Fracture evolution in single-joint granite under uniaxial compression: A 3D grain-based discrete element and AE–MTI study

Enhanced geothermal systems (EGS) are a promising source of clean energy owing to their vast reserves and sustainable potential. Their efficient development is vital for diversifying the energy mix and mitigating supply shortages. Reservoir stimulation in EGS commonly relies on hydraulic fracturing, which enhances rock mass permeability but simultaneously elevates the risk of induced seismicity, underscoring the trade-off between production efficiency and seismic safety. In naturally fractured reservoirs, the geometry and mechanical attributes of pre-existing joints exert a primary control on crack propagation and failure. While prior studies have addressed the mechanical response of jointed rocks, quantitative characterization of energy evolution and seismic magnitude during fracturing remains limited. Here, we employ a PFC3D grain-based model (GBM) of granite with a single joint to simulate uniaxial compression. Acoustic emission (AE) events are extracted and analyzed through moment tensor inversion to resolve focal mechanisms. Results show that joint angle dominates both strength and failure mode: higher inclinations induce more localized failure sources, greater energy release, larger seismic magnitudes, and shear-dominated cracking. At maximum weakness, specimen strength reduces to 51.2% of intact granite. AE activity is also sensitive to mineral-scale failure: boundary cracking yields abundant but low-energy AE events, whereas quartz failure generates fewer but higher-magnitude events. By linking the mesoscopic fracture processes with mineral-scale mechanisms, this study enhances the understanding of the failure evolution and mechanisms of single-jointed granite. The findings may provide theoretical reference for further understanding the fracture behavior of naturally jointed granitic reservoirs under complex stress conditions relevant to hydraulic fracturing. • A 3D grain-based discrete element model with SBM–SJM contacts is developed to capture progressive post-peak softening and enhance AE consistency. • A reproducible modeling workflow explicitly maps mineralogical information to contact constitutive laws through Voronoi-based grain partitioning. • A dual-window paradigm (cumulative vs segmented) is proposed for AE analysis, enabling both early crack initiation detection and staged mechanism interpretation. • Joint angle and mineral-scale failure jointly govern strength degradation, energy release, and AE–MTI signatures, establishing a mechanistic basis for fracture evolution in single-joint granite.