-Leakage of CO2 into the atmosphere is the most crucial concern for geological storage of anthropogenic CO2. Leakage routes could develop through existing wells and pipelines, but also by natural migration of CO2 rich pore-fluid through the caprock and in fault zones. Therefore, a thorough geological characterization of the prospective formation identifying seal capacity, integrity and possible migration pathways must be performed prior to injection of CO2. A caprock is a low permeable confining layer trapping CO2 stored in a reservoir rock. Although the caprock acts as a seal, its lower boundary will be in contact with CO2 saturated pore water or even pure CO2. Chemical interaction between the pore fluid and the caprock may change its material properties. The aim of this study is to increase our understanding of the interaction between CO2 and caprock, focusing on the microstructural properties of the rock. A flow through cell is used to flood a shale core (40 mm long and 38 mm diameter) with supercritical CO2 at a temperature of 35 ∘C and at pressures above 7.5 MPa. A pressure gradient is applied across the sample to obtain a breakthrough of CO2 in the core. During flooding both axial and radial strain of the core are measured together with the acoustic velocities in the axial direction. The measurements are performed both for the brine saturated core and at various stages during flooding with CO2. The experimental results suggest flow of CO2 along defined pathways within the shale. These pathways are likely to be controlled by the increasing pore pressure at the bottom of the sample which allows reopening of the cracks in the lower part of the sample, allowing CO2 to enter the shale. Flow of CO2 into the cracks in the lower part of the sample is followed by percolation of CO2 in the upper part of the sample where the effective pressure is higher and crack reopening is less likely to occur

Supercritical CO ₂ breakthrough and flow mechanisms in shale have been investigated in laboratory experiments using a high pressure flow cell and cylindrical samples of shale from the Draupne formation in the North Sea. The main objective is to study the basic mechanisms involved in the breakthrough process and define the controlling parameters for supercritical CO ₂ flow in a low permeable shale. Experimental testing provides new insight into the CO ₂ breakthrough process through simultaneous measurements of deformation and ultrasonic velocities in the sample. A marked sample dilation associated with the CO ₂ breakthrough is identified accompanied with a pronounced drop in ultrasonic velocities. X-ray images of the sample using a high resolution 3D computer tomography (CT) scanner provide information on macroscopic fracture distribution inside the sample before and after testing. The CO ₂ breakthrough pressure for the Draupne material seems to depend on confining pressure and effective pressure rather than pore pressure difference across the sample. After breakthrough the effective CO ₂ permeability was found to follow a simple model for permeability in fractured rock. The drop in ultrasonic velocity was associated with mechanical changes and possible micro fracturing inside the sample. Based on our observations we conclude that pressure-induced opening of micro-fractures during the breakthrough process is an important mechanism for flow in addition to capillary displacement. Our findings may have important consequences for later testing and estimation of CO ₂ breakthrough pressure and flow in shale.

-Laboratory core flooding experiment was run to investigate the joint use of electrical resistivity, ultrasonic velocities and 3D images of fluid distribution to improve current understanding of CO2 and brine behaviour during drainage and imbibition in reservoir rocks. The test specimen was cylindrical Rothbach sandstone measuring 100 mm in length and 38 mm in diameter, with a porosity of 23% and an average permeability of 400 mD. The brine saturated specimen was drained by injecting CO2 and later imbibed with brine while monitoring changes in resistivity and ultrasonic velocity measurements. Actual fluid saturation level and distribution have been mapped simultaneously using X-ray CT scan. CO2 saturation calculations based on CT values showed a steep saturation gradient at start of drainage and the gradient flattened with more CO2 injected into the sample. The ultrasonic compressional velocity (V p) measurements depicted a continuous decrease with increase in CO2 saturation while the resistivity of the sample increased proportionally with increase in CO2 saturation. A CO2 saturation of 53% was achieved at the end of drainage after injection of 20 pore volume (PV) CO2. This resulted in a decrease of V pby 7.2% while the amplitude decreased by as much as 48%. At the end of drainage, the resistivity of the sample increased to 13.9 Ω-m from full brine saturated value of 3.2 Ω-m. During imbibition, the sample was re-saturated close to 100% after 10 PV brine injection. Change in V p due to CO2 saturation level variation was relatively consistent during drainage and imbibition (no hysteresis) where as the resistivity of the sample was more affected by the flooding history resulting in hysteretic variation in resistivity. Resistivity index (RI) values did not show a consistent pattern and a single Archie’s saturation exponent (n) could not be assigned