Examining Supercritical Methane Adsorption on Nanoporous Shales Using Constant and Varying Adsorbed Phase Density Approaches

Abstract

Supercritical adsorption studies are crucial, as adsorbed methane has a significant contribution to the total gas-in-place along with the free gas-in-shale reservoirs. Due to the low specific adsorption of shales, Gibbs excess isotherms that decrease after a certain pressure are being reported. Corrections to the excess adsorption amount can be made using models that predict absolute adsorption through empirically estimated adsorbed phase density. However, due to the inherent heterogeneity of shale and uncertainties in the adsorbed phase density estimated from different models, the quantitative characterization of adsorption is becoming increasingly challenging. The assumption of constant adsorbed phase density is being challenged with varying adsorbed phase density approaches, which are theoretically correct; however, their applicability has not been analyzed distinctively. In this work, supercritical adsorption models like supercritical BET (SBET), supercritical Dubinin–Radushkevich (SDR), Ono–Kondo mono- and multilayer, fixed volume Ono–Kondo (FV-OK1), and variable density Langmuir (VD-Langmuir) are revisited, analyzed, and compared for their applicability in shale reservoirs. For this, methane adsorption isotherms at two different temperatures (45 and 90 °C) are generated for four samples with varying geochemical and petrophysical properties selected from the Cambay and Krishna–Godavari (KG) basins of India. Through this analysis, it has been found that fixed-density approaches better explain the isotherms in comparison to varying density approaches. The average adsorbed phase density estimated using SBET and SDR models was closer to 0.2 g/cm3, which is very low in comparison to the theoretical limit. The SBET and Ono–Kondo theories have displayed that the probability of adsorption after the third layer is low. The VD-Langmuir and FV-OK1 could fit perfectly to highly specific adsorbing solids; however, they failed to explain adsorption in low-adsorbing solids. The mathematical drawbacks of variable density approaches have not been analyzed previously and are explained in this work.