From Microscopic Failure to Macroscopic Fracture: Understanding Energy Dissipation in Tensile Failure of Rocks

The deformation of Earth's crust is largely governed by tensile fractures, making the prediction of their propagation and arrest critical for a wide range of geological processes. Traditionally, the onset of crack growth is described using a single material property, the fracture energy. In geomaterials, however, fracture energy varies significantly with temperature, pressure, loading conditions, and scale. This variability reflects the activation of different microscopic dissipative mechanisms near the crack tip that control the energy required for macroscopic fracture propagation. In this thesis, we first quantify the variability of strength and elastic properties of rocks through the creation of the first nationwide database (Ch. 2), and subsequently investigate how microstructural interactions give rise to toughening mechanisms that govern tensile fracture in rocks. We investigate how pore fluids influence fracture resistance (Ch. 4). Experiments in saturated porous solids reveal that rapidly advancing cracks generate sharp pore-pressure drops ahead of the tip, reducing the effective stresses carried by the solid skeleton. This poromechanical coupling delays microscopic failure and increases apparent fracture energy through a localization of dissipative processes near the crack tip. As conditions transition from drained to undrained regimes, fluid-solid interactions fundamentally reshape dissipation in an otherwise simple system in agreement with poromechanical theory. We examine how stress biaxiality and structural scale influence fracture (Ch. 5). Under biaxial tension, cracks develop deflections, branches, and bridging microstructures that raise fracture energy through front roughening and fragmentation. Scaling tests (Ch. 6) show that the extent of bridging grows with crack size, confirming that fracture resistance increases systematically with size. These findings identify biaxial stress states as drivers of microstructural instabilities that couple directly to size effects, yielding scale-dependent fracture energy. Together, these results demonstrate that fracture energy is not a fixed material property, but an emergent quantity shaped by dissipative processes acting from the microscopic scale to the scale of the propagating fracture, and manifesting as systematic variability across lithologies and scales.