Abstract Sorption, strain, and flow-related laboratory experiments combined with numerical modeling have been conducted with CO 2 and other gases, including N 2 , CH 4 , H 2 , H 2 S, and SO 2 on a variety of coal cores and coal powders to investigate the interplay of the parameters controlling storage, migration, and permeability changes during sequestration. The experiments include sorption isotherms, volumetric swelling or shrinkage of coal matrix during sorption (with a variety of gases on solid coal cores), N 2 flow-through experiments on CH 4 -saturated coal cores, and N 2 effects on permeability. The order of adsorption capacity for a given coal in ascending order was H 2 < N 2 < CH 4 < CO 2 < H 2 S < SO 2 . The ratio of adsorptive capacity of the gases to various coals is rank dependent, which our experiments show is mainly attributable to declining moisture content with increasing coal rank. In low volatile bituminous rank coals, the ratio of CO 2 to CH 4 adsorption capacity at a given pressure is about 2:1, but this is about 10:1 in subbituminous coals. Moisture content in the coal reduces the adsorption capacity of CH 4 whereas increased adsorption capacity was observed with CO 2 , H2S, and SO2 with increasing moisture content. Because these gases have high Henry’s solubility coefficient, moisture from the coal micropore surface is stripped off to react with gases making moisture-occupied sorption sites available for the gases to adsorb resulting high adsorption capacity with high-moisture coals. Sorption-related strain experiments with N 2 , CH 4 , CO 2 , and H2S show that adsorption of gases on coal causes swelling of the coal matrix, which is directly proportional to the amount of gas adsorbed onto the coal and hence increases with rank. The average volumetric strain of the samples tested in decreasing order is H 2 S (2.5 × 10 –3 g/cm 3 )> CO 2 (9.9 × 10 –4 g/cm 3 )>CH 4 (6.9 × 10 –4 g/cm 3 )>N 2 (3.1 × 10 –4 g/cm 3 ). Adsorption of CO 2 relative to CH 4 causes a relatively higher volumetric strain of the coal matrix and in turn reduces cleat permeability causing significant reduction in the sequestration capacity into coalbeds. Injection of N 2 into coalbed significantly improves the permeability while displacing the CH 4 because of its lower adsorption and associated swelling. Our experiments with associated analytical and numerical modeling using real data clearly indicate that sequestering pure CO 2 into most coal seams results in volumetric strain and associated loss of permeability that quickly inhibit further or significant sequestration. Hence, it is very unlikely that in-situ sequestration of significant amounts of pure CO 2 will be possible in any but the most permeable coals such as those of the Powder River Basin. However, mixing of N 2 with CO 2 significantly enhances the sequestration potential into coalbeds. Based on our results, a new numerical model was developed, which takes into consideration the shrinkage coefficients derived from experimental results with various gases coupled with mechanical properties of rocks, which closely predict the behavior of CO 2 sequestration in coalbeds. The N 2 flow-through experiments on CH 4 -saturated coal cores confirm the modeling results that N 2 displaces the methane while inhibiting the permeability reduction because of its low sorption property. However, this process requires a minimum permeability to start with and has to be coupled with the drawdown of CH 4 ; otherwise N 2 sorbs into coalbeds because of increased pressure in the overall system without an associated decrease in the partial desorption of CH 4 pressure. Based on our experimental and modeling experience, analytical and numerical solutions provide a good approximation of the behavior of the multicomponent, multiphase flow of gases in coalbeds. However, much more work is required in understanding the sorption behavior of multicomponent gases and their effects on volumetric strain vis-à-vis their sensitivity to permeability on a variety of coals.

The Upper Jurassic-Lower Cretaceous Mist Mountain Formation in the southeastern Canadian Cordillera is a nonmarine succession up to 670 m thick that includes as many as 15 major seams of high volatile bituminous to semi-anthracite coal. Coals at the base of the formation were deposited in coastal and delta plain environments, whereas those of the upper part are interpreted as upper delta plain and alluvial plain deposits. The coal seams are thicker, more abundant, and laterally less continuous in the upper part of the formation. The geometry of the coal seams is influenced by the presence of adjacent channels that have locally thinned or washed out some seams. The effect of differential compaction on coal seam geometry is variable; some seams thin over paleo-channels, whereas others are thicker and/or contain fewer partings. The ash content of most coals shows no predictable lateral or vertical variation that can be related to the overall sedimentology, nor is there a correlation between seam thickness and ash content. The sulfur content of all seams is low (<1 percent), suggesting the absence of marine influence during peat accumulation. There is a general increase in vitrinite and a decrease in inertinite and semi-fusinite from the base to the top of the formation, which may reflect a greater contribution of herbs to coals formed from coastal marsh-swamp complexes at the base. Variations in roof conditions in underground mines are related to the structural fabric of the coal measures, which in turn reflects the kinematics and dynamics of tectonic deformation and roof rock lithology. In the Vicary Creek Mine, the roof rock comprises two lithofacies: a thin-bedded, very fine grained, carbonaceous sandstone lithofacies interpreted as distal crevasse splay deposits, and a thick-bedded sandstone lithofacies interpreted as proximal splay deposits. The thin-bedded lithofacies includes carbonaceous partings that were preferred horizons for intrastratal slip along which cohesion of the roof rock has been lost. The thick-bedded sandstone lithofacies is well jointed, leading to a blocky roof rock that localized intrastratal slip within the underlying coal seam. In the Balmer North and Five Panel Mines, the roof rock is composed of carbonaceous siltstone and very fine grained sandstone interpreted as crevasse splay and overbank deposits. During flexural-slip folding, slip was localized along carbonaceous partings that have destroyed the cohesion between successive beds in the roof rock. The intersections of slickensided bedding surfaces and major shear and extension fracture systems have resulted in unstable roof rock, particularly in rooms and roadways developed parallel to their intersection.