Reservoir capacity estimates in shale plays based on experimental adsorption data

Abstract

Fine-grained sedimentary rocks, such as mudrocks, are characterized by a complex porous framework containing pores in the nanometer range that can store a significant amount of natural gas (or any other fluids) through adsorption processes. Unfortunately, although the adsorbed gas can take up to a major fraction of the total gas-in-place in these reservoirs, the ability to produce it is limited, and the current technology focuses primarily on the free gas in the fractures. A better understanding and quantification of adsorption/desorption mechanisms in these rocks is therefore required, in order to allow for a more efficient and sustainable use of these resources. Additionally, while water is still predominantly used to fracture the rock, other fluids, such as supercritical CO2 are being considered; here, the idea is to reproduce a similar strategy as for the enhanced recovery of methane in deep coal seams (ECBM). Also in this case, the feasibility of CO2 injection and storage in hydrocarbon shale reservoirs requires a thorough understanding of the rock behavior when exposed to CO2, thus including its adsorption characteristics. Shale reservoirs are widely distributed in the U.S. with existing infrastructures used for gas and oil production operation; should CO2 injection in such reservoirs prove feasible, the capacity of geologic formations for CO2 storage will increase significantly. The main objectives of this Master's Thesis are as follows: (1) to identify the main controls on gas adsorption in mudrocks (TOC, thermal maturity, clay content, etc.); (2) to create a library of adsorption data measured on shale samples at relevant conditions and to use them for estimating GIP and gas storage in shale reservoirs; (3) to build an experimental apparatus to measure adsorption properties of supercritical fluids (such as CO2 or CH4) in microporous materials; (4) to measure adsorption isotherms on microporous samples at various temperatures and pressures. The main outcomes of this Master's Thesis are summarized as follows. A review of the literature has been carried out to create a library of methane and CO2 adsorption isotherms on shale samples from various formations worldwide. Large discrepancies have been found between estimates of the adsorbed gas density from different measurement techniques using representative fluids (such as CH4 and CO2) at elevated pressures, and the adsorbed density can range anywhere between the liquid and the solid state of the adsorbate. Whether these discrepancies are associated with the inherent heterogeneity of mudrocks and/or with poor data quality requires more experiments under well-controlled conditions. Nevertheless, it has been found in this study that methane GIP estimates can vary between 10-45% and 10-30%, respectively, depending on whether the free or the total amount of gas is considered. Accordingly, CO2 storage estimates range between 30-90% and 15-50%, due to the larger adsorption capacity and gas density at similar pressure and temperature conditions. A manometric system has been designed and built that allows measuring the adsorption of supercritical fluids in microporous materials. Preliminary adsorption tests have been performed using a microporous 13X zeolite and CO2 as an adsorbing gas at a temperature of 25oC and 35oC and at pressures up to 500 psi. Under these conditions, adsorption is quantified with a precision of +/- 3%. However, relative differences up to 15-20% have been observed with respect to data published in the literature on the same adsorbent and at similar experimental conditions. While it cannot be fully explained with uncertainty analysis, this discrepancy can be reduced by improving experiment practice, thus including the application of a higher adsorbent's regeneration temperature, of longer equilibrium times and of a careful flushing of the system between the various experimental steps. Based on the results on 13X zeolite, virtual tests have been conducted to predict the performance of the manometric system to measure adsorption on less adsorbing materials, such as mudrocks. The results show that uncertainties in the estimated adsorbed amount are much more significant in shale material and they increase with increasing pressure. In fact, relative uncertainties in the adsorbed amount can reach up to 80 and 200% at 500 and 1600 psia, respectively. The latter can be reduced (i) by increasing the mass of adsorbent material (15.2% and 42.3% when the mass of adsorbent is doubled as compared to the experiment with 13X zeolite) and/or (ii) by increasing the precision of the pressure transducers (uncertainty is further reduced to 3% and 8.4% from case (i) when the transducers with 0.05% accuracy are used. These experiments are justified by the need of extending the current data set on gas adsorption of mudrocks, thus enabling a more reliable estimate on the available gas reserves in shale reservoir and the potential of carbon dioxide storage.