Numerical modelling of brittle fracture and step-path failure: from laboratory to rock slope scale

Recent research indicates that brittle fracture and step-path failure are important considerations in both natural high-mountain and engineered rock slopes. Newly developed techniques for field survey and numerical modeling of brittle fracture and step-path failure are presented in this research in an attempt to overcome many of the limitations of traditional approaches. Research primarily focuses on the simulation of brittle fracture and step-path failure at both the laboratory and large slope scale, and the application of LiDAR and digital-imaging techniques in the field characterization of brittle fracture. BEM, FEM, DEM(UDEC Voronoi) and hybrid FEM/DEM codes are all used in an attempt to realistically simulate laboratory step-path brittle fracture. The selected approaches are constrained through comparison of simulated fracture coalescence patterns with published physical model experiments. The combined utility of a diverse suite of data interpretation methods is shown to aid in the identification of fracture initiation and propagation. Simulation of brittle fracture propagation using both the hybrid FEM/DEM code and the FEM code PHASE2 is presented for a conceptual large open pit slope with the specific objective of investigating the relationship between intact rock bridge width and the potential for toe breakout. A major high mountain rock slope failure, the Randa rockslide, is used to demonstrate the potential of the hybrid code in modelling the influence of discontinuity persistence and step-path failure on progressive rock slope instability. The importance of characterizing three dimensional step-path failure modes in the field and the subsequent simulation using a three dimensional distinct element code are illustrated. A variety of block movement modes (i.e. translation, rotation, buckling and toppling) are shown to be highly sensitive to joint orientation, joint spacing, the assumed joint friction angle, and acting boundary constraints. The insights gained from this research provide a major contribution toward understanding of the limitations and advantages of varying numerical approaches in the simulation of brittle fractures and step-path failure. It is hoped that by examining the process of brittle fracture and step-path failure using state-of-the-art numerical codes and field characterization methods, this thesis, will provide a foundation for improved hazard assessment and rock slope design.