Rock grain-scale mechanical properties influencing hydraulic fracturing using Hydro-GBM approach

Rock cracking is of key concern to many geological applications. On a grain scale, rock fracturing depends on not only the external load but also the mechanical properties of the mineral grain and grain boundary. In this study, we investigated the effect of rock grain-scale mechanical param-eters on fluid-driven cracking behaviours and discussed the identification of micro mechanical parameters in the grain-based model. A coupled hydro-grain-based DEM model (Hydro-GBM) is used to reconstruct rock microstructures and simulate hydraulic fracturing. We analyzed the influences of the main micro-mechanical parameters of mineral grain and grain boundary and explored the responsible micro-mechanisms. Results including crack initiation pressure, break-down pressure, partitions of intragranular and grain boundary cracks in tension/shear are pre-sented in detail. Then, based on the parameter analysis, some issues in identifying micro parameters in existing DEM simulations are discussed. We proposed a formulation to determine contact friction angle, which could eliminate the long-standing mismatch of shear cracking be-tween experiment and simulation. We also suggested the importance of calibrating micro results in grain-based modelling. The presented study systematically revealed the effects of rock grain-scale properties on hydraulic fractures and could provide valuable references to the selection of micro mechanical parameters in future modelling.

Automated summary

Rock cracking and fracturing have significant implications for geological applications. This study investigates the influence of rock grain-scale mechanical parameters on fluid-driven cracking behaviors and proposes a formulation to eliminate the mismatch between experimental and simulated shear cracking. The results provide valuable insights into the effects of rock grain-scale properties on hydraulic fractures and guide the selection and calibration of micro mechanical parameters in future modeling.