Application of contrast techniques in X-ray microtomography to characterisation of rocks and hydrocarbon recovery

Abstract

X-ray microtomography (micro-CT) is the principal imaging tool for digital analysis of rock samples and provides the basis for simulation of rock properties and transport processes within the pore space. The accuracy of flow simulation predictions can be improved by extracting more input information from rock tomograms and by direct imaging of flowexperiments at their end states or during their course. These advances require development of techniques to selectively enhance the X-ray attenuation contrast of the features of interest, which may be rock surfaces or fluids in their pores. The ability to spatially register the 3D tomograms before and after an experiment then allows the subtle changes highlighted by contrast enhancement to be isolated and analysed. This thesis deals with three such contrast enhancement strategies and their applications to characterization of flow properties relevant to hydrocarbon recovery from reservoir rock. The first part of the thesis is directed at furthering understanding of the pore-scale mechanisms by which waterflooding using low salinity brine can improve oil recovery. A clay-containing outcrop sandstone was prepared in a mixed-wet state and micro-CT scanned after spontaneous imbibition of high salinity brine followed by low salinity brine. In this case the contrast between residual crude oil and brine in the pore space was achieved by moderately doping the brine by ion exchange of chloride for iodide. Oil saturation decreased by 10% due to imbibition of low salinity brine, and image analysis of the sequence of three registered tomograms (in the dry state and after the two imbibition steps) was used to characterize the pore-scale changes. All image metrics, including pore oil saturation, oil connectivity, oil-rock and oil-brine interfacial areas and oil-brine meniscus curvature confirmed that the additional recovery on exposure to low salinity brine was driven by a change in wettability state towards more water-wetting. Two techniques were then used to identify the minerals most responsible for this wettability change. One approach attempted to segment mineral classes within the rock phase guided by comparison of attenuation to registered mineral maps from SEM-EDS. However, this lacked the sensitivity to make statistically reliable conclusions. The second approach imaged the drained state with the irreducible brine strongly highlighted by sodium iodide to map the microporosity and thus the clays hosting it. This approach gave sufficient surface sensitivity to reveal that oil preferentially released from surfaces of clays rather than grains in the low salinity brine. The second part of the thesis focuses on improving the ability of micro-CT to extract finescale information from rock pores. The current resolution limit of micro-CT is around 1 pm, however many pores (micropores) in reservoir rocks lie below this size. It is well understood that registration of the tomograms of a rock before and after saturation with an X-ray dense liquid provides on subtraction a 3D map of porosity within sub-resolution pores. The aim of the second part of the thesis was to complement this pore volume map with a corresponding map of sub-resolution pore surface area. In this case iodine adsorption was used as the X-ray contrast technique, such that subtraction of the registered tomograms before and after adsorption isolated this local surface area contribution. A set of 10 rocks was prepared and analysed in this way using micro-CT supported by spectrophotometry. It was shown that the adsorbed iodine layer was so thin (below 1 nm) that only rocks of high internal surface area, such as from tight gas reservoirs, were well suited to this method. One complication was that carbonate rocks adsorbed less iodine than sandstones, although pre-treatment of carbonates by adsorption of asphaltenes from crude oil could partly reduce this difference. Application of the iodine adsorption method to a shale sample exposed the further difficulty of decoupling the adsorption contribution to internal surface area from bulk uptake of iodine within the solid organic matter in shales. The third part of the thesis targets further development and utilisation of contrast enhancement techniques tailored to the unique challenges of shales. Upscaling of simulated matrix transport properties from resolved nano-pore networks in tiny FIBSEM cubes to representative elementary volumes and core plugs requires accurate micro-CT images of pore space - hosted by minerals and by organic matter - over scales from microns to centimetres. 3D porosity mapping by differencing of registered tomograms before and after saturation was adapted to this purpose by using a very dense liquid, diiodomethane, to provide sufficient contrast in these very tight, low porosity rocks and to wet organic matter. A second tomogram difference was generated after selective staining of the organic matter by iodine, to provide a 3D map of local organic content and thus distinguish its internal porosity from that hosted by minerals. Further, bottom-up approaches to transport simulation within pores can be tested against top-down direct imaging of laboratory-prepared states during or after transport processes. In particular, diffusion is thought to be a key transport mechanism for recovering gas from the matrix of fractured shale reservoirs. To this end, the diiodomethanesaturated shale was used as the initial state for dynamic micro-CT imaging of diffusion of a miscible, X-ray transparent second liquid into the sample. Alternatively, the end state after drainage of diiodomethane in air by centrifugation was scanned to quantify wetting-phase saturation versus capillary pressure at each point in the sample, yielding a 3D map of pore throat diameter. Workflows for integration of these tomographic maps with high-resolution SEM images and SEM-EDS mineral maps of sections and FIBSEM cubes were applied to shales from three formations. Show more