Capillary trapping mechanism in strongly water wet systems: Comparison between Experiment and Percolation theory

Abstract:

To understand capillary trapping mechanism, we conduct a real Monte-Carlo experiment by using packed glass beads with nearly the same pore size distribution, but different stochastic realizations. We study gas phase trapping during imbibition for capillary number from 2×10^-7 to 10^-6 by X-ray computer tomography (μ-CT) and compare the experimental results with predictions from percolation theory. We found excellent agreement. Percolation theory could explain (i) that the capillary desaturation curves are not dependent on flow rate, (ii) the linear dependence of the total gas surface on gas saturation that is a direct consequence of the linear relationship between cluster surface and cluster volume, which is a prediction from percolation theory for large finite clusters, (iii) the power-like cluster size distribution with an exponent τ_exp = 2.15 that only deviates by 2% from the theoretical one (τ_theor = 2.19), and (iv) that the maximal z-extension of trapped large gas cluster is described by the cut-off correlation length ξ_B (B – bond number).

In order to support the findings from μ-CT-experiments and to study the dynamics of capillary trapping, we conduct visualization experiments using monolayer- and microstructure-models.

The Figure shows the residual trapped air (red colored) after water imbibition: left: 2D-cut through a 3D-reconstructed image, right: 3D-slice of a 3D-reconstructed image.