Research by John Elliston changes evereything!

ASSESSING THE ROLE OF SILICA GEL AS A FAULT WEAKENING MECHANISM IN THE TUSCARORA SANDSTONE

etd.ohiolink.edu/apexprod/rws_etd/send_file/send?accession=bgsu1429218108&disposition=inline

32. BACKGROUND 2.1.
Formation of a silica gel

Silica gel (SiO2.nH2O) is amorphous porous silica (Shevkina et al., 2012) that forms as a result of amorphization during comminution (the action of reducing a material, an ore, to minute particles or fragments.) of quartz in presence of water (Tullis and Goldsby, 2002). Sources of water are limited to adsorbed water, atmospheric water and water in pores and fluid inclusions, which may be released during sliding (Goldsby and Tullis, 2002). Silica particles within the gel consist of layers of various types of hydroxyl groups (Shevkina et al., 2012).

According to de Freitas (2009), gels begin as a molecular dispersion in water forming a soft solid capable of hardening. SiO4loosely-bound with OH is the molecular dispersion of H4SiO4, which initiates silica gel formation. As H2O molecules condense from this system, Si from the original structure bond with other Si by sharing a single O to form polymers (Si-O-Si).

These polymers combine to form colloidal spheres of silanol/silicic acid in suspension which, in sufficient numbers, may produce a sol. Being an unstable and incoherent system, sol can coalesce to form a more stable and coherent system referred to as gel.

Silica gel is capable of dehydration, even in water, a process that can lead to formation of crystalline silica phases. A hydrous amorphous form of silica (SiO2.nH2O) that forms from a silica gel consists of silica spheres bound together to generate an orderly 3D structure of opal.

Naturally, the solid generated will have pore spaces due to coalescing of spherical particles of similar sizes ( 150 nm) (de Freitas, 2009). The rate-limiting step for the precipitation of silanol is the breaking of Si-O bonds (Rimstidt and Barnes, 1980). This implies that the rate and amount of gel formation is dependent on the rate and amount of comminution (Rimstidt and Barnes, 1980). A large activation energy is required to produce silanol due to the difficulty in breaking strong Si-O bonds. However, 4 frictional processes promote the reaction even at moderate temperatures due to tribochemical and mechanochemical effects (Hayashi and Tsutsumi, 2010).

These effects are most active at high pressure points at contacts of asperities on the fault surface (Hayashi and Tsutsumi, 2010). Shear at these points allows distortion and breaking of Si-O bonds, making them highly reactive, particularly for the OH groups. Presence of strained bonds on the surface due to mechanical abrasion results in high solubility of the comminuted particles and therefore rapid dissolution to produce silanol (Rimstidt and Barnes, 1980).

Amorphous material has been reported in silicate gouges resulting from friction sliding experiments (Yund et al., 1990; Tullis and Goldsby, 2002). According to de Freitas (2009), amorphization of a silicate particle involves breaking the lattice structure and increasing the surface area to volume ratio. Once broken, other elements hosted in the lattice structure (i.e. Al, Fe, Ca, Mg, K, Na, etc.) escape and leave behind a damaged lattice structure enriched in silica.

This enrichment and lack of order in broken lattice structures provide the environment for development of a sol that coalesces to form a gel on the surface (de Freitas, 2009). Strained Si-O-Si bonds on the surface attract hydroxyl groups and preferentially react with water to form hydrated amorphous silica on silicate particles, pore spaces and collectively on friction surfaces resulting in dynamic weakening (Nakamura et al., 2012). This may explain why most of the silica precipitates in cataclasites and fractures occur as cements and overgrowths on silicate substrates that provide nucleation sites for growth. Silica phases precipitating from silica-undersaturated fluids are generally restricted to overgrowths on preexisting quartz surfaces while those precipitating at higher saturation levels are deposited on other surfaces in addition to quartz and may form one or more silica polymorphs by homogenous nucleation (Okamoto et al., 2010).

Page 5
Although silica gel on sliding surfaces generated in rock friction experiments is formed by amorphization of quartz in presence of water during comminution, it is not necessarily the only origin of silica gel. Silica gel on natural fault surfaces may be derived from an external source where silica-saturated hydrothermal fluids are injected into open-space cavities.

Stress drop accompanying Mode I fracture formation (Engelder, 1992) causes a sudden drop in the solubility and an already silica-saturated solution becomes supersaturated. This is a particularly favored origin in cases where the source rock is silica-depleted as reported in the Shimanto accretionary complex (Uijie et al., 2007). Although the host rock does not contain enough silica to generate a silica gel through comminution, silica-depletion is compensated by mineralization in preexisting fractures or faults (Power and Tullis, 1989; Faber et al., 2014) or as injections of silica-supersaturated fluids, providing a potential for this mechanism to occur.

Hydrous silica (opal) is common in veins and breccia fills and may by formed by fault slip processes or independent fluid flow from a different source (Onasch et al., 2010; Faber et al., 2014). Regardless of the source of silica gel on faults, experiments and studies on natural faults demonstrate that once formed, slip-weakening effects are significant due to the mechanical properties of silica gel (Faber et al., 2014).

########################################


2.3. Silica gel and natural faults
Slickensides are smooth fault surfaces with or without linear striations parallel to the slip direction (Power and Tullis, 1989). Slickensides that form during rapid sliding between rock surfaces may be characterized by natural, highly polished surfaces referred to as fault mirrors (FM) (Siman-Tov et al., 2013). Studies that extrapolate results from friction experiments to those formed in nature attribute formation of FM’s to ancient seismic slip (Fondriest et al., 2013). Although certain slickensides are indicative of different stages of the earthquake cycle (Power and Tullis, 1989), highly reflective rock surfaces do not necessarily indicate rapid fault slip (Evans et al., 2014). Other than presence of a pseudotachylyte, there is no universally accepted indicator of ancient seismic slip along faults (Fondriest et al., 2013; Faber et al., 2014). To demonstrate a seismic origin for a fault surface, its mineralogy and microstructures must be linked to temperature, strain rate and/or equilibrium pressure relative to the setting of the fault zone (Evans et al., 2014).