Search

Geological CO2 sequestration is one of the most effective methods to counter global climate change. Coal matrix shrinkage, swelling, gas diffusion and permeability are the key phenomena associated with geological CO2 disposal in coal. In this study, permeability experiments were approached using the supercritical and liquid phases of CO2 (less understood; most likely insitu phases) for naturally fractured bituminous coal. Experiments were performed under triaxial conditions using four sets of various confinement conditions corresponding to variable depths. Injection pressure was varied gradually and the permeability changes were calculated using the Darcy's equation for subcritical and supercritical CO2 exclusively. Changes in CO2 phases were obtained by changing the system temperature from 26 degrees C for liquid CO2 to 34 degrees C for supercritical CO2. N-2 was alternatively injected at the start and between the injections of CO2 to analyze the changes in the permeability from a relatively very less sorptive medium's perspective. It was observed that the supercritical CO2 flow reduced the permeability significantly and this behavior was greatly attributed to the highly viscous nature of the supercritical phase and the high volumetric deformation or the swelling of coal under supercritical CO2 as compared to liquid CO2. The injection pressures were observed to reduce the effective stress behavior, which in turn pushed the permeability evolution at each confinement to a positive trend. However, the permeability of CO2 reduced exponentially with increasing effective stresses. Two different empirical equations were proposed for permeability of both the phases of CO2 with effective stresses. (C) 2016 Elsevier Ltd. All rights reserved.

Building upon the previous modelling work on coalbed methane reservoirs by the authors, recently reported experimental horizontal stress data for two coal samples tested under uniaxial strain conditions have been analysed. As well as CH₄, non-adsorbing Helium has also been used in the experiments. Deviation between the model and the experimental CH₄ stress data at low pore pressures (< 3 MPa) are observed, which, as in the previous study, may be attributed to the onset of shear failure experienced by the coal samples. In addition, CO₂ was also used on one of the samples after the CH₄ experiment. The post-failure CO₂ stress data could be matched using a combination of a reduced Young's Modulus and an increased Poisson's ratio. The results of this study have provided further evidence for the validity of the equation for computing the changes in the effective horizontal stress in producing coalbed reservoirs, which forms an integral part of the Shi and Durucan permeability model.