Abstract

Chamber-and-throat pore networks etched in glass (or plastic) are often used as model porous media to study the pore-scale mechanisms and their cooperative effects on macroscopic transport coefficients of important multiphase processes, such as hydrocarbon recovery from rock reservoirs, pollution of soils and aquifers by liquid organic wastes, etc. Such models have special built-in characteristics, notably, small pore depth (compared to the pore widths and diameters), equivalent capillary diameter that is nearly equal to the pore depth and has small variation and small chamber to throat equivalent capillary diameter ratio. Given that all these features strongly affect the various multiphase transport processes that are studied experimentally in model porous media, it is necessary to take them into account quantitatively. The present work develops a 2-D network mercury intrusion–retraction simulator that is adapted to the specific geometrical and topological characteristics of model pore networks. Based on experimental observations, the mercury meniscus motion in chambers and throats during intrusion and the mercury disconnection events during retraction are analyzed at pore-scale. Mercury intrusion or retraction in chamber-and-throat networks is simulated as a sequence of flow events occurring at progressively increasing or decreasing external pressures, respectively. It is found that the location and the degree of hysteresis of mercury intrusion-retraction curves are determined primarily by the depth of pores and the values of contact angles. The pressure of residual air may influence the high pressure region of intrusion curve and the low pressure region of retraction curve. Mercury retraction from the pore network is carried out through cluster growth, over a narrow pressure region and the residual mercury saturation is low because of the strong dependence of capillary pressures of retraction from pore chambers on the local fluid topology. The simulator predicts the experimental capillary pressure curves of two glass models satisfactorily; furthermore, the simulated patterns of mercury intrusion–retraction are in agreement with corresponding experimental ones. The experimentally validated simulator provides a reliable tool for the thorough characterization of the structure and the prediction of the capillary properties of model porous media as well as of real porous media with similar pore scale characteristics.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call