Combining molecular simulation and experiment to understand the effect of moisture on methane adsorption in kerogens

Abstract

There is a limited understanding of the critical impact moisture has on shale gas resource estimation by affecting gas adsorption and pore structure. Laboratory experiments on dry and 95% relative humidity (R.H.) isolated kerogens are combined with Grand Canonical Monte Carlo (GCMC) and Molecular Dynamic (MD) simulations for kerogen models, including matrix and slits (0.5, 1.0, 1.5, and 2.0 nm) with a range of moisture contents (0–42 wt% on a total organic carbon content (TOC) basis) to better understand how moisture impacts methane adsorption. Higher methane adsorption capacities (Qm) and micropore volumes (Vmicro) are observed for simulated kerogens since all pores in GCMC are accessible. Moisture has a negative effect on Qm, displaying ‘rapid’, ‘gentle’, and ‘slow’ stages with increasing moisture in simulation. Reductions in Qm (61–75%) and Vmicro (88–93%) are obtained for isolated kerogens containing moisture of 38–70 wt% TOC with up to 56% of the moisture in micropores. The same Qm and Vmicro reductions can be reached for the simulated kerogens with moisture contents of 4–24 wt% TOC for matrix and slits. The relative coordination number (Cr) from MD simulation indicates water has a stronger affinity than methane for all functional groups with preferred sorption sites like carboxyl (COOH) under reservoir conditions. The microporosity controls condensed water cluster size. Water adsorbed in ultra-micropores (<0.7 nm) leads to ‘rapid’ reduction, the ‘gentle’ Qm reduction stage arises from water condensing, and filling of remaining pores at the highest moisture is related to the ‘slow’ Qm reduction stage. Therefore, water reduces the methane adsorption capacity of kerogen mainly by occupying and blocking the pore volume rather than competing directly with methane for sorption sites.