Contrasting Mechanical and Hydraulic Properties of Wet and Dry Fault Zones in a Proposed Shale‐Hosted Nuclear Waste Repository

Abstract

The underground disposal of high‐level nuclear waste is a pressing issue for several countries. In Switzerland, the Opalinus Clay formation is a shale with favorable barrier properties. However, small‐to‐large faults intersecting the formation bring the long‐term integrity of the future repositories into question. Here we present the first systematic laboratory study on the frictional strength, stability, dilatancy, and permeability of simulated Opalinus Clay gouge under typical repository conditions. Wet gouges exhibit an extremely low coefficient of friction (μf~0.16), velocity‐strengthening behavior, and shear‐enhanced dilatancy at the onset of slip, and permeability increase. Conversely, dry gouges remain weak (μf~0.36) but exhibit a transition from unstable to stable sliding with increasing sliding velocity. Thus, we infer that faults hosted in Opalinus Clay could be easily reactivated via aseismic creep, possibly acting as poor fluid conduits. However, if temporarily dried, the faults become potentially unstable, at least, at low sliding velocities (<~10 μm/s).

Laboratory Study on Fluid‐Induced Fault Slip Behavior: The Role of Fluid Pressurization Rate

“…The generation of a thin layer fault gouge in the presence of water potentially forming a gel or suspension resembling a highly viscous fluid may partly support normal stress acting between fault walls (Brodsky & Kanamori, 2001). This lubrication process is expected to result in a drop of friction beyond the critical slip distance (Cornelio et al, 2019;Di Toro et al, 2011;Reches & Lockner, 2010), as observed in wet gouges (Orellana et al, 2019;Sammis et al, 2011). Additionally, the real contact area between fault planes is expected to be reduced due to the gradual decrease of effective normal stress during fluid pressurization (Rubinstein et al, 2004).…”

Mathematical Modelling of Fault Reactivation Induced by Water Injection

“…In this paper, many fault parameters, such as the friction angle, were derived through a calibration procedure because they were not readily available at the time this research was conducted. Recent experimental data for fault properties [7][8][9][10][11][12] will be used in future work. The conditions leading to the generation of seismic events, and estimates of their magnitudes will also be studied.…”

“…The shear strength parameters are the key parameters that determine the onset of fault failure, while the damage enhancement parameter determines the magnitude of the permeability increase in the post-failure phase. The friction angle is directly related to the friction coefficient, which has been shown to be highly variable [8,9,12]. Therefore, we performed a sensitivity study in this section, and also consider it spatial variability in the heterogeneous case described in a subsequent section.…”

“…The shear strength of the fault depends largely on the asperities and contact conditions of the surfaces of the individual fractures that are present in the fault. Therefore, a shear strength parameter such as the friction angle would be spatially variable [7][8][9]12], and its spatial distribution is a reflection of the topology of the fracture wall asperities, contacts and degree of wetness and the characteristics of gouge material that might be present. In practice, if measured data on such parameter as the friction angle are available at different fault locations, the fault friction angle can be approximated as a function of position, using a finite Fourier series of the form:…”

“…Fault re-activation can either trigger seismic events (seismic situation) or not (aseismic situation), depending on such factors as the "seismic efficiency" [6], i.e. the fraction of energy stored in the fault released as seismic waves, and the friction properties of the fault [7][8][9][10][11][12] Mathematical modelling of induced seismicity has been proposed [13] based on the basic theory of poromechanics and the consideration of Coulomb's friction as the threshold for fault failure. Although the above theoretical framework is well established, its application to the prediction of fault behaviour is a complex endeavour, and is the subject of many current research activities [14,15].…”