Experimental Study of Confinement Effect on Hydrocarbon Phase Behavior in Nano-Scale Porous Media Using Differential Scanning Calorimetry

Abstract

Abstract Phase behavior in shale remains a challenging problem in petroleum industry due to many complexities. One complexity arises from strong surface-fluid interactions in shale nano-scale pores. These interactions can lead to a heterogeneous distribution of molecules, which conventional bulk-phase thermodynamics fails to describe. Phase behavior in shale is altered from that characterized in PVT cells. The majority of current models are based on bulk-phase thermodynamics and efforts have been made using molecular simulation to gain insight into the nano-structure of confined fluids. However, to our best knowledge, the experimental data for hydrocarbon phase behavior in shale systems is severely absent. In this work, we investigated the phase change in nano-scale capillaries using experiments. The controlled pore glasses (CPGs) were applied to model the nano-porous structure of shale reservoirs. CPGs (pore diameters 4.3 and 38.1 nm) infiltrated with hydrocarbons (octane, decane, and the binary mixture) are subject to differential scanning calorimetric (DSC) analysis. It's observed that the bubble point is affected by pore size dramatically: at 38.1 nm the confinement effect is insignificant, but at 4.3 nm two distinct bubble points appear with deviations as great as ±15 K relative to the bulk, suggesting two populations of evaporating fluid. Based on experiments and simulations, a two-state model for the nanoconfined hydrocarbons is proposed. The bubble point is modeled using Peng-Robinson equation of state (PR-EOS) with the capillary pressure considered. The flash calculation is based on isofugacity and an interfacial tension model is accommodated. The modeling shows a general trend of increasing bubble point temperature with decreasing pore diameter, inconsistent with the experimental results. Besides, the “dual bubble points” behavior observed at 4.3 nm is not predicted by the model. This indicates the incapabilty of the bulk-phase thermodynamics in describing the behavior of nanoconfined fluids and the needs for molecule-scale simulation.