imaging, which has become more readily accessible over the years and allows for non-destructively imaging geomaterials in three dimensions over many length scales (~10−9 to 100 m). While the number of studies involving X-ray CT scanning has been steadily growing, there is a paucity of rules regarding the conduct of that imaging work compared to the well established set of protocols that exist around all aspects of experimental geomechanics. To correct for this imbalance and build a common experience within the community, the identification of image and lab data integration as a discipline could be of tremendous help as it would allow to devise principles and focus efforts on areas most in need of improvement. More importantly, this new discipline would naturally take on the role of building image-based predictive petrophysical and geomechanical models by leveraging many decades of research on the microstructural control of physical properties. Such image-based modeling effort would require to shift the role of high performance computing tools from simulation to description, with two immediate advantages: (1) A very efficient predictive capability compared to numerical simulations (through constitutive laws) and (2) The associated benefit of phenomenological understanding.
X-ray CT images can be used to assess sample integrity, as well as to study the relationship between microstructures and measured properties, either through ad hoc comparisons or by applying a variety of effective medium models. More recently, X-ray CT images have been increasingly employed for conducting direct numerical simulations, providing property estimates for digital rock samples that would otherwise be difficult or simply too time consuming to physically obtain and test. While a number of standard protocols exist to conduct laboratory testing on rock samples, there usually are no specific expectations on the part of imaging in terms of e.g. field of view, resolution or calibration. Moreover, as sample dimensions are usually fixed, some amount of heterogeneity should always be expected, what can substantially impact laboratory results. Such heterogeneity information can only be gained through imaging of the entire tested volume.
In-situ horizontal stresses are key variables for planning hydraulic fractures in the Bakken Play of North Dakota, and other unconventional assets. The oil and gas industry typically uses a poroelastic model to estimate horizontal stresses. This model requires knowledge of poroelastic properties, the vertical or overburden stress, pore pressure, and the two horizontal elastic principal strains (assuming one principal strain is vertical). Usually some of these parameters are poorly known; most workers make simplifying assumptions about the poroelastic properties and about the horizontal elastic strains. Typical simplifying assumptions are (a) one or both of the horizontal strains are assumed to be zero, and (b) the two independent Biot-Willis effective stress coefficients are assumed to be 1 or, if less than 1, they are assumed to be equal. To test these assumptions, we used extensive laboratory and field measurements to constrain the poroelastic model for the Middle Bakken reservoir in North Dakota. We measured the relevant poroelastic properties on core plugs and found that the Biot-Willis effective stress coefficients were less than 0.8, with significantly different horizontal and vertical components. In conjunction with previously measured in-situ stresses, these properties imply that the horizontal elastic strains are on the order of 10−4 or greater. These strains imply a difference of 103 psi (7 MPa) or more between the actual and the computed minimum horizontal stress that assumes zero horizontal strain. This difference is quite significant in the context of hydraulic fracturing and demonstrates that the common assumptions about strain and effective stress coefficients are incorrect for the Middle Bakken. This study is an initial step toward understanding the variations in lateral strain across the area of interest.
Presentation Date: Thursday, September 28, 2017
Start Time: 9:20 AM
Presentation Type: ORAL
ABSTRACT: This study reports on strength measurements performed on sandstone samples from a shallow offshore field and is part of a larger petrophysical evaluation including static and dynamic elastic properties under various stress conditions. The samples tested were retrieved from three one-foot-long preserved core sections spanning a depth range of about 20 meters. For each depth, three types of mechanical tests were performed: brazilian tensile test, unconfined compression test (UCS) and confined compression test with ultrasonic wave velocity monitoring along and perpendicular to the plug axis. The mechanical data are augmented with X-ray CT image volumes acquired pre-test and post-test in order to elucidate some of the observations as well as to outline possible elements of a laboratory workflow that would comprise image-based heterogeneity assessment and prediction. We emphasize in particular the fact that if such workflow is to be standardized, care should be taken in acquiring properly calibrated plug scale X-ray images in order to allow heterogeneity quantification and its use as an input into future predictive models.
Geomechanical testing of rock samples in the laboratory is typically limited in that every measurement accounts for the sample behavior as a whole with seldom acknowledgement of its internal degree of preexisting or induced heterogeneity. Preexisting heterogeneities such as mechanically weak layers in strongly laminated rocks may cause early failure. This effect, which causes a rock to behave in a strongly anisotropic fashion since the weaker surfaces need to be presented at an appropriate angle to the maximum stress to fail, is exemplified with the data of Figure 1 borrowed from Paterson and Wong . In order for such an effect to be predicted, weak layers need to be identified, which involves imaging since no bulk measurement such as porosity, density or composition is able to capture that information. On the other hand, induced heterogeneity such as localized shear bands formed during brittle failure is known to strongly impact fluid flow, and cannot be accurately predicted in terms of e.g. number, thickness and orientation. Nondestructive 3D imaging greatly complements laboratory measurements by providing access to the internal make-up of a rock at various stages of a test. Moreover, it does seem sensible to expect that a core plug will often exhibit some degree of heterogeneity, as plug dimensions are not determined based on an assessment of the representative elementary volume (REV) for the material of interest.
ABSTRACT: In this paper, we report on a scoping study that was prompted by operational issues through an Oligocene smectite-rich shale that involved changes in borehole inclination with respect to the bedding. A core characterization workflow is used to specifically probe geomechanical heterogeneity and anisotropy for static and dynamic elastic properties as well as failure strength. Initial petrophysical scanning of the core surface provides a first indication of existing heterogeneity for properties of interest and assists in devising an efficient sampling strategy. Over the three-foot section analyzed, and despite its apparent homogeneity, the core exhibits a two-fold variation in reduced Young’s modulus between softer and stiffer zones, which is tied to slight changes in carbonate content. Confined elastic and mechanical measurements reveal strength anisotropy of the order of 20% and P-wave and S-wave velocity anisotropies of about 20% and 30%, respectively. Moreover, testing shows that the shale is weakest at oblique angle to the bedding due to weak bed parallel surfaces which activate when favorably oriented. These results suggest that anisotropy and heterogeneity both need to be accounted for in borehole stability models involving smectite-rich material.
Accurate wellbore stability prediction in geomechanically unstable formations requires thorough understanding of the drilled rock properties. This includes the ability to predict failure in deviated wells associated with bedding heterogeneity or to better assess the relationship between intrinsic elastic properties and stress/strain boundary conditions for e.g. in situ stress computations and log-based geomechanical forecasting.
This paper presents a geomechanical core analysis workflow that includes petrophysical core scanning for heterogeneity assessment and sample picking, as well as geomechanical testing for anisotropic static/dynamic elastic properties and strength. In particular, the petrophysical scanning includes a mechanical probe called the Impulse Hammer which functions by analyzing the force-time function of a hardened steel sphere mounted on an accelerometer dropped on the surface of the rock. This analysis produces a reduced Young’s modulus at a resolution on the order of the millimeter revealing fine scale heterogeneity. Using the profiles obtained during petrophysical scanning, locations of interest can be chosen for further geomechanical evaluation on plugs.
ABSTRACT: The objective of this paper is to provide an overview of a new data-based approach to understanding and modeling strength anisotropy in porous sandstone with potential application to a wider range of geomaterials. The rationale for the contribution is the interest in quantifying the impact of mechanical anisotropy in practical industrial applications such as the prediction of mechanical behavior of hydrocarbon reservoirs during depletion or injection. The onset of yielding can have profound impact on flow and elastic properties, and it is still unclear to what extent anisotropy may affect those predictions. After recalling an extensive mechanical data set obtained previously on samples of Rothbach sandstone, we define our modeling strategy using other experimental and theoretical work from the literature. This results in the necessity of determining what an appropriate isotropic yield envelope might be. We propose an approach whereby isotropic and anisotropic models are defined and parametrized sequentially. We obtain a consistent framework where the microstructural controls on strength, including anisotropy, may be understood better than previously, allowing the revisiting of legacy data sets as well as the designing of better informed geomechanical testing programs in core analysis.
Mechanical anisotropy is known to adversely affect strength predictions, especially when associated with the presence of weak layers [1-3], which can result in unexpected borehole failure. In homogeneous materials, anisotropy has been showed to directly affect the shape of yield surfaces [4-7] and can also cause the strain field to be oblique to the macroscopic stress tensor. In this latter case, an important question to be answered is to what extent mechanical anisotropy should be incorporated into geomechanical models, and how does it ultimately compare with work that has been done on the effect of true triaxial states of stress .
The present paper builds on an extended petrophysical and mechanical data set that was acquired over the years in the Rothbach sandstone, a Triassic 20% porosity cross-bedded fluvial sandstone belonging to the Buntsandstein Formation in Northern Europe [4; 9-16]. The Rothbach sandstone is anisotropic with respect to key physical properties such as permeability, acoustic velocities and mechanical strength. As far as mechanical strength is concerned, this sandstone presents the particularity of being stronger when loaded perpendicular to the bedding, and its bedding plane does not appear to constitute a plane of weakness when favorably oriented with respect to the maximum compressive stress. Samples cored at 45 degrees to the bedding exhibit intermediate strength with respect to the ones cored parallel and perpendicular to it .
ABSTRACT: Nanodarcy permeability is common in many oil- and gas-bearing unconventional formations requiring stimulation technologies, such as hydraulic fracturing, which can connect the isolated hydrocarbon-rich pores to a wellbore through induced fracture networks. Even though a dense induced fracture network is capable of accessing economic amounts of hydrocarbons, much of the rock between and very close to the fracture network is still un-accessed. Inducing fracture networks into rock can create large amounts of micro failures in surrounding regions that are not connected to the wellbore. Regions of rock containing microcracks near coalesced macro-scale fractures oftentimes behave differently than the original matrix material due to the permanent structural changes. These changes can manifest themselves in mechanical and petrophysical alterations from the original matrix conditions. Understanding that the coalesced fractures must drain the reservoir rock through these regions containing microcracks requires the characterization of damage within rock in terms of mechanical and petrophysical changes. In this study, a laboratory hydraulic fracture test was performed on a two-block system separated by a discontinuity as an analogue to a large natural fault. The induced hydraulic fracture was monitored with acoustic emissions (AE) throughout initiation, propagation, and interaction with the large fault. Individual AE event source characterization was performed to obtain mode of failure and relative volumetric deformation. Source characteristics were used in conjunction with cloud-based event density techniques to determine regions of differing damage within the cloud of microcracks. Quantitative three-dimensional event density imaging results were compared with permeability measurements on sub-cores taken from the sample post-test. Inverse relationship between AE event densities and permeability of sub-cores was observed, meaning that reductions in permeability were found nearest coalesced hydraulic fractures in the crystalline rock tested.
Hydraulic fracturing has become a standard practice in oil- and gas-bearing formations with low permeability reservoirs especially in unconventional hydrocarbon resource development. Fracture complexity and/or fracture network generation is desired to create a dense hydraulically conductive pathway reaching as many isolated hydrocarbon-rich pores as possible. Although economic amounts of hydrocarbons can be reached, much of the rock very near the fractures, and sometimes within the network, is not accessed because of the very low permeability of the reservoir rock. It has become desirable to understand the changes induced within the surrounding rock not directly or hydraulically connected by a conductive pathway or fracture. This rock is typically associated with the longer-term fluid transport commonly seen in the slow tail of well production curves.
The present paper concerns itself with the use of morphological information from the 3D image of a rock microstructure to extract parameters needed to model the compaction of porous sandstones. We propose to test on the mineral framework a tool that is already employed in the simulation of mercury injection capillary pressure experiments (MICP) (see [1-2] for implementations and  for background on MICP) and, in particular, we investigate the existence of a characteristic grain contact radius. As a starting point, results from hydrostatic loading of two porous sandstones of similar porosities, the Castlegate sandstone (?~26%) and the Boise sandstone (?~29%), are presented. The mechanical data reveals a factor of almost 4 between the values measured for the critical grain crushing pressure P*. As a way to connect microstructural parameters to the observed strength contrast, we test the use of the morphological analysis on high resolution X-ray CT images of both rocks. In comparing our findings with the already existing model of Zhang and Wong [4-5], we propose that the intergranular contact radius information extracted from the image analysis be explicitly incorporated into the modeling of the strength of porous sandstones.
Mechanical response to depletion, which comprises irrecoverable volumetric strain as well as elastic deformation, strongly depends on in situ conditions and on the nature of the corresponding perturbation in terms of stress path, strain rate, fluid substitution, etc. The ability to forecast this behavior, whether for pressure support or subsidence risk assessment, hinges on our understanding of deformation mechanisms at the scale of the aggregate, their interplay with preexisting heterogeneities and their manifestation at the scale of the reservoir.
Modeling of mechanical properties traditionally relies on microstructural parameters such as porosity, mineralogy, coordination number, cemented contact area, grain size and shape, which are combined to account for trends obtained in laboratory measurements. The now widespread availability of 3D pore scale imaging techniques allows one to access the intimate make-up of a rock, offering in principle a means to fully quantify and validate the parameters used for pore-scale modeling. It also provides an opportunity to identify which of these parameters control resulting behavior, whether redundancies exist from a physical point of view, and whether they can even be measured in a meaningful way. The option of performing direct numerical simulations based on pore scale images is being increasingly utilized to complement costly laboratory measurements . However, an understanding of the key controls of the observed behavior remains essential for generalizations to be made.
The knowledge of the in situ stress state around a borehole is of primary importance when estimating the stability of the borehole, or designing an optimal drilling trajectory in order to minimize the circumferential stress around the borehole. In this paper, we present an approach for calculating the stress field surrounding a borehole in rock with nonlinear anisotropic elastic properties and for predicting how that stress field changes with applied tectonic stress. Using these results we then resolve the complex velocity field in the region near the borehole. I
One of the primary uses of acoustic logs, in particular, dipole logs, is to measure the formation anisotropy. Previous studies have shown that the formation shear wave anisotropy varies with the distance away from the borehole, and can be used to distinguish between intrinsic and stress-induced anisotropy. The knowledge of the in situ stress state around a borehole is of primary importance when one needs to estimate the stability of the borehole, or to design an optimal drilling trajectory in order to minimize the circumference stress around the borehole. Sinha and Kostek (1995) calculated the changes in the local stresses away from the borehole along the principal stress directions. They then use a non-linear stress-strain relationship to calculate the shear wave velocity changes along these two directions. The velocity profiles are then used to estimate the flexural wave dispersion characteristics. Winkler (1996) studied the velocity variations around a borehole in a laboratory specimen and found a clear relationship between velocity anisotropy and the principal stress directions. Plona et al. (2004) gave a good summary of the use of flexural wave dispersion to estimate the in situ stress state around the borehole. However, the approach used by Sinha and Kostek (1995) is only an approximation to a rather complex problem. A complete analysis must consider the complex constitutive relationship between an anisotropic applied stress field and the stiffness tensor for a rock with micro-cracks embedded in the matrix. In this paper, we present a more general approach for calculating the stress field surrounding a borehole in rock and how that stress field results in a complex distribution of velocities near he borehole.
A classic problem in the field of linear elasticity is the computation of the stress concentration around a hole due to an imposed stress field. For a homogeneous linear elastic material the solution provides the functional form of enhanced tensile and compressive stresses around the hole, their location with regard to the imposed stress, and how these stresses fall off with distance away from the hole. This solution has myriad uses in applied engineering problems, including the geomechanics problem of calculating the locations of potential borehole failure with azimuth. However despite the marked stress concentrations, the dynamic elastic moduli will not change in an ideal homogeneous linear elastic material, thus the acoustic velocities will be unaffected by the hole. A thorough understanding of the rock and damage mechanics can provide a model in this case.