ABSTRACT: Fiber optic based distributed acoustic sensors (DAS) provide a new approach for monitoring signals of interest in the subsurface with unprecedented spatiotemporal resolution. These sensors produce measurements that are fundamentally different from their traditional counterparts, such as geophones, and produce significantly larger volumes of data. To interpret these data, we begin by using the physics-based thermal-hydraulic-mechanical (THM) model in the GEOS code to simulate synthetic DAS measurements for a range of subsurface conditions (fracture propagation, fault slip, etc.) and sensor configurations (e.g.: horizontal or vertical well deployments). These synthetic DAS measurements are then algorithmically labeled based upon features of interest within their parent models, such as the extent of any generated hydraulic fractures or the distribution of proppant particles, and are compiled into a database. The synthetic database can be used to train and test an initial deep neural network (DNN) representation of the subsurface, which can then be optimized by incorporating any available field measurements through transfer learning. This hybrid, physics informed DNN model is capable of interpreting DAS measurements in near-real time, making it a useful tool for decision making by field engineers, and works under both data-rich and data-poor conditions. To demonstrate this approach, we consider the problem of imaging hydraulic fracture propagation in an unconventional oil and gas reservoir. Our results indicate that we are can use a trained DNN model to accurately estimate the extents of a hydraulic fracture using the location information and DAS measurements for a single fiber-optic sensor as inputs.

ABSTRACT: In this paper, the algorithm of the hybrid finite-element meshfree method (FEMM) is first used to simulate three-dimensional thermal cracking in the context of the TOUGH-FEMM simulator. The TOUGH-FEMM simulator combines two existing codes sequentially, TOUGH2 and FEMM. The temperature distribution of the system is analyzed using TOUGH2, which is able to solve coupled problems of nonisothermal, multiphase, multicomponent fluid flow in complex geological systems. Recently, hydro-mechanical models based on FEMM and TOUGH2 have been coupled together for the analysis of hydraulic fracturing in fractured rock. The thermal stress, rock deformation and cracking induced by temperature change are modeled by FEMM. The finite element meshfree method (FEMM) is presented to model mechanical deformation and discontinuities. With the concept of partition of unity, the accuracy of global approximation can be improved without adding extra degrees of freedom (DOF), and various local approximation functions can be used for modeling strong or weak discontinuities. Linear elastic fracture mechanics (LEFM) with mixed-mode stress intensity factors are employed to determine the initiation and direction of crack propagation. Several benchmark examples, including stress intensity factor calculation, steady-state heat conduction, transient heat conduction, and deformation induced by thermal stress, are provided to validate TOUGH-FEMM in dealing with thermo-mechanical problems with cracking. Additionally, a numerical experiment of crack propagation induced by thermal stress is also given, and its results are compared with the experimental results. The simulation results agree well with the analytical solutions or experimental results, indicating that the proposed approach is effective for thermal cracking simulation.

ABSTRACT: In this study, we present the results of numerical simulations of washout behavior of granular geomaterials. The washout of soil grains, which is driven by a fluid flow, is considered to be one of the most significant reasons of ground subsidence particularly in urban areas accompanies by underground excavations since the excavation may disturb and cause a groundwater gradient. For this, a computational fluid dynamics (CFD) module was implemented into particle flow code (PFC) so as to include a seepage force in the particle displacement calculation of the conventional PFC. Through this coupling, both particle dislocation created by fluid flow and porosity/permeability change induced by the particles redistribution are considered. To verify the validity of the current coupled PFC-CFD analysis, we compared our numerical estimation of washout rate to that of the previous laboratory experiments. They showed well comparable results for two comparative cases of stable and unstable conditions. We then performed a parameter sensitivity analysis to identify the influential parameters on internal stability against a washout behavior and showed it is the effect of particle size distribution, the area where fluid flows out and infiltration flow rate that affect a washout rate.

ABSTRACT: The dissolution of rocks in underground flow paths is transport-controlled if their dissolution rates are relatively high, while the dissolution is reaction-controlled if their dissolution rates are relatively low. Transport-controlled dissolution is a common process in the formation of gypsum karst and oil reservoir acid stimulations. As an initial step to understand groundwater flow and dissolution in fractures and in wormholes, pipe flow is used as a simplified representation of groundwater flow. The extended Graetz solution was developed by Li and Einstein, 2017 to simulate the transport-controlled dissolution process when water flows through a tube in gypsum. This model can predict both the effluent concentration and the evolving geometry of the tube during the dissolution process. This paper focuses on the experimental validation of this model. An effluent chemistry monitoring system (ECMS) was developed and integrated in the triaxial system at MIT. Flow tests were conducted with cast gypsum tubes in this triaxial system. The experimental results of the effluent chemistry and tube geometries confirmed the validity of the extended Graetz solution.

ABSTRACT: Account for fluid-solid interaction based on a locally averaged approach (Anderson & Jackson, 1967, Tsuji et al., 2008, Furuichi et al., 2014) Directly simulate hydrodynamic forces on the solid particles (Noble & Torczynksi, 1998, Holdych, 2003, Holzer & Sommerfeld, 2009) Scour of rock in unlined rock spillway channels is a critical issue facing many of the world’s dams. From a modeling point of view this poses a challenging and interesting problem that combines rock mechanics and hydraulics of turbulent flow. We analyze this interaction between the blocky rock mass and water by directly modeling the solid and fluid phases—the individual polyhedral blocks are modeled using the Discrete Element Method (DEM) while the water is modeled using the Lattice Boltzmann Method (LBM). The LBM mesh is entirely independent of the DEM discretization, making it possible to refine the LBM mesh such that transient and varied fluid pressures acting of the rock surface are directly modeled. This provides the capability to investigate the effect of water pressure inside the fractured rock mass, along potential sliding planes, and can be extended to rock falls and slides into standing bodies of water such as lakes and reservoirs. Herein we present preliminary results to demonstrate the capabilities of the methodology.

ABSTRACT: Dynamic friction slip of rock joint is prevalent in seismic wave propagation and plate sliding. Great concerns have been taken to reveal the whole process of dynamic friction slip, i.e. the wave-induced micro slip, the inertial slip, and the plate sliding. Series of measuring techniques including the multi-point acoustic emission method and the multi-point infrared temperature method are established based on the modified split Hopkinson pressure device (SHPB), and the characteristics of acoustic emission and heat production in the initial stage of friction sliding on the joint surfaces are investigated. The results show that wave profiles corresponding to the wave-induced micro slip is located at the higher frequency domain of the acoustic emission signals, e.g. frequencies ranging from 78.125 kHz to 312.5 kHz for granite specimens with size 50×50×50 mm 3 . While the signal measured at the lower frequency domain characterizes the propagating of the Rayleigh wave. Wave profiles corresponding to the heat production is located at the lower frequency domain in the temperature signals. The measured amplitudes of elevated temperature are in order of 0.01 °C, which increases slightly with increasing the joint angle. To reveal the intrinsic rules of these measured signals, a reasonable analysis model based on the numerical method is established. The results show that when dynamic slip occurs, a slow front generates and propagates along the joint surfaces. With increasing the joint angle, the amplitude of elevated temperature on the joint surface increases. The calculation results are in agreement with the experimental results.

ABSTRACT: In the framework of the In-situ Stimulation and Circulation (ISC) experiment Fiber-Bragg-Grating (FBG) and Brillouin strain sensing systems were installed to monitor deformation during six hydraulic shearing and six hydraulic fracturing experiments. Three boreholes were dedicated to strain monitoring. Both systems are installed in the same boreholes, offering a unique opportunity to compare these systems with respect to their applicability in hydraulic stimulation tests. A total of 60 FBG sensors with 1 m base length were installed across fractures, shear zones and intact rock. Along the entire borehole length, pre-stressed optical cables for Brillouin distributed strain (DBS) sensing were embedded in grout with two installation methods: a bare cable and a cable packed and fixed with glue every 0.65 m. The strain signals were compared as time series for a given borehole depth and as profiles along the borehole axis. The study reveals that the FBG system gives a high accuracy (0.04 µ-strain) and temporal resolution (>1s) with pointwise measurements. The bare DBS leg yield good quantitative strain data with poorer strain accuracy (>500 times poorer than FBG) and poorer temporal resolution (factor of >100). The packed DBS leg provide no meaningful information about the strain field

ABSTRACT: For the storage of natural gas, compressed air or hydrogen in rock salt caverns it is necessary to carry out numerical calculations for the dimensioning of the relevant operation parameters. These are on the one hand the allowable maximum and necessary minimum internal pressure to ensure a gastight and stable cavern over decades and on the other hand the limitation of the withdrawal and refill rates between these geomechanical limits.

ABSTRACT: Underground cavities in rock salt have received increased attention for the storage of oil, gas, and compressed air energy. In this study, the transition between secondary and tertiary creep in salt is determined by a micro-macro model: The initiation of grain breakage is correlated with the acceleration of viscoplastic deformation rate and with the initiation of damage at the macroscopic scale. Salt stiffness decreases when macroscopic damage increases, which allows predicting the evolution of the damage zone around salt caverns used for geological storage. After implementing the phenomenological model into the Finite Element Method (FEM) program POROFIS, two thermo-mechanical coupled stress paths are simulated to analyze stress concentrations and viscous damage around a 650-m-deep cavern in axisymmetric conditions. Numerical results indicate that, despite the pressurization or depressurization-induced temperature variation, internal gas temperature always tends to approach the primary surrounding rock mass value. The viscous deformation induced by thermo-mechanical couplings significantly affects the original stress field at the cavern wall and induces high damage at the most concave sections of the cavern. Results reveal the significant influences of idle time, gas pressure range, and injection and withdrawal cycles on stress, strain and temperature distributions in the vicinity of the cavern. More analyses are needed to confirm the influence of thermo-mechanical cycles of pressurization and depressurization, and to design long-term cavern operations. 1. INTRODUCTION

ABSTRACT: Tracking the development of hydraulic fracture (HF) propagation in the laboratory and in the field has always been difficult, especially in cases where multiple HFs are propagating simultaneously. A new method is presented using fiber optic (FO) technology to monitor the hydraulic fracturing process during real time. The method has been proven using FO cables ‘woven’ in a grid pattern and cast inside 46 cm (18 inch) cubic blocks of cement. Some of these blocks were then placed in a large cubic polyaxial load frame and subjected to a true triaxial state of stress during the fracturing operation. The FO cable acts essentially as a long continuous ‘strain gage’ in the direction of the cable. As a HF propagates across each FO cable segment it sharply strains each one. This process allows a real time tracking of the position of the growing hydraulic fracture crack tip during testing. The technique is being used to track single HFs, multiple HFs, and leak- off haloes in polyaxial HF block testing.

ABSTRACT: Recent research at Montana Tech has focused on demonstrating the effectiveness of the emerging fiber-optic-based distributed strain and temperature (DST) and distributed acoustic sensing (DAS) technologies in an underground environment, facilitating their adoption by the mining industry and thereby contributing to mine safety. Deployment of a fiber optic array underground was used as base to compare the performance of this new technology to that of traditional instruments deployed at the site, and to compare performance of different brands of fiber optic cable. From the lessons and techniques learned during this deployment, an underground fiber optic array protocol has been developed to aid future engineers in the utilization of this tool.

Abstract: Understanding thermal properties is necessary for the development of both shallow and deep geothermal systems. Thermal properties can also be used for understanding the development and assessing weathering in a variety of materials. An experimental apparatus has been developed to determine the thermal properties of rock samples. The apparatus consists of a ring heater, aluminum adaptor, and a stainless steel base. Three thermocouples are used to measure heat flow through the base. The rock specimen sits on top of the stainless steel base. Twelve thermocouples, arranged vertically in sets of three at the cardinal positions around the rock sample, are used to measure heat flow through the sample. The apparatus is wrapped in insulation to prevent heat loss. The top of the apparatus is open so a thermal camera can capture the temperature increase at the top of the specimen. Thermal analyses conducted using finite element modeling have verified the design of the apparatus. Two limestone specimens have been tested and experimentally derived thermal conductivities are within experimental ranges of thermal conductivities presented in the literature. Introduction Understanding the thermal properties (thermal conductivity, thermal diffusivity, and heat capacity) of rock is important for regional geothermal and ground source heat pump applications. Ouali (2009) studied the variation in thermal conductivity of a stratigraphic column with varying porosity in Reykjavik, Iceland. Goto and Matsubayashi (2009) investigated the correlation of thermal properties in the Juan de Fuca Ridge as a function of porosity and depth. Additionally thermal properties may provide insight into the weathering state of specimens. Banks et al. (2010) studied the use of thermography for structural health monitoring of subsurface anomalies in porous materials. The thermal properties of rock can be estimated based on the rock porosity with varying accuracy.

Abstract: The mechanical response of the target formations during Carbon geological storage strongly depends to the fluid pore pressure alterations in the formation. The storage formations must have sufficient capacity and could avoid migration of CO2 to the surface. When the highly pressurized CO2 is injected into the geological repository the fluid pressure increases in the formation, which results in changes in stresses and deformation of the medium. From geomechanics point of view, the pressure caused by CO2o injection, should not exceed the formation strength and should not cause activation of existing faults. Therefore, finding a realistic estimation of alteration of stresses during the CO2 injection job, has been subject of several studies in the literature. Most of the analytical methods available in the literature to calculate the stress distribution are well established for single-phase flows, but it requires extension when a second fluid, in this case CO2, is also flowing. In the current work, an approximate analytical model is developed for calculating different stresses caused by injection of CO2 in a saline formation, assuming two phase flow pressure regime. Here, we investigated the stress regime under induced fluid pressure and temperature alterations during the injection time. Introduction Carbon dioxide storage or carbon dioxide sequestration refers to the processes by which captured CO2 is securely stored in deep geologic formations. Carbon dioxide storage in geologic formations includes oil and gas reservoirs, unmineable coal seams, and deep saline reservoirs. It has been reported by The Intergovernmental Panel on Climate Change (IPCC) that the global capacity of deep saline storage sites is more than thousands of gigatons of CO2, which is hundreds of times greater than the annual CO2 emissions from industrial sources [1–3]. Therefore studying the behavior of CO2 when stored in these sites are crucial for the industry.

Abstract: High injection rate hydraulic fracturing can have Reynolds number as high as 10 4 . For such high Reynolds numbers, turbulent flow is likely to occur. In this paper, we investigate the effect of turbulence on the propagation of hight contained hydraulic fractures, commonly referred to as PKN fractures. We discuss different scalings for the fracture width, length and pressure under limiting laminar, turbulent smooth and turbulent rough flow regimes. We implement an explicit, central numerical scheme to solve the continuity and friction factor based momentum conservation equations, taking into account the full variation of friction factor with Reynolds number and relative fracture roughness. The scheme is validated against the analytical solution of the PKN model. The results show that the local Reynolds number evolves from a maximum value at the inlet to zero at the tip, with a transition from turbulent to laminar at some point along the fracture length, depending on the value of inlet Reynolds number. Results showing the effect of smooth and rough turbulence on the fracture length and fracture width depending on the Reynolds number are finaly presented. Introduction The usual approximation of laminar flow in a propagating hydraulic fracture sometimes breakdown. This is the case of fluid-driven fracture propagation at glacier beds where Reynolds numbers above 10 5 are expected (Tsai and Rice (2010)). At such large Reynolds numbers, the flow is well within the turbulent rough regime. Recently, Ames and Bunger (2015) have investigated the effect of a fully rough turbulent flow on the propagation of height contained hydraulic fracture (Nordgren (1972)) and derived scaling relationship for that flow regime. However, in practice for a hydraulic fracture treatment in an oil and gas reservoir using a low viscosity fluid injected at very large injection rate, the Reynolds number for a height contained fracture are expected to be in the range 10 3 – 10 4 at most (see Table 1 and 2), right in the transition between the laminar to the fully turbulent regime. We focus solely here on a height contained fracture as depicted in Fig. 1 (i.e. the so-called PKN geometry). In parallel to the problem formulation, we also review the available experimental data and models covering the complete flow conditions spanning the complete spectrum of Reynolds numbers from laminar to fully turbulent. We notably discuss the use of an equivalent Reynolds number proposed by Jones (1976) to use the available data and models developed for flow in pipes to the case of elliptical cross-section encountered in the PKN geometry.

Abstract: A program of experimental and analytical studies was carried out to investigate the influence of geologic structure on the erodibility of blocky rocks. Until this study, very little data existed regarding hydraulic loads on 3D blocks or regarding the mechanics of erodibility of 3D blocks beyond simple cubes or prisms. Proper understanding of the block scour processes is necessary to ensure the safe and reliable operation of key civil infrastructure, such as dams, bridges and water conveyance tunnels. This paper covers the experimental results and observations from physical hydraulic model experiments to determine 3D block erodibility. The results of the hydraulic modeling are used to develop a block theory framework for analysis of block erodibility. Introduction Scour processes play an important role in the evolution of natural rock landscapes as well as present a potential hazard to civil infrastructure such as dams, spillways, bridges and water conveyance tunnels (Fig. 1). Removal of individual blocks of rock is a primary mechanism by which scour can occur and, accordingly, proper understanding of block scour processes is necessary to ensure the safe and reliable operation of such structures. Evaluation of 3D block erodibility requires detailed knowledge regarding hydraulic loads applied to the block in both a spatial and temporal sense. Data of this nature, however, are extremely limited in current literature. Prior research has focused on simplified rectangular or cubic block shapes (e.g., Reinius, 1986, Bollaert, 2002, Frizell, 2007, Federspiel et al., 2011, Duarte, 2014). However, in general, 3D blocks present a variety of kinematic failure modes, e.g., lifting, sliding, and rotation (Goodman, 1995), and the influence of the kinematic failure mode on block erodibility has not been explored. In that context, a set of flume experiments with different flow rates and block configurations was carried out as part of a comprehensive study of the hydraulics of 3D rock block scour. Detailed measurements and observations of dynamic block response were made to characterize scour mechanics and to better understand the implications for block erosion thresholds.

Abstract: Advancement in laser technology and material engineering have progressively increased the output power and beam quality of solid-state lasers and reduced the price per kilowatt to single digit figures. These developments have the potential to make downhole photonics a reality. To reach its full potential, there are several challenges to overcome for this technology, ranging from material characterization to the accurate modeling of the underlying electromagnetic-thermal-mechanical dynamics among lasers, rocks and the downhole environment. In this paper we present the first results of our investigation, including analysis of the significant variables and their interdependence. Later we explore the advantages and disadvantages of each method, and show some preliminary results of continuous mechanics modeling of laser-rock interaction. Finally, we posit how all these physical phenomena can be coupled in a comprehensive numerical model. Introduction In the first femtoseconds after a laser beam impinges on the surface of a material a series of nanoscopic effects commence and give rise to the absorption, transmission and scattering of the incident rays. If the material is opaque – at the wavelength of the incident radiation – a portion of the absorbed electromagnetic energy transforms into thermal energy. In the next microseconds the material heats up, and a new chain of microscopic events takes place. The material begins to melt, then it disassociates and spalls; finally, it evaporates. Sometimes the thermal energy is large enough to sublimate the material. Through these phase changes the morphology of the material reshapes and its fundamental physical properties vary (see Fig. 1). This ensues in a complex, intense and short-lived dynamic. How to control it remains a hot topic in applied research and engineering with variegated applications; e.g., drilling, perforation, heat treatment, and nano texturing (Gahan et al., 2001; Graves et al., 2002; Batarseh et al., 2004; Lepski and Bruckner, 2009; Batarseh et al., 2012; Assuncao and Williams, 2013; Yang et al., 2015; Sharma et al., 2015; Senesi et al., 2016). In the case of laser-rock thermodynamic interaction, Gahan et al. (2001) first proposed a direct relation between the volume of rock removed and the energy input (termed specific energy), they used it to predict the depth of a laser perforation in different rock samples. Later, Graves et al.(2002) studied the effect of laser-drilling in the porosity and permeability of rocks and observed that rocks permeability and porosity increased proportionally with thermal conductivity. Later, Batarseh et al. (2004) measured the relation between specific energy, time of laser incidence, thermodynamic phase-change and hole geometry; and showed that rocks melted under laser irradiation had larger specific energies than those that were dissociated. Thereafter, numerous studies have been conducted to observe the effects of laser irradiation on saturated and unsaturated rocks (Ahmadi et al., 2011), formation of stress waves inside the rock (Sakakura et al., 2011), and ejection of rock vapor/plasma plumes and rock dissociation (Yan et al., 2013).

Abstract Supercritical carbon dioxide (SC-CO 2 ) jet can break rocks with higher rates and lower threshold pressures than water jet due to the unique fluid properties. The abrasive SC-CO 2 jet technique, which is formed by adding abrasive particles to the SC-CO 2 jet, was proposed and studied. It is expected to achieve better cutting performance than water jet and SC-CO 2 jet. Considering the low fluid viscosity, whether the SC-CO 2 jet has satisfactory particle carrying capacity has to be determined. In effort to investigate the motion of the abrasive particle and the feasibility of abrasive SC-CO 2 jet, numerical simulations on the flow field of the jet under different ambient conditions were carried out, as well as the comparisons with another three particle-laden jets. The results show that SC-CO 2 jet has better particle carrying capacity than the other three fluids. Moreover, temperature rise slightly impairs the capacity meanwhile both the fluid and particle velocities increase, exerting better impinging effect of the abrasive jet. Besides, SC-CO 2 jet can fully carry particles with the diameter in the range of 0.1~2.4mm to form the abrasive SC-CO 2 jet. Particles with larger diameter will obtain higher impinging velocity on the impinged target. 1. INTRODUCTION Inman [1] pointed out that the large-scale, multi-stage hydraulic fracturing, coupled with horizontal drilling, has led to a boom of the development of the shale gas in the US and may help achieve "a thriving US economy" in the future. However, Hughes [2] assumed that it will also bring difficulties like sustainability, high cost, water contamination and other environmental concerns and so forth. Likewise, according to EIA [3], shale gas industries in China in early stages of development are facing both technical and water resource challenges, in addition to the difficult terrain and high overall costs. Considering the huge consumption and exploration of natural gas worldwide, Wang et al [4] thought that it is in urgent need to make progresses in operating fluids. According to Rogala et al [5], emphasis has been placed on the alternatives to hydraulic fracturing or the non-aqueous fracturing technologies due to its technical and environmental advantages. Among the fluids, supercritical carbon dioxide (SC-CO 2 ) is hopeful to replace water as a drilling fluid or the abrasive-laden fluid in hydraulic fracturing in shale gas development.

Abstract Rock mass treatment using fluid injection is common in various industrial applications, including enhanced recovery methods in the oil and gas industry, rock mass pre-conditioning in the mining industry, and heat extraction in geothermal systems. Non-isothermal fluid injection requires consideration of the thermomechanical perturbation as well as hydro-mechanical processes. Thermal effect is rarely included in injection analysis for geothermal application and thermal enhanced oil recovery methods, although with long times their impact becomes of first-order. In this paper, a fully-coupled, hybrid numerical model is implemented to study the effect of cold fluid injection into a conductive fracture under different injection/cooling schemes. The results show that the thermoelastic effect soon overwhelms the hydroelastic effect adjacent to the injection source, whereas far from the injection point, hydroelastic effect dominates because the pressure front always moves faster than the cold front. In addition, the fracture becomes more susceptible to shear failure in the presence of both thermoelastic and hydroelastic induced stresses for the case of cold fluid injection. The magnitude of the changes implies that an appropriate thermo-hydromechanical (THM) model is an essential key to address the physical behavior and potential impairment of fracture conductivity under thermal stimulation. 1. INTRODUCTION Rock mass treatment using fluid injection is common in various industrial applications. Fluid is usually injected under a constant rate or injection pressure to fulfill the intended engineering goal of the stimulation practice. In the oil and gas industry, as an example, large volumes of water are injected at constant pressure during secondary recovery phase for waterflooding the oil-bearing reservoirs with low pressure (with respect to the minimum principal stress) to displace part of the remaining oil toward the production well [1]. In the mining industry, rock mass pre-conditioning using hydraulic fracturing may soon become standard practice for stope mining to facilitate caving propagation and fragmentation management [2]. On-going research interests are pursuing fluid injection as an in-situ stress and rock mass stiffness management tool [2, 3].

Abstract The paper presents results of a micro-mechanical Discrete Element Method (DEM) study of the hydraulic fracture initiation and propagation in Enhanced Geothermal System (EGS) performed using the Particle Flow Code (PFC). Hydraulic fracturing is the main means to stimulate and create flow paths to extract heat in hot dry rocks with insufficient permeability to inject and circulate fluids. Hydro-thermo-mechanical coupled modeling is performed to analyze stress and strain changes on fracturing from a wellbore for improving the understanding of the role of thermal stresses on fracture propagation processes and the resulting fracture geometry. Bonded particle model (BPM) that is used for modeling the mechanical response and fracturing of solids was modified for capturing mechanical effects of temperature difference between rock and fracturing fluid infiltration in the propagating fracture. Heat exchange between fluid and rock and fracture is fully coupled processes. As the fracture propagates from the pressurized borehole, both fluid and rock adjacent to a newly formed fracture change temperature. The results show that thermally induced stresses can significantly change the fracture initiation and the fracture propagation pattern. Thermal stresses cause shallow randomly oriented cracks. The study evaluated fracture geometry and orientation with respect to fracturing fluid temperature, viscosity, density, pressure, rock parameters and in-situ stress difference.

Abstract We conducted experiments in which a high concentration (50% v/v) of granular solids suspended in a non-Newtonian carrier fluid (0.75% guar gum in water) flowed through a parallel-plate fracture. Digital imaging and particle-imagevelocimetry analysis provided a detailed map of velocities within the fracture. Results demonstrate development of a strongly heterogeneous velocity field within the fracture. We observed the highest velocities along the no-flow boundaries of the fracture and the lowest velocities along the centerline of the fracture. Computational fluid dynamics (CFD) simulations using a recently developed model of the rheology of dense suspensions of mono-disperse solids in Newtonian carrier fluids closely reproduced experimental observations of pressure gradient versus flow rate. Results from additional simulations suggest that small (3%) variations in solid volume fraction within the fracture could lead to significant (factor of two) velocity variations within the fracture with negligible changes in observed pressure gradients. The variations in solid volume fraction persist over the length of the fracture, suggesting that such heterogeneities may play a significant role in the transport of dense suspensions. Our work suggests that a simple average conductivity parameter does not adequately represent the flow of high solid content suspensions in a fracture, as the flow develops strong three-dimensional structure even in a uniform rectangular channel.