Coal reservoir characteristics and fluid transport properties for coalbed methane (CBM) evaluation

Coalbed methane (CBM) is an unconventional resource and as such, requires a fundamental understanding of parameters and processes that control fluid transport. In order to examine the feasibility of CBM in Germany, a reservoir characterization study was conducted. This dissertation represents work carried out in the project “RWTH-EMPG Investigation of Coal Seam Gas: Gas storage and transport processes in coal seams”, which was part of the CBM reservoir characterization study. Priority was placed on investigating controls on permeability and determining end members for permeability, and whether these are above an economic threshold. A set of coal samples from active deep coal mines as well as core samples from explorations wells drilled in Germany were collected for this study and complemented with coal samples from other countries. In particular, meso- and macropores and cleats were examined visually and results were interpreted for their influence on fluid flow. In turn, fluid flow in some of the same samples was determined experimentally and the results were related to production. The insights gained from these three parts of the study -between pores, cleats and permeability- are integrated and their relevance to CBM exploration and development is discussed.Broad-ion beam (BIB) milling combined with scanning electron microscopy (SEM) was applied for the first time to coals to study their meso- and macroporosity. BIB milling prepares a low relief surface (±5 nm), improving recognition of pore structure, which can be indistinguishable from shadows cast by surface topography of non- milled samples. High resolution SEM is used to view pores down to the nanometer scale. Sample with varying degrees of coalification were selected to test the applicability of this new method. Samples were investigated as a function of maceral type and their relationship to bedding. Less than 2 % of all the meso- and macropores studied were found in vitrinites. Pore morphology of inertinite depends on the original maceral, e.g., fusinite yields large, elongated pores and macrinite yields small, round pores. A power law relationship was fit to the pore size distribution. Albeit, micropores and macropores belong to different size distributions. BIB-SEM is a useful tool to study meso- and macropore morphology in coals. However, more maceral types should be studied at different stages of coalification to be able to upscale the results for fluid flow to the heterogeneous matrix composition of coals. The coal cleat system was studied by applying fracture characterization techniques, commonly applied to reservoir rocks. Data on cleat aperture, height and spacing, as well as on cleat type, e.g. face or butt, and their relationship to lithotype, were gathered using a 1D scanline to determine fracture permeability and porosity assuming a cubic relationship. Sixteen different coals were examined, creating a large data set (>8000 individual measurements). Polished sections of the samples were digitally scanned under the microscope. The examined cleat system spanned several orders of magnitude from micrometers to centimeters. Butt cleats are more strongly related to lithotype, and thus bed thickness and mechanical strength of the bed, than face cleats. Cleat aperture and spacing depend on cleat type. A face to butt cleat ratio of 5:1 given an average aperture of 12 µm was found, which is similar to published permeability anisotropies. Calculated cleat porosity from aperture and spacing measured on un-stressed samples is several times greater than cleat porosity calibrated to laboratory permeability under stress. Aperture is predicted to decrease by as much as four times due to an increase in stress from surface to in-situ conditions. Calibrated porosity ranged from 0.01 % to 1.8 % and agrees with published values. This data can be used in reservoir simulation models. To study permeability of low permeable coal and sandstone as a function of stress, a new steady-state method was developed and apparent permeability coefficients are compared with results from two established methods. The new method combines two counter-acting piston pumps (CAPP) at opposite ends of the sample, which to date, has only been applied to shales and limestones. This method was applied to coal and sandstone for the first time. The advantage of this method is that it maintains a constant, low pressure differential between upstream and downstream compartments at high mean pore pressures. The second method, referred to as TAP (“to atmospheric pressure”), is a steady state method with adjustable upstream pressure and downstream pressure being atmospheric pressure. This method has been applied to coal in several previous studies. The third method used in the comparison, uses a “true triaxial stress coal permeameter” (TTSCP) developed at the University of Queensland, Australia. This method allows measurements of fluid transport in three dimensions using cubic coal samples under true triaxial stress conditions. The permeability experiments were performed on six coal samples and one sandstone sample. Laboratory permeability experiments with the TAP and CAPP use 38 mm cylindrical plugs, while the TTSCP uses 80x80x80 mm3 coal blocks. Apparent permeability coefficients as a function of mean pore pressure from the three methods generally resulted in TAP>CAPP and TAP>TTSCP but are all within the same order of magnitude. The TAP method restricts the fluid flow pathways on the downstream side, whereas the CAPP and TTSCP methods keep cleat apertures open. The TAP method is appropriate for samples with a rigid matrix, e.g. sandstone, while for more elastic materials, e.g. coal, CAPP and TTSCP are more suitable to distinguish the effect of differential pressure gradient and change in mean pore pressure. CAPP-derived apparent permeaiblity coefficients are better suited for CBM reservoir modeling since the downstream pressure is seldom atmospheric in a wellbore. Greater permeability from TTSCP than CAPP method is likely due to the larger area for fluid flow. Differences in apparent permeability derived from experiments performed on coal samples in face cleat and butt cleat direction were observed, indicating variable degree of horizontal permeability anisotropy. Low rank coal from Colorado showed greater apparent permeability coefficients in the face cleat direction than in the butt cleat direction, whereas for the anthracite from Germany the opposite trend was observed. This can be related to microscopically observed differences in cleat geometry between the two coals, or can indicate different connectivity of the cleat network. Differences in coal cleat system can be due to different thermal maturity or different tectonic history, or both. Cleat volume compressibility was determined as a function of the change in apparent permeability with a change in effective stress under constant pore pressure. Cleat compressibility is comparable to effective porosity and can be used to describe how sensitive coalbed permeability is to a change in reservoir stress, associated with the production of reservoir fluids. Results of the permeability study can be applied in CBM exploration, development and reservoir simulation. From the laboratory experiments of permeability as a function of stress, the samples from Colorado and the Ruhr basin (Girondelle) are above a given economic permeability threshold. Cleat compressibility is used in reservoir simulators to predict the time to first gas production and knowledge of the permeability anisotropy is important in the planning phase of CBM development and can have an impact on the drilling program, well placement and reservoir stimulation