Multiscale characterization of three-dimensional pore structures in a shale gas reservoir: A case study of the Longmaxi shale in Sichuan basin, China

Abstract A typical Longmaxi (LMX) shale sample from Sichuan basin, China, is investigated in three dimensional through multiscale tomography methods spanning pixel sizes from 3.25 μm to 10 nm, combining X-ray computed tomography and focused ion beam-scanning electron microscopy imaging techniques. Pores, organic matter (OM) and pyrite were segmented and reconstructed at multiple scales and the representative elementary volume (REV) of each phase at different scales based on the convergence of error bars in the volume fraction are determined. Results show that REVs at each scale varied between different phases. The determination of REV defined the applicable scale of different methods as well as the appropriate scales of specific structures. Porosity of LMX shale determined at different scales shows significant difference at two orders of magnitude. This difference likely was caused by structural heterogeneity together with the limitations of methods. This study confirms that REV is a feasible and powerful approach in modeling shale using multiscale 3D imaging. The application of multiscale 3D imaging techniques to worldwide shale gas reservoirs could allow for significant insights into the assessment of migration mechanisms within low-permeability systems.