Distinguishing friction- from shock-generated melt products in hypervelocity impact structures

ABSTRACT Field, microtextural, and geochemical evidence from impact-related melt rocks at the Manicouagan structure, Québec, Canada, allows the distinction to be made between friction-generated (pseudotachylite) and shock-generated melts. Making this distinction is aided by the observation that a significant portion of the impact structure’s central peak is composed of anorthosite that was not substantially involved in the production of impact melt. The anorthosite contrasts with the ultrabasic, basic, intermediate, and acidic gneisses that were consumed by decompression melting of the >60 GPa portion of the target volume to form the main impact melt body. The anorthosite was located below this melted volume at the time of shock loading and decompression, and it was subsequently brought to the surface from 7–10 km depth during the modification stage. Slip systems (faults) within the anorthosite that facilitated its elevation and collapse are occupied by pseudotachylites possessing anorthositic compositions. The Manicouagan pseudotachylites were not shock generated; however, precursor fracture-fault systems may have been initiated or reactivated by shock wave passage, with subsequent tectonic displacement and associated frictional melting occurring after shock loading and rarefaction. Pseudotachylites may inject off their generation planes to form complex intrusive systems that are connected to, but are spatially separated from, their source horizons. Comparisons are made between friction and shock melts from Manicouagan with those developed in the Vredefort and Sudbury impact structures, both of which show similar characteristics. Overall, pseudotachylite has compositions that are more locally derived. Impact melts have compositions reflective of a much larger source volume (and typically more varied source lithology inputs). For the Manicouagan, Vredefort, and Sudbury impact structures, multiple target lithologies were involved in generating their respective main impact melt bodies. Consequently, impact melt and pseudotachylite can be discriminated on compositional grounds, with assistance from field and textural observations. Pseudotachylite and shock-generated impact melt are not the same products, and it is important not to conflate them; each provides valuable insight into different stages of the hypervelocity impact process.

Probing the Structure of Melts, Glasses, and Amorphous Materials

Liquids, glasses, and amorphous materials are ubiquitous in the Earth sciences and are intrinsic to a plethora of geological processes, ranging from volcanic activity, deep Earth melting events, metasomatic processes, frictional melting (pseudotachylites), lighting strikes (fulgurites), impact melting (tektites), hydrothermal activity, aqueous solution geochemistry, and the formation of dense high-pressure structures. However, liquids and glassy materials lack the long-range order that characterizes crystalline materials, and studies of their structure require a different approach to that of conventional crystallography. The pair distribution function is the neutron diffraction technique used to characterize liquid and amorphous states. When combined with atomistic models, neutron diffraction techniques can determine the properties and behavior of disordered structures.

Pseudotachylyte veins in accretionary complexes: melt or mechanical wear?

<p>Whether seismic rupture propagates over large distances to generate mega-earthquakes or on the contrary slows down quickly, is heavily dependent on the slip processes operating within the fault core, such as frictional melting or intense grain-size reduction and amorphization. The record, in fossil fault zones, of seismic slip, consists in many instances in Black Faults Rocks (BFR), that consists in a generally thin dark and aphanitic veins similar to volcanic glasses, which cross-cuts sharply a weakly foliated tectonic mélange, and have been interpreted as resulting from quenching of a melt (i.e. pseudotachylytes). Such interpretation has nevertheless been questioned because identical (micro- and nano-) textures have been observed on intensely comminuted natural fault rocks and on slow creep experiments on crustal rocks.</p><p>In this study, we report a new dataset of high spatial-resolution Raman Spectroscopy of Carbonaceous Materials (RSCM) profiles across natural BFR from two accretionary complexes. RSCM is sensitive to both temperature and deformation. We have carried out analyses on Okitsu and Nobeoka BFR from the Shimanto Belt and Kodiak BFR from the Kodiak Accretionary Complex to discriminate the slip weakening process. The Raman Intensity Ratio (i.e. R1 in Beyssac et al., 2002) and the Area ratio (RA1 in Lahfid et al., 2010) show a drastic and discontinuous stepped increase along profiles across the BFR, revealing a higher crystallinity. Moreover, in spite of scattering, highest values have been measured on the rim between the BFR and the host-rock. Fluidization structures, interpreted as injection veins, show similar values to the ones in the host rock. Additionally, using an experimentally calibrated kinetics 1D modelling of Intensity ratio evolution with temperature, we compared the natural Raman spectroscopy profiles to different scenarios of temperature increase during seismic slip. In the three examples of BFR from accretionary complexes interpreted as natural pseudotachylytes, RSCM profiles are not consistent with a molten origin and must reflect mechanical wear during deformation.</p><p>Consequently, these results bear major consequences on the dynamics of faulting in accretionary complexes, as the slip-weakening processes that occur during seismic slip rely on extreme grain-size reduction and fluidization rather than melting.</p>

Impact of Coseismic Frictional Melting on Particle Size, Shape Distribution and Chemistry of Experimentally-Generated Pseudotachylite

In natural friction melts, or pseudotachylites, clast textures and glass compositions can influence the frictional behavior of faults hosting pseudotachylites, and are, in turn, sensitive to the processes involved in pseudotachylite formation. Quantification of these parameters in situations where the host rock composition and formation conditions are well-constrained, such as analogue experiments, may yield calibrations that can be employed in analysis of natural pseudotachylites. In this paper, we experimentally-generated pseudotachylites in granitoid rocks (tonalite and Westerly granite) at Pconf = 40 MPa and slip rates of ∼0.1 m s−1, comparable to the conditions under which natural pseudotachylite is known to form in Earth’s upper crust. We find variations in both clast textures and glass compositions that reflect formation processes, and probably influence the frictional behavior of similar natural faults hosting pseudotachylite. Quantification of particle size and shape distribution with a semi-automatic image analysis method, combined with analysis of glass and host-rock composition of these experimentally generated pseudotachylites, reveals that the textures of pseudotachylite material evolved by combinations of 1) comminution, 2) heterogeneous frictional flash melting, and 3) homogeneous (diffusive) clast melting and/or marginal decrepitation. Fractal dimensions of pseudotachylite-hosted clasts (D ∼ 3) that are greater than those of marginal fragmented host rock particles (gouge, D ∼ 2.4), reflect an increase of the intensity of comminution by slip localisation during a pre-melting phase. Chemical analyses demonstrate that these pseudotachylite glasses were generated by frictional flash melting, where host rock phases melt individually. Biotite is the least resistant to melting, feldspar intermediate, and quartz is the most resistant. The peudotachylite glass generated in these experiments has an alkaline composition, is depleted in SiO2 compared to the bulk host-rock, and shows heterogeneous compositions in a single sample related to proximity to host-rock minerals. The percentage contributions of host rock phases to the melt, calculated by a mixing model, shows that glass compositions are dominated by plagioclase and biotite. Within the melt, margins of clasts were dissolved uniformly by diffusion and/or affected by marginal decrepitation, resulting in convex and round shapes with convexities averaging ∼0.8 and circularities averaging ∼0.65.

Lower crustal earthquake facilitated by overpressurized frictional melts

<p>Earthquakes are among the most catastrophic geological events that last only several to tens of seconds. During earthquakes, many processes may occur including rupturing, frictional sliding, pore fluid pressurization and occasionally frictional melting. However, little direct records of these fast processes remain preserved through geological time. During rapid shearing, frictional melt may form that lubricates the rocks and facilitates further sliding. The frictional melt layer may quench quickly within seconds to minutes depending on its thickness. After quenching, the product pseudotachylyte preserves valuable information about the conditions when the frictional melt was generated. Here, we study pseudotachylyte from Holsnøy Island in the Bergen Arcs of Western Norway, an exhumed portion of the lower continental crust. The investigated pseudotachylyte vein is ca. 1-2 cm thick and free of injection veins along the 2 m visible length of the fault. The pseudotachylyte matrix is made up of fine-grained omphacite (Jd<sub>38</sub>), sodic plagioclase (Ab<sub>83</sub>) and kyanite with minor rutile and sulphides. Many dendritic garnets are found within the pseudotachylyte showing gradual grain size reduction towards the wall rock. This suggests that the garnets crystallized during rapid quenching. The stability of epidote, kyanite and quartz in the wall rock plagioclase, and omphacite and albitic plagioclase together with quartz in the pseudotachylyte matrix constrains the ambient P ca. 1.5-1.7 GPa and T ca. 650-750°C. Using Raman elastic barometry, the constrained pressure condition from quartz inclusions in the dendritic garnets in the pseudotachylyte is > 2 GPa. Based on an elastic model (Eshelby’s solution), it is not possible to maintain 0.5 GPa overpressure within a thin melt layer by thermal pressurization or melting expansion. A potential explanation is that GPa level differential stress was present in the wall rocks and the melt pressure approached the normal stress when shear rigidity vanished during frictional melting. Our study illustrates how overpressure can be created within frictional melt veins under conditions of high differential stress, and offers a mechanism that facilitates co-seismic weakening during lower crustal earthquakes.  </p>