Fracture growth using fully coupled thermo-mechanical model in brittle rocks during thermal shock and resulting network patterns

The utilisation of geothermal heat as a form of clean energy is experiencing global growth. Many sites designated for geothermal energy extraction are in regions with elevated heat gradients, such as Iceland and Japan. However, the repercussions of the migration of cold fluids in rock formations at high temperatures, and the subsequent fracturing of the host rock, which is relevant in the context of deep geothermal energy systems, remains insufficiently understood.  In this study, we explore the three-dimensional evolution of fractures in 50×9.8×1 mm³  homogeneous slabs of various brittle rocks, subjected to a thermal shock of ΔT=580°C, through numerical simulations over 10 seconds. Initially, we validate our numerical approach using a benchmark of Al₂O₃  slab, and subsequently, we examine fracture development in granite, basalt, and shales. Our numerical methodology employs a three-dimensional finite-element-based simulator to model thermo-mechanical deformation. The in-house code is a fully coupled THM code which considers damage to predict fracture initiation. Fracture growth is predicted per fracture tip using stress intensity factors. The code uses adaptive meshing and NURBS for fracture surfaces to facilitate mesh-independent fracture growth. In our simulations, we apply triangles and tetrahedra elements to discretise surface and volume elements, respectively. Our results show the development of dozens of fractures in bi- or tri-modality, which penetrate up to 85% of the depth of the slab, for the various simulations. These results demonstrate both qualitative and quantitative agreement between the simulated slab and the benchmark by reproducing the same intertwined short-long fracture patterns and modal distribution of fracture lengths. Furthermore, they illustrate how the fracturing rates (ranging between 1-100 mm/sec), fracture length distribution (unimodal, bimodal or trimodal), and penetration depth of fractures in front of the shock front vary among the different brittle rock types.