A critical component of natural gas in organic-rich shales is adsorbed gas within organic matter. Quantification of adsorbed gas is essential for reliable estimates of gas-in-place in shale reservoirs. However, conventional high-pressure adsorption measurements for coal on the volumetric method are prone to error when applied to characterize sorption kinetics in shale-gas systems due to limited adsorption capacity and finer pores of shale matrix. An innovated laboratory apparatus and measurement procedures have been developed for accurate determination of the relatively small amount of adsorbed gas in the Marcellus shale sample.
The custom-built volumetric apparatus is a differential unit composed of two identical single-sided units (one blank and one adsorption side) connected with a differential pressure transducer. The scale of the differential pressure transducer is ± 50 psi, a hundred-fold smaller than the absolute pressure transducer measuring to 5000 psi, leading to a significant increase in the accuracy of adsorption measurement. Methane adsorption isotherms on Marcellus shale are measured at 303, 313, 323 and 333 K with pressure up to 3000 psi. A fugacity-based Dubinin-Astakhov (D-A) isotherm is implemented to correct for the non-ideality and predict the temperature-dependence of supercritical gas sorption.
The Marcellus shale studied displays generally linear correlations between adsorption capacity and pressure over the range of temperature and pressure investigated, indicating the presence of a solute gas component. It is noted that the condensed phase gas storage exists as the adsorbed gas on shale surface and dissolved gas in kerogen, where the solute gas amount is proportional to the partial pressure of that gas above the solution. To our best understanding, it is the first time to observe the contribution of dissolved gas to total gas storage. With adsorption potential being modeled by a temperature dependence expression, the D-A isotherm can successfully describe supercritical gas sorption for shale at multiple temperatures. Adsorption capacity remarkably decreases with temperature attributed to the isosteric heat of adsorption. Lastly, the wide applicability of the proposed fugacity-based D-A model is also tested for literature adsorption data on Woodford, Barnett, and Devonian shale. Overall, the fugacity-based D-A isotherm provides precise representations of the temperature-dependent gas adsorption on shales investigated in this work. The application of the proposed adsorption model allows predicting adsorption data at multiple temperatures based on the adsorption data collected at a single temperature. This study lays the foundation for accurate evaluation of gas storage in shale.