A Multimodal 3D Imaging Study of Natural Gas Flow in Tight Sands

Abstract

Abstract Pore scale phenomena, especially in multi-phase flow, have a strong impact on the mechanisms and efficiency of gas recovery. We studied the impact of the pore space geometry on single- and two-phase flow properties of tight-sands. X-ray computed tomography (CT) and scanning-electron microscopy (SEM), along with other imaging tools provide insights into pore structure at multiple scales. Segmented micron-scale resolution CT 3D images were used as input data for simulations based on the Maximal Inscribes Spheres (MIS) method. Finite-difference flow simulations of creeping flow on MIS-evaluated equilibrium fluid distribution predict of the relative permeability curves as functions of saturation. A pore-scale model of two-phase gas flow accounts for condensate dropout based on thermodynamic equilibrium. The pore-scale analysis results are applied in optimal-control model of wet gas flow with liquid skin. This study leads to the following conclusions: (a) the estimated minimal wetting fluid threshold saturation, at which the gas phase is disconnected, is much lower than that for conventional sandstones; (b) pore-scale flow simulations suggest that in imbibition, the gas and brine (or other wetting fluid) can block the other phase's flow ("Permeability Jail"), where the blockage of gas is of capillary nature, and the blockage of liquid phase is dynamic; (c) pore-scale simulations of drainage do not show mutual blockage of the fluids; (d) the rate of the local condensate dropout is proportional to the dot product of the local flow velocity and pressure gradient; (e) upscaling of conclusion (d) to Darcy scale leads to condensate dropout rate proportional to the pressure gradient; (f) the optimal-control model suggests that gas well production-rate control may have potential for considerably increasing the total gas recovery.