Microtectonic analysis of an incipient thrust fault in Opalinus Clay.
Abstract:The microfabric of a fault rock controls the fault's mechanical and hydrological properties. Knowing the fabric is thus essential for estimating seismic behavior and potential fluid flow.
We studied well-preserved core and outcrop samples from the Main Fault, an up to 3 m wide zone of approximately 10 m offset in the Mont Terri Underground Research Laboratory (CH), a site to evaluate long-term safety of radioactive waste disposal. We found four main structural elements: (1) slickensided shear surfaces, (2) veins, (3) fine-grained gouge, and (4) scaly clay fabric. We investigated each element by ultra-thin section microscopy, by broad-ion-beam scanning electron microscopy (BIB-SEM) and focused-ion-beam transmission electron microscopy (FIB-TEM), by X-ray diffraction crystallography (XRD) and by naked-eye analysis.
We found extremely thin shear zones (<4µm) along which several samples broke, revealing slickensides. BIB-SEM and FIB-TEM showed that these thin shear zones comprise strongly aligned nano-sized clay particles. The porosity of the shear zones is dramatically reduced compared to the protolith. The strong alignment of clay particles, which wrap larger grains as quartz, calcite fossils and feldspar, yields a shiny, smooth surface morphology of the slickensides. Occasionally, calcite and celestite veins are associated to releasing sections such as risers of the slickenside. Gouge comprises much finer particles, a higher fabric intensity and a strong reduction in porosity and calcite content compared to the protolith. These findings suggest that gouge evolved by a cataclastic deformation mechanism aided by pressure solution of calcite. Scaly clay occurs in varying intensity and comprises thin shear zones, which sometimes act as flexural-slip faults of microfolds and C'-type shear bands.
We propose that next to cataclastic processes, pressure solution and precipitation are important micro-scale mechanisms in faulting in Opalinus Clay and thus need to be considered in extrapolating laboratory results to long-term mechanical behavior.