Sotiris Alevizos

Thermo-poro-mechanics of chemically active creeping faults. 1: Theory and steady state considerations

In this paper, we study the behavior of a fluid-saturated fault under shear, based on the assumption that the material inside exhibits rate- and temperature-dependent frictional behavior. A creeping fault of this type can produce excess heat due to shear heating, reaching temperatures which are high enough to trigger endothermic chemical reactions. We focus on fluid-release reactions and incorporate excess pore pressure generation and porosity variations due to the chemical effects (a process called chemical pressurization). We provide the mathematical formulation for coupled thermo-hydro-chemo-mechanical processes and study the influence of the frictional, hydraulic, and chemical properties of the material, along with the boundary conditions of the problem on the behavior of the fault. Regimes of stable-frictional sliding and pressurization emerge, and the conditions for the appearance of periodic creep-to-pressurization instabilities are then derived. The model thus extends the classical mechanical stick-slip instabilities by identifying chemical pressurization as the process governing the slip phase. The different stability regimes identified match the geological observations about subduction zones. The model presented was specifically tested in the Episodic Tremor and Slip sequence of the Cascadia megathrust, reproducing the displacement data available from the GPS network installed. Through this process, we identify that the slow slip events in Cascadia could be due to the in situ dehydration of serpentinite minerals. During this process, the fluid pressures increase to sublithostatic values and lead to the weakening of the creeping slab.

Thermo-poro-mechanics of chemically active creeping faults: 2. Transient considerations

This work studies the transient behavior of a chemically active, fluid-saturated fault zone under shear. These fault zones are displaying a plethora of responses spanning from ultrafast instabilities, like thermal pressurization, to extremely slow creep localization events on geological timescales. These instabilities can be described by a single model of a rate-dependent and thermally dependent fault, prone to fluid release reactions at critical temperatures which was introduced in our companion work. In this study we integrate it in time to provide regimes of stable creep, nonperiodic and periodic seismic slip events due to chemical pressurization, depending on the physical properties of the fault material. It is shown that this chemically induced seismic slip takes place in an extremely localized band, several orders of magnitude narrower than the initial shear zone, which is indeed the signature field observation. Furthermore, in the field and in laboratory experiments the ultralocalized deformation is embedded in a chemical process zone that forms part of the shear zone. The width of this zone is shown here to depend on the net activation energy of the chemical reaction. The larger the difference in forward and backward activation energies, the narrower is the chemical process zone. We apply the novel findings to invert the physical parameters from a 16year GPS observation of the Cascadia episodic tremor and slip events and show that this sequence is the fundamental mode of a serpentinite oscillator defined by slow strain localization accompanying shear heating and chemical dehydration reaction at the critical point, followed by diffusion and backward reaction leading the system back to slow slip.

A Framework for Fracture Network Formation in Overpressurised Impermeable Shale: Deformability Versus Diagenesis

Abstract Understanding the formation, geometry and fluid connectivity of nominally impermeable unconventional shale gas and oil reservoirs is crucial for safe unlocking of these vast energy resources. We present a recent discovery of volumetric instabilities of ductile materials that may explain why impermeable formations become permeable. Here, we present the fundamental mechanisms, the critical parameters and the applicability of the novel theory to unconventional reservoirs. We show that for a reservoir under compaction, there exist certain ambient and permeability conditions at which diagenetic (fluid-release) reactions may provoke channelling localisation instabilities. These channels are periodically interspersed in the matrix and represent areas where the excess fluid from the reaction is segregated at high velocity. We find that channelling instabilities are favoured from pore collapse features for extremely low-permeability formations and fluid-release diagenetic reactions, therefore providing a natural, periodic network of efficient fluid pathways in an otherwise impermeable matrix (i.e. fractures). Such an outcome is of extreme importance the for exploration and extraction phases of unconventional reservoirs.

Bifurcation Criteria for Strain Localization in Multiphysical Systems

The study of bifurcation criteria for non-isothermal processes in geomaterials requires approaches that deviate from the classical material bifurcation approach. Indeed, in a quasi-static stress state of the medium, the admissible equilibria are of steady-creep-type and are governed also by the energy balance. Furthermore the steady-state temperature profiles are far from homogeneous as shown by [3], leading the analytical stability analysis to be rather complex. In this contribution, we adopt an overstress plasticity approach and present approximations for the criteria of the onset of localisation of deformation in a plane strain setting, using numerical continuation methods.

The Effect of Rotational and Isotropic Hardening on the Onset of Compaction Bands

Compaction bands are localized failure patterns that appear in highly porous rock material under the effect of relatively high confining pressure. Being affected mainly by volumetric compression, these bands appear to be almost perpendicular to the most compressive principal stress at a stress state at the so-called “cap” of the yield surface (Issen and Rudnicki, J Geoph Res 105:21529–21536 (2000) [4]). In this study we focus on the mechanism that leads to the onset of compaction bands by using a viscoplasticity model able to describe the post-localization response of these materials. The proposed constitutive framework is based on the overstress theory of Perzyna (Adv Appl Mech 9:243–377 (1966) [7]) and the anisotropic clay plasticity model of Dafalias (Mech Res Commun 13(6):341–347 (1986) [1]) as modified by Dafalias and Taiebat (Geotechnique 63(16):1406–1418 (2013) [2]) which provides not only the necessary “cap” of the yield surface, but introduces a rotational hardening mechanism thus taking into account possible anisotropic phenomena. Following the analysis of Veveakis and Regenauer-Lieb (J Mech Phys Solids 78:231–248 (2015) [8]) we identify the compaction bands as “static” cnoidal wave formations in the medium that occur at a post-yield regime and we study the effect of rotational and isotropic hardening on their onset. Moreover, we determine a theoretical lower limit of confining pressure in triaxial compression tests for the compaction bands to develop.