Russell T. Ewy

Wellbore Stability Predictions Using a Modified Lade Criterion

The paper presents a 3D rock-failure criterion that is a modification of a criterion originally developed by Lade. The criterion proposed has the following desirable qualities : (1) it correctly describes the influence of σ 2 on rock strength and therefore on wellbore stability ; (2) it can be rearranged to give a closed-form solution for critical mud weight for any wellbore orientation ; and (3) only two rock-strength parameters, such as cohesion and friction angle are required.

ISRM Suggested Method for Uniaxial-Strain Compressibility Testing for Reservoir Geomechanics

Please send any written comments on this ISRM Suggested Method to Prof. Resat Ulusay, President of the ISRM Commission on Testing Methods, Hacettepe University, Department of Geological Engineering, 06800 Beytepe, Ankara, Turkey at resat@hacettepe.edu.tr .

Nonlinear Elastic Response of Partially Saturated Gas Shales in Uniaxial Compression

Elastic properties of gas shales are fundamental for the geomechanical characterization of unconventional reservoirs. Due to the presence of microcracks, a nonlinear elastic behaviour with hysteresis is usually exhibited when gas shales are subjected to unloading/reloading paths in laboratory tests for the determination of the elastic moduli. When gas shales are tested in partially saturated conditions with total suction control to reproduce in situ hydraulic conditions, the hysteretic behaviour results to be significantly affected by the wetting or drying processes imposed to change the water saturation of the material; an opening of the loop is observed; and different secant elastic moduli are measured on the unloading and reloading parts. Experimental evidence from a uniaxial compression test performed on a gas shale specimen is presented in this paper, where several unloading–reloading paths are carried out at different total suction values. This study demonstrates that the swelling (or shrinkage) behaviour experienced during total suction variations imposed before performing the axial stress variation influences significantly the sliding cracks mechanism during the unloading phase. In particular, when a wetting process is applied, the nonlinearity of the response decreases significantly. On the other hand, the reloading part exhibits always a nonlinear trend. Although the discrepancy between the unloading and reloading curves of the performed paths, a systematic impact of total suction on the elastic moduli is highlighted. The presented analysis demonstrates that not considering these mechanisms may lead to misleading interpretation of the experimental results.

Shale Swelling/Shrinkage, Suction and Osmosis

Samples were cut from nine different preserved shales, representing in situ states from 11% to 29% porosity and native water contents from 5% to 15%. In an air environment, each shale shows well-defined relationships among suction, water content, bulk volume and saturation. The native state for some shales corresponds to RH<0.8, even though the samples are fully-saturated. The shales will always shrink if placed into RH lower than native state. In contrast, when placed in direct contact with brines, the shales nearly always swell even when the brine water activity is less than the shale activity. The shales are confirmed to be unable to exclude ions; however, the swelling is correlated with the amount of water gain. Brine contact results in water gain that is difficult to explain through osmotic theory. However, some osmotic-type effects are observed.

Gas Shale Water Imbibition Tests with Controlled Suction Technique

Water loss during flowback operations represents one of the main challenges related to the use of hydraulic stimulation to exploit shale gas resources. About 20% of the injected fracturing fluids are usually recovered after stimulation. Fluid imbibition is expected to be one of the main mechanisms responsible for the water uptake of shale gas reservoirs. Imbibition tests are typically performed to analyse this issue. This study presents a new experimental methodology based on the control of total suction to quantify the impact of the swelling response of gas shales on the water uptake during imbibition processes. The obtained results demonstrate that a precise quantification of the gas shale water uptake cannot be performed neglecting the volumetric behaviour and the presence of the mechanical stress during the imbibition process.

Consolidated-Undrained Triaxial Test Results of Opalinus Clay and Comparison with Caprock Shales

Specific equipment and procedures developed for geomechanical testing of hydrocarbon caprocks were adopted to conduct truly undrained triaxial tests with Opalinus Clay. The amount of pore pressure development during consolidation, and the resulting effective stress, is managed by equilibrating the samples in vacuum desiccators of different relative humidities (vapor equilibration technique) prior to assembling into the test apparatus. We present test results of five Opalinus Clay samples covering a laboratory mean effective consolidation stress range from 5 MPa to 50 MPa. A drained consolidation test was first conducted to determine the appropriate strain rate for consolidated-undrained (CU) triaxial testing. The Skempton ‘B’ parameter was quantified prior to the deformation tests and found to be stress dependent. A distinct stress dependency of elastic moduli is also observed, but normalized with the undrained shear strength there is only a relatively small variation. Within the explored stress range the different stress paths to peak indicate a transition from over consolidated to rather normally consolidated state. However, failure is in all cases dilatant, i.e. associated with a drop in pore pressure and strain-softening (more so at low effective stress). Caprock shales of similar porosity to the Opalinus share many similarities in overall behavior, but also exhibit some slight differences.

Shale Capillarity, Osmotic Suction and Permeability, and Solutions to Practical Testing Issues

For typical shales (void ratio 0.15 to 0.42), modal pore throat sizes range from a few nm to a few tens of nm. Unstressed shales, even when fully saturated, have negative pore water pressure (capillary tension). However, the total suction is often greater than this, especially for highly-compacted shales with extremely small pore size. The additional suction is due to effects associated with clay surfaces. Osmotic pressures can be directly measured, and they can easily be several MPa, a combination of solute suction and clay-related effects. The small pore sizes in shales also result in extremely low values of permeability and of consolidation coefficient. All these characteristics directly impact testing protocols. The first step in any test should be to apply sufficient confining stress to raise the pore pressure up to a positive, measured value. Undrained consolidation, combined with undrained triaxial compression and with small sample sizes (and drainage screens when necessary), results in acceptable test durations. A range of effective consolidation stress values is attained by first equilibrating shale samples in varying amounts of suction, to vary the water content. Non-aqueous fluids are required when sampling, to avoid swelling, and are often necessary for pore lines if osmotic pressures are to be avoided.

Investigation of Stress-Induced Borehole Enlargement Mechanisms by a Liquid-Metal Saturation Technique

Stress-induced borehole enlargement was investigated with thick-walled hollow rock cylinders and a liquid-metal sample-preservation technique that permits observation of the microstructure of the broken rock. The observed breakouts are oriented and occur on the two opposite sides of the hole with the highest compressive-stress/compressive-strength ratio. Triangular (dog-eared) breakouts result from progressive spalling of thin rock slabs. The fundamental mechanisms leading to this failure is the growth and interaction of microscopic cracks oriented parallel to the hole wall. Overbalance pressure in the hole changes this failure mechanism and greatly strengthens and stabilizes the hole wall. The apparent rock strength observed next to these small laboratory-scale model holes is higher than expected for field-scale holes.