Applied stress reduces the CO2 sorption capacity of coal

Though the adsorption of CO2 by coal has been extensively studied in experiments, few systematic studies have been done on the effects of the stress state within the coal on CO2 sorption. To investigate whether or not the CO2 sorption capacity of coal is influenced by the application of an effective stress to its solid framework, we performed five experiments on pre-compacted, porous aggregates of crushed coal matrix material, at a fixed temperature of 40 °C and at varying CO2 pore pressures between 10 and 20 MPa. The samples were loaded in a stepwise manner up to an applied effective stress of 35 MPa in excess of the CO2 pressure in the pores, while the volume expelled from the samples was monitored using a syringe pump. One control experiment was conducted using non-adsorbing helium at 15 MPa. Our data show that CO2 is reversibly expelled from our samples by the application of an effective stress, and that this is accompanied by reversible compaction, with both effects being partly time-dependent. Analysis of these data, assuming the maximum versus the minimum possible reduction of pore volume occupied by free CO2 during compaction of the aggregates, demonstrates that the CO2 sorption capacity of our samples was reduced by an amount in the range between 0.0014 and 0.014 mmol gcoal− 1 MPa− 1. Under the temperature and CO2 pressure conditions investigated, this is equivalent to a reduction of about 5–50% of the initial sorption capacity at zero applied effective stress. The observed desorption was accompanied by a reduction in apparent axial stiffness of the samples by about ⅓ from 1.52 to 0.95 GPa. The microphysical mechanism responsible for desorption was investigated by deriving a simplified (hydrostatic) thermodynamic model representing the coal as an adsorbent with a finite density of adsorption sites, interacting with CO2 (the adsorbate). Our model is based on the principle that the free energy change associated with adsorption of a single CO2 molecule in stressed coal, is increased, relative to the unstressed state, by an amount reflecting the extra mechanical work done due to swelling of the coal per adsorbed molecule. This model predicts a reduction in CO2 sorption capacity and in apparent axial stiffness of the samples, of similar magnitude to those observed in our experiments, suggesting that the basic physics of the effect of stress on sorption are successfully captured by the model. In coal basins considered for ECBM production, the overburden stress increases with depth, and self-stressing of coal seams will occur due to swelling after CO2 injection. As current reservoir models do not consider the effects of stress on adsorption, our observations and modelling results may have important implications for predictions of in situ CO2 storage capacity