The Evolution of a Fracture in a Dolomite Sample During Dissolution Induced by a CO2-Saturated Solution Flow at Reservoir Conditions: a Dynamic Synchrotron X-Ray Microtomography Study

Abstract:

Due to their importance as both GCS reservoir rocks and seals, the geochemical behavior of carbonates, such as limestones and dolomites, in acidic environments is relatively well-studied topic. However, the feedbacks between dissolution processes and permeability are not fully understood, particularly for fractured carbonates at in-situ stress conditions. A key open problem is the evolution of fracture aperture (and therefore permeability) for induced fractures in low permeability carbonates. We present the results of a dynamic synchrotron X-ray microtomography (SXR-μCT) experiment investigating changes in fracture aperture in a dolomite sample during dissolution induced by CO₂ saturated water. The dolomite sample used for the experiment was sourced from the Duperow formation of North central Montana, and consisted of a 3/8” cylindrical sample, 7/8” long, with a single vertical fracture. CO₂ bearing solution was flowed through the the sample in a μCT high pressure vessel for ~ one week at a constant flow rate of 5 μl/min under 1400 psi pore pressure and 1700 psi confining pressure. XR tomographic scans were taken at different time steps to follow the evolution of fracture apertures. Results show that the evolution of the fracture is extremely complex: the calcite fraction present in the sample (~5%) dissolves readily showing a prompt retreat in the fragments contacting the fracture surface. In contrast, dolomite zones develop a leached layer ~250 μm thick, rich in micropores along the surfaces of the fracture. With the evolution of the system, a wormholing effect becomes more evident with the development of channels of preferential dissolution along the paths of maximum fluid velocity, where the leached layer becomes less evident and faster bulk dissolution of the dolomite is observed. The main aperture of the fracture remains fairly constant during dissolution; even after an increase in confining pressure (2300 psi), no closure was observed. This highlights how the weakening of the contact points, under these experimental conditions, is not as strong as some models might suggest. Likewise, these observations indicate that reactive fracture models may require inclusion of a porous altered zone in the near-fracture region to adequately represent feedbacks between dissolution and permeability.