Geological observations, seismic data as well as laboratory experiments have shown that faults lithify and recover their strength (heal) during inter-seismic periods. The mechanical-chemical process of fault healing is a key in understanding many aspects of fault behavior, such as earthquake recurrence and rupture dynamics, but is also pivotal for the application of deep geothermal energy, CO₂ sequestration and the underground storage of radioactive waste. In this contribution we investigate the mechanical-chemical recovery of fractures in carbonates at upper crustal conditions. In the upper crust, fractures are dominantly sealed through mineral precipitation from supersaturated fluids that are chemically out of equilibrium with the host rock. Such fracture sealing processes are inferred to be either relatively fast under high fluid availability and through mineral precipitation from supersaturated fluids (advective transport) or relatively slow by dissolution-precipitation processes, also known as self-sealing (diffusive transport). In order to improve the understanding of the mechanism and rate of the damage-recovery cycle of fractures, this project investigates how different sealing mechanisms (advection vs. diffusion) as well as fracture sealing rates are coupled. For this purpose, observations from natural vein systems are combined together with fracture sealing experiments with a percolation cell apparatus. In order to quantitatively document the sealing process, the selected rock samples are analyzed by laboratory-based X-ray computed microtomography and scanning electron microscopy. In summary, this work will advance knowledge about the damage-recovery cycle in fractured carbonates through the investigation of sealing processes active at the microscale.

The weak anisotropic fabrics of granites can be used to reconstruct emplacement and strain conditions to infer the paleostress field during their development. For that purpose, case studies commonly analyse the anisotropy of magnetic susceptibility (AMS) or the crystallographic preferred orientations (CPOs) of rock-forming minerals. Here, we present a systematic investigation of a range of granitoids from Germany, integrating AMS data with CPOs of feldspar and quartz (EBSD) and universal-stage measurements of microfracture orientations. Primary magmatic fabrics, as defined by AMS and feldspar CPOs, exhibit local variability in orientation and range between prolate and oblate shapes. In contrast, late- to post-magmatic fabrics are revealed through fluid inclusion planes (FIPs) and quartz CPOs. While c-axis distributions in low-strain samples are typically weak and irregular, distributions of the rhomb-faces consistently display orthotropic symmetry, characteristic of stress-induced Dauphiné twinning. These patterns, previously proposed as potential paleostress indicators, align systematically with FIP orientations, which are independently known to record coaxial strain at low differential stresses. Remarkably, both rhomb-face distributions and FIP orientations are spatially consistent across several kilometres, highlighting their potential as mappable late- to post-magmatic paleostress indicators. We discuss the implications of these findings for regional paleostress interpretations and the role of late-magmatic fluid migration pathways during the structural evolution of granites.

Antigorite is a phyllosilicate exhibits a marked mechanical anisotropy: the crystals are relatively stiff along the foliation plane (001), while being very compressible along the [001] crystallographic direction. Moreover, under shear deformation, antigorite develops a strong crystallographic preferred orientation (CPO) with the [001] direction oriented perpendicularly to the shear plane. Given its mechanical anisotropy, antigorite crystals hosted in subducting slabs are predicted to align their [001] direction perpendicular to the slab dip. Other physical properties of antigorite might show similar anisotropic behavior. Lattice thermal conductivity Λ, for example, is closely related to the elastic stiffness of a mineral. For this reason, we measured the in-plane Λ^[010], and cross-plane Λ^[001] thermal conductivities of antigorite using the Time-Domain Thermo-Reflectance (TDTR) technique^[1]. We found that Λ^[001] is ~2-3 times lower than Λ^[010], even at the high-P,T conditions of subducting slabs. To investigate the large-scale effects of antigorite’s anisotropic Λ we designed a finite difference heat diffusion model of a 2D vertically subducting slab, in which we prescribed the presence of a 3-km-thick layer of serpentinite. This insulating layer hinders the propagation of thermal energy toward the cold core of the slab. This effect would reduce the stability field of hydrous phases in the external part of the slab, thus promoting dehydration embrittlement. Furthermore, the presence of antigorite in pre-existing faults can trap the frictional heating inside shear zone, thus favouring the onset of thermal runaway. Potentially antigorite is a key mineral for the development of double seismic zones (DSZs).

Most rocks show direction-dependent seismic wave velocities, which are largely controlled by the content and CPO (crystallographic preferred orientation) of elastically anisotropic minerals, especially phyllosilicates. In order to unravel the effect of complex structural features like folding, re-folding and the scale difference between seismic experiments and lab measurements, we investigate polyphasely deformed phyllosilicate-rich samples from the NW-Tauern Window. Based on anisotropy and modeled seismic velocity data we evaluate the effect of folding and hence upscaling to a larger scale.

The CPO of millimeter-thick foliated sample cylinders was measured using high energy X-ray diffraction at the German and European Synchrotron Facilities (DESY, ESRF). Folding structures were quantified from thin sections, drill cores and field data. CPOs were then synthetically folded and weighted based on observed and hypothetical geometries. Overall seismic velocities were computed using µXRF-based modal composition, single crystal stiffness tensors and folding-derived CPOs and considered to be representative up to kilometer-scales.

Folding and crenulation reduce the Vp-anisotropy by approximately 40% at interlimb angles of 60° and 45°, respectively. Polyphase folding around parallel fold axes shows only a slight further reduction. Re-folding at distinctly different fold axis orientations leads to an elastic anisotropy reduction up to 80%. In our samples, only wide interlimb angles of >120° occur that reduce the Vp anisotropy by <20%. Shear wave splitting results in a reduction of <30%. Hence, folding may have a significant effect on the elastic anisotropy of foliated rocks and can be crucial for any interpretation of seismic data.

The occurrence of jadeite in metagranites is typically interpreted as an indicator of high-pressure conditions during metamorphism. Moreover, the albite breakdown reaction to jadeite and quartz has long been employed for the calibration of piston-cylinder. It is therefore paradoxical that jadeite is frequently absent in metagranitic rocks, even when these are associated with mafic eclogites. Possible explanations for the absence of jadeite include sluggish reaction kinetics under fluid-absent conditions, retrograde overprinting, and mechanical pressure differences between mafic eclogites and adjacent granitic lithologies.

This experimental study investigates the reaction mechanisms and the role of fluid in nominally dry jadeite formation. Piston-cylinder experiments were conducted at 2.2 GPa and 800 °C on two natural granitic rocks. One sample is a fresh granite from Finland, and the other a metagranite from Monte Rosa (Western Alps) with a complex metamorphic history involving two orogenic cycles. The history of the latter led to exsolution of magmatic plagioclase into albite and zoisite and partial replacement of biotite by white mica.

While jadeite formed in the fresh granite at albite–biotite interfaces, no jadeite developed in the Monte Rosa metagranite under identical conditions. Although the albite breakdown reaction is thermodynamically overstepped, jadeite formed not through simple albite breakdown but through localized reactions at albite–biotite interfaces. The results indicate that jadeite formation is controlled by several factors, including biotite stability dependent on aluminium content, slow reaction kinetics in the absent of fluid, and the metamorphic history of the rock, which influences local equilibrium.