Publication
Failure in brittle, porous materials initiates when structural damage localizes along an emergent failure plane in a transition from stable crack growth to dynamic rupture. Due to the rapid nature of this critical transition, the precise micro-mechanisms involved are not fully understood and difficult to capture. However, these mechanisms are crucial drivers for earthquakes, including induced seismicity. Here we observe these micro-mechanisms directly by controlling the rate of micro-seismic events to slow down the transition in a unique triaxial deformation apparatus that combines acoustic monitoring with contemporaneous in-situ x-ray imaging of the microstructure. The results provide the first integrated picture of how damage and associated micro-seismic events evolve together during sample weakening, allowing us to directly constrain the partition between seismic and aseismic deformation at the micro-scale. The evolving damage in the 3D x-ray volumes and local strain fields undergoes a breakdown sequence involving self-organized exploration of candidate shear zones, spontaneous tensile failure and rotation of individual grains within a localized shear zone, and formation of a proto-cataclasite, highlighting the importance of aseismic mechanisms such as grain rotation in accommodating bulk shear failure. Dilation and shear strain remain strongly correlated throughout failure confirming the existence of a cohesive zone but with crack damage distributed rather than concentrated solely at the propagating front of a discontinuity. Seismic amplitude is not correlated with local imaged strain intensity, and the seismic strain partition coefficient is very low overall. We explain the stress evolution in terms of a new, sub-critical fracture mechanics model, and compare the seismic signatures between this experiment and a sister experiment carried out under constant strain rate loading. Compared with loading under a constant strain rate, reactive loading to maintain a constant micro-seismic event rate increases the seismic b-value, decreases the maximum event magnitude, suppresses the number of events of all sizes, and reduces the seismic strain partition coefficient. Qualitative inspection of comparable grey-scale x-ray volumes between the two experiments (peak stress and post-failure after unloading) showed a greater degree of brittle damage and earlier localization of en-echelon microcracks along the eventual failure plane in the sample deformed under a constant strain rate. Our results explain the effectiveness of seismic event rate control for managing the risk from induced seismicity in mining settings and imply that it may be more effective for pre-emptive risk management than magnitude based ‘traffic light’ systems, which adjust operations (green: proceed; amber: proceed with caution; red: suspend) in response to extreme events that exceed pre-defined magnitude thresholds. Effective management of induced seismicity is essential for safe operati