Alternative methods for geomechanical characterisation of sub-bituminous coal, Surat Basin, Australia

Abstract

This thesis examines alternative methods for testing the geomechanical behaviour of subbituminous coal that is a target for coal seam gas extraction as well as mining in the Surat Clarence-Moreton basins in eastern Australia. nAlthough the coal seams can vary from l1 m and g10 m, the coal is thin bedded and interbedded with mudstones and tuffaceous claystones that contribute to its strength anisotropy both in compression and tension. Uniaxial Compressive Strength tests are commonly conducted on standard size core and scaled to the rock mass for predicting borehole stability, whereas indirect tensile strength tests are needed to model response to fracture stimulation. For prediction of rock behaviour, the critical parameters for fracture initiation and propagation might be hardness or fracture toughness, and these properties are not commonly available or are derived from empirical correlation to compressive strength results. Because coal is an inherently fractured rock, smaller specimens might be required to test a truly intact material. n nnThe approach of this thesis was to go back to the basics to understand the rock material as an intact specimen. This study showed that alternative laboratory tests to derive rock parameters such as Uniaxial Compressive Strength on mini-cylinder specimens, fracture toughness and tensile strength using Short Impact Load Cell on mini-core specimens and chips/irregularparticles, and hardness testing using a Shore Scleroscope provide reliable geomechanical information obtained at the scale of interest. The experimental work was performed and evaluated using a certain number of specimens to obtain statistical sets and/or controlled specimens to draw correlations between data available on the literature, when possible. Tests were conducted on different coal lithotypes and stone to capture the heterogeneity based on composition. Tests were also conducted on fabricated 3D-printed sandstone to investigate the validity of calculations and performance of the different shape, size, and fabric orientation free from the natural variability of the material. These tests provided a valuable contribution for low rank, sub-bituminous coals; moreover, the same approach can be extended to other coal ranks and rock material. nA relationship between coal specimen size and its uniaxial compressive strength was verified for mini-cylinder specimens. The intact coal strength for the mini-cylinder specimens was upscaled to larger sizes and to the rock mass, with resulting properties comparable to coal data tested on other ranks available in the literature. In each of the short impact load cell tests, size and shape was varied and used to inform the upscaling. Despite the short impact load cell results being affected by the varying sizes, shapes, and fabric orientation of the specimens, it is a promising test to obtain the tensile strength and fracture toughness of shape-controlled specimens, as well as to observe their failure characteristics. Specific fracture energy estimated from short impact load cell utilizing heterogeneous rock of irregular-particles or shaped specimens can be correlatednto fracture toughness, whereas for homogeneous 3D-printed sandstone, the specific fracture energy can be widely different for specimens of distinct shape. For the short impact load cell tests, ultra-high speed camera was used to capture fracture initiation and propagation. Last, because fracture propagation might find a pathway through the matrix and natural fractures of coal, microscopical observation was used to understand the brittle-ductile behaviour of sub-bituminous coal organic components (macerals and microlithotypes) and inorganic components (in particular clay minerals that were often much softer than the coal).nOverall, the investigations provided confidence in the geomechanical properties derived and failure criterion for coal and interbedded rock material utilizing non-conventional experiments and provide alternatives that can be used in the coal seam gas and mining industries. For both industries, the rock characterisation requires a site-specific investigation to develop a realistic representation for numerical modelling simulations. This thesis shows that small scale specimens can be used, and tests for deformation in tension, such as SILC, and simple indentation tests can capture the brittle and ductile characteristics of sub-bituminous coal for engineering behaviour. Together these data can inform future numerical models of rock mass behaviour during drilling and reservoir stimulation for gas extraction.n