Peter Ibemesi, Philip Benson

Hydraulic fracture in deep rock masses is used across a variety of disciplines, from unconventional oil and gas to geothermal exploration. The overall efficiency of this process requires not only knowledge of the fracture mechanics of the rocks, but also how the newly generated fractures influence macro-scale pore connectivity. We here use cylindrical samples of Crab Orchard sandstone (90mm length and 36mm diameter), drilled with a central conduit of 9.6mm diameter, to simulate hydraulic fracture. Results show that the anisotropy (mm-scale cross-bedding orientation) affects breakdown pressure, and subsequent fluid flow. In experiments with samples cored parallel to bedding, breakdown pressures of 11.3MPa, 27.7MPa and 40.5MPa are recorded at initial confining pressures at injection of 5MPa, 11MPa and 16MPa respectively. An increase in confining pressure (from 5 MPa to 26 MPa) after the initial fracture event results in a flow rate decrease from 1.67 mL/min to 0.043 mL/min. For samples cored perpendicular to bedding, breakdown pressure of 15.4MPa, 27.4MPa and 34.2MPa were recorded at initial confining pressure at injection of 5MPa, 11MPa and 16MPa respectively. As confining pressure increases from 5 MPa to 26 MPa, flow rate through the newly generated fracture decreases from 0.043 mL/min to 0.0073 mL/min. We note that fluid flow recovers during a confining pressure “re-set” and that the ability of flow to recover is strongly dependent on sample anisotropy and initial confining pressure at injection.