Monte Carlo simulations of phase distribution in porous materials

Abstract

The interplay and absorption of gases and fluids within a pore space is a fundamental phenomenon of natural systems. In particular, the equilibrium phase distribution of these components within the porous material affect many of the constitutive properties of these materials. These effects can be far ranging from stability of hill slopes resulting from partially saturated soil to the recovery of gas and oil within sediments for energy applications. Therefore a critical understanding and investigation of the equilibrium phase distribution within porous materials is desired. To address this, we have developed a robust Monte Carlo simulation methodology. The methodology uses a lattice gas with nearest neighbor interactions to model the equilibrium phase distribution of the constitutive materials. A primary goal in developing this methodology was to allow direct input of experimental pore space information and to allow prediction of properties for specific porous systems. Input variables such as the pore geometry and interfacial free energy values are thus all calculated directly from experimental values. Systems of water, air, and solid at both constant volume (canonical) and constant humidity (grand canonical) simulation conditions gave results that were in excellent agreement with experiment. Specifically, quantitative agreement was achieved for the soil-water retention curve in both canonical and grand canonical systems and equilibrium phase distribution between simulation and experimental imaging. The methodology was also applied to systems of gas hydrate and oil recovery from within sediments.