Phase transformations are widely invoked as a source of rheological weakening during subduction, continental collision, mantle convection and various other geodynamic phenomena. However, despite more than half a century of research, the likelihood and magnitude of such weakening in nature remain poorly constrained. Here we use experiments performed on a synchrotron beamline to reveal transient weakening of up to three orders of magnitude during the polymorphic quartz to coesite (SiO2) and olivine to ringwoodite (Fe2SiO4) phase transitions. Weakening becomes increasingly prominent as the transformation outpaces deformation. We suggest that this behaviour is broadly applicable among silicate minerals undergoing first-order phase transitions and examine the likelihood of weakening due to the olivine-spinel, (Mg,Fe)2SiO4, transformation during subduction. Modelling suggests that cold, wet slabs are most susceptible to transformational weakening, consistent with geophysical observations of slab stagnation in the mantle transition zone beneath the western Pacific. Our study highlights the importance of incorporating transformational weakening into geodynamic simulations and provides a quantitative basis for doing so. Subducted slabs may experience transient weakening during mineral phase transitions, consistent with slab stagnation in the mantle transition zone, according to a synchrotron beamline study of the quartz-to-coesite and olivine-to-ringwoodite phase transitions.

Abstract. Natural ices often fail at stresses much lower than those measured in laboratory settings, complicating our understanding of glacial failure and icy moon crustal fracture. This may be because strength models, which depend on the size of individual ice grains, do not account for the saltwater commonly found in terrestrial and planetary ices. We conducted grain growth experiments, finding that saltwater always modifies grain growth compared to pure ice, and that increasing volume of saltwater introduces a pinning effect limiting this growth. Ice grain size therefore depends directly on liquid fraction, controlled by salinity and temperature. Modeled effects of grain growth on tensile strength following saltwater infiltration find that low-salinity water can reduce ice strength by up to 46 % within 24 hours, narrowing gaps between observations and experiments.

Laboratory studies of rock rheology rely on purpose-built devices that can apply planetarily relevant pressures, temperatures, and non-hydrostatic stresses. Generating these pressures and stresses requires the application of large forces over small specimen areas. However, because rocks are generally polymineralic and deformation microstructures form across many length scales, it is advantageous to study relatively large (millimetric) specimens. In addition, many microstructures continue to evolve with progressive strain, so it is vital that some apparatus are able to generate enough shear strain to study these deformation phenomena. This contribution describes two new rock deformation apparatus—the Large Volume Torsion apparatus—at Washington University in St. Louis, which are capable of deforming geological specimens at high pressure and temperature (P = 3 GPa; T = 1800 K). Deformation is imposed in a torsional geometry, which enables the generation of extremely large shear strains (γ > 100) relevant to Earth’s plate boundaries and convecting mantle. A large specimen (diameter up to 4.2 mm) permits detailed postmortem microstructural analysis. Apparatus design, calibration, experimental procedures, and some examples of applications are reviewed.

Abstract Seismology is increasingly used to infer the magnitude and direction of glacial ice flow. However, the effects of interstitial meltwater on seismic properties remain poorly constrained. Here, we extend previous studies on seismic anisotropy in temperate ices to consider the role of melt preferred orientation (MPO). We used the ELLE numerical toolbox to simulate microstructural shear deformation of temperate ice with variable MPO strength and orientation, and calculated the effective seismic properties of these numerical ice‐melt aggregates. Our models demonstrate that even 3.5% melt volume is sufficient to rotate fast directions by up to 90°, to increase Vp anisotropy by up to +110%, and to modify Vs anisotropy by −9 to +36%. These effects are especially prominent at strain rates ≥3.17 × 10−12 s−1. MPO may thus obscure the geophysical signatures of temperate ice flow in regions of rapid ice discharge, and is therefore pivotal for understanding ice mass loss.

Talc is expected to be an important water carrier in Earth's upper mantle, and understanding its electrical and seismic properties under high pressure and temperature conditions is required to detect possible talc-rich regions in subduction zones imaged using geophysical observations. We conducted acoustic and electrical experiments on natural talc aggregates at relevant pressure-temperature conditions. Compressional wave velocity (Vp) was measured using ultrasonic interferometry in a Paris-Edinburgh press at pressures up to 3.4 GPa and temperatures up to 873 K. Similar Vp values are obtained regardless of the initial crystallographic preferred orientation of the samples, which can be explained by talc grain reorientation during the experiment, with the (001) plane becoming perpendicular to the uniaxial compression axis. Electrical conductivity of the same starting material was determined using impedance spectroscopy in a multi-anvil press up to 6 GPa and 1263 K. Two conductivity jumps are observed, at similar to 860-1025 K and similar to 940-1080 K, depending on pressure, and interpreted as talc dehydroxylation and decomposition, respectively. Electrical anisotropy is observed at low temperature and decreases with increasing pressure (similar to 10 at 1.5 GPa and similar to 2 at 3.5 GPa). Comparison of acoustic and electrical results with geophysical observations in central Mexico supports the presence of a talc-bearing layer atop the subducted Cocos plate.

Phase transformations are widely invoked as a source of rheological weakening during subduction, continental collision, mantle convection, and various other geodynamic phenomena. However, the likelihood and magnitude of such weakening in nature remains undetermined. Here, experiments performed on a synchrotron beamline reveal dramatic weakening across the polymorphic quartz↔coesite phase transition. Under non-hydrostatic conditions, we observe a transient decrease in effective viscosity of up to two orders of magnitude. Such weakening occurs only when the transformation outpaces deformation. We suggest that this behavior is broadly applicable among silicate minerals and examine the likelihood of slab weakening due to the olivine-spinel phase transformation. Our model suggests that cold, wet slabs are most susceptible to transformational weakening, consistent with geophysical observations of slab stagnation beneath the western Pacific.

We deformed samples with varied proportions of olivine and orthopyroxene in a deformation-DIA apparatus to test the applicability of subgrain-size piezometry to polymineralic rocks. We measured the stress within each phase in situ via X-ray diffraction during deformation at a synchrotron beamline. Subgrain-size piezometry was subsequently applied to the recovered samples to estimate the stress that each phase supported during deformation. For olivine, the final in-situ stresses are consistent with the stresses estimated via subgrain-size piezometry, both in monomineralic and polymineralic samples, despite non-steady state conditions. However, stress estimates from subgrain-size piezometry do not reliably record the in-situ stress in samples with grain sizes that are too small for extensive subgrain-boundary formation. For orthopyroxene, subgrain boundaries are typically sparse due to the low strains attained by orthopyroxene in olivine-orthopyroxene mixtures. Where sufficient substructure does exist, our data supports the use of the subgrain-size piezometer on orthopyroxene. These results do, however, suggest that care should be taken when applying subgrain-size piezometry to strong minerals that may have experienced little strain. Stresses estimated by X-ray diffraction also offer insight into stress partitioning between phases. In mixtures deformed at mean stresses > 5 GPa, orthopyroxene supports stresses greater than those supported by olivine. This stress partitioning is consistent with established theory that predicts a slightly higher stress within a ‘strong’ phase contained in a material consisting of interconnected weak layers. Overall, these results demonstrate that subgrain-size piezometry is a valuable tool for quantifying the stress state of polymineralic rocks.

During plastic deformation, strain weakening can be achieved, in part, via strain energy reduction associated with intragranular boundary development and grain boundary formation. Grain boundaries (in 2D) are segments between triple junctions, that connect to encircle grains; every boundary segment in the encircling loop has a high (>10 degrees) misorientation angle. Intragranular boundaries terminate within grains or dissect grains, usually containing boundary segments with a low (<10 degrees) misorientation angle. We analyze electron backscatter diffraction (EBSD) data from ice deformed at -30 degrees C (Th approximate to ${T}_{h}\approx $ 0.9). Misorientation and weighted Burgers vector (WBV) statistics are calculated along planar intragranular boundaries. Misorientation angles change markedly along each intragranular boundary, linking low- (<10 degrees) and high-angle (10-38 degrees) segments that exhibit distinct misorientation axes and WBV directions. We suggest that these boundaries might be produced by the growth and intersection of individual intragranular boundary segments comprising dislocations with distinct slip systems. There is a fundamental difference between misorientation axis distributions of intragranular boundaries (misorientation axes mostly confined to ice basal plane) and grain boundaries (no preferred misorientation axis). These observations suggest during progressive subgrain rotation, intragranular boundaries remain crystallographically controlled up to large misorientation angles (>>10 degrees). In contrast, the apparent lack of crystallographic control for grain boundaries suggests misorientation axes become randomized, likely due to the activation of additional mechanisms (such as grain boundary sliding) after grain boundary formation, linking boundary segments to encircle a grain. Our findings on ice intragranular boundary development and grain boundary formation may apply more broadly to other rock-forming minerals (e.g., olivine, quartz).

Kinking can accommodate significant amounts of strain during crystal plastic deformation under relatively large stresses and may influence the mechanical properties of cold planetary cryosphere. To better understand the origins, mechanisms, and microstructural effects of kinking, we present detailed microstructural analyses of coarse-grained ice (~1300 µm) deformed under uniaxial compression at -30°C. Microstructural data are generated using cryogenic electron backscattered diffraction (cryo-EBSD). Deformed samples have bimodal grain size distributions, with thin and elongated (aspect ratio ≥ 4) kink domains that develop within, or at the tips of, remnant original grains (≥ 300 µm, aspect ratio < 4). Small, equiaxed subgrains also develop along margins of remnant grains. Moreover, many remnant grains are surrounded by fine-grained mantles of small, recrystallized grains (< 300 µm, aspect ratio < 4). Together, these observations indicate that grain nucleation is facilitated by both kinking and dynamic recrystallization (via subgrain rotation). Low- (< 10°) and high-angle (mostly > 10°, many > 20°) kink bands within remnant grains have misorientation axes that lie predominantly within the basal plane. Moreover, previous studies suggest the kinematics of kinking and subgrain rotation should be fundamentally the same. Therefore, progressive kinking and subgrain rotation should be crystallographically controlled, with rotation around fixed misorientation axes. Furthermore, the c-axes of most kink domains are oriented sub-perpendicular to the sample compression axis, indicating a tight correlation between kinking and the development of crystallographic preferred orientation. Kink band densities are the highest within remnant grains that have basal planes sub-parallel to the compression axis (i.e., c-axes perpendicular to the compression axis)—these data are inconsistent with models suggesting that, if kinking is the only strain-accommodating process, there should be higher kink band densities within grains that have basal planes oblique to the compression axis (for low kink-host misorientation angles, e.g., ≤ 20°, as in this study). One way to rationalize this inconsistency between kink models and experimental observations is that kinking and dynamic recrystallization are both active during deformation, but their relative activities depend on the crystallographic orientations of grains. For grains with basal planes sub-parallel to the compression axis, strain-induced GBM is inhibited, and large intragranular strain incompatibilities can be relaxed via kinking, when other processes such as subgrain rotation recrystallization are insufficient. For grains with basal planes oblique to the compression axis, strain-induced grain boundary migration (GBM) might be efficient enough to relax the strain incompatibility via selective growth of these grains, and kinking is therefore less important. For grains with basal planes sub-perpendicular to the compression axis, kink bands are seldom observed—for these grains, the minimum shear stress required for kinking exceeds the applied compressive stress, such that kinks cannot nucleate.

Dynamic recrystallization is an important mechanical weakening mechanism during the deformation of ice, yet we currently lack robust quantitative tools for identifying recrystallized grains in the “migration” recrystallization regime that dominates ice deformation at temperatures close to the ice melting point. Here, we propose grain boundary irregularity as a quantitative means for discriminating between recrystallized (high sphericity, low irregularity) and remnant (low sphericity, high irregularity) grains. To this end, we analysed cryogenic electron backscatter diffraction (cryo-EBSD) data of deformed polycrystalline ice, to quantify dynamic recrystallization using grain boundary irregularity statistics. Grain boundary irregularity has an inverse relationship with a sphericity parameter, Ψ, defined as the ratio of grain area and grain perimeter, divided by grain radius in 2-D so that the measurement is grain size independent. Sphericity (Ψ) typically decreases with increasing grain size, up to a threshold grain size, above which Ψ either plateaus (at temperature, T< -10°C) or increases much more gradually (at T≥ -10°C). There is no apparent relationship between grain boundary sphericity and grain c-axis orientation even at very high temperatures (-4 and -5°C), where GBM dominants, suggesting little crystallographic control on the activity of grain boundary migration (GBM). Decreasing sphericity up to the threshold grain size can be explained by newly-formed, small, spherical recrystallized grains growing via strain-induced GBM and thereby developing increasingly irregular grain boundaries. We suggest that the plateau (or gradual decrease) in sphericity at larger sizes represents a population of original grains (i.e., remnant grains) that becomes increasingly irregular (at similar rates) due to the balance between GBM and nucleation. In this interpretation, the threshold grain size represents the largest grain size reached by a growing recrystallized grain by the end of an experiment. Thus, grain boundary irregularity provides a means for discriminating between recrystallized and remnant grains—a capability that is potentially useful for evaluating dynamic recrystallization processes in ice deformed at temperatures close to the melting point. The threshold grain size and experiment duration can be used to calculate the rates of recrystallization and grain size evolution associated with GBM. Grain size evolution rates are similar at high and low temperatures, suggesting similar GBM rates. Previous studies show that grain boundary mobility decreases with decreasing temperature. The driving force of GBM, on the other hand, has a positive correlation with stress, which increases with a decreasing temperature if strain rate remains unchanged. The balance between boundary mobility and driving force is likely the cause of similar GBM rates between high and low temperatures.

Strain weakening leads to the formation of high‐strain shear zones and strongly influences terrestrial ice discharge. In glacial flow models, strain weakening is assumed to arise from the alignment of weak basal planes—the development of a crystallographic preferred orientation, CPO—during flow. However, in experiments, ice strain weakening also coincides with grain size reduction, which has been invoked as a weakening mechanism in other minerals. To interrogate the relative contributions of CPO development and grain size reduction toward ice strain weakening, we deformed initially isotropic polycrystalline ice samples to progressively higher strains between −4 and −30°C. Microstructural measurements were subsequently combined with flow laws to separately model the mechanical response expected to arise from CPO development and grain size reduction. Magnitudes of strain weakening predicted by the constitutive flow laws were then compared with the experimental measurements. Flow laws that only consider grain size do not predict weakening with strain despite grain size reduction. In contrast, flow laws solely considering CPO effects can reproduce the measured strain weakening. Thus, it is reasonable to assume that strain weakening in ice is dominated by CPO development, at least under high temperature ( Th ${T}_{h}$ ≥ 0.9) and high stress (>1 MPa), like those in our experiments. We speculate that at high homologous temperatures ( Th ${T}_{h}$ ≥ 0.9), CPO development will also govern the strain weakening behavior of other viscously anisotropic minerals, like olivine and quartz. Overall, we emphasize that geodynamic and glaciological models should incorporate CPOs to account for strain weakening, especially at high homologous temperatures.

In order to better understand ice deformation mechanisms, we document the microstructural evolution of ice with increasing strain. We include data from experiments at relatively low temperatures (−20 and −30 ∘C), where the microstructural evolution with axial strain has never before been documented. Polycrystalline pure water ice was deformed under a constant displacement rate (strain rate ∼1.0×10-5 s−1) to progressively higher strains (∼ 3 %, 5 %, 8 %, 12 % and 20 %) at temperatures of −10, −20 and −30 ∘C. Microstructural data were generated from cryogenic electron backscattered diffraction (cryo-EBSD) analyses. All deformed samples contain subgrain (low-angle misorientations) structures with misorientation axes that lie dominantly in the basal plane, suggesting the activity of dislocation creep (glide primarily on the basal plane), recovery and subgrain rotation. Grain boundaries are lobate in all experiments, suggesting the operation of strain-induced grain boundary migration (GBM). Deformed ice samples are characterized by interlocking big and small grains and are, on average, finer grained than undeformed samples. Misorientation analyses between nearby grains in 2-D EBSD maps are consistent with some 2-D grains being different limbs of the same irregular grain in the 3-D volume. The proportion of repeated (i.e. interconnected) grains is greater in the higher-temperature experiments suggesting that grains have more irregular shapes, probably because GBM is more widespread at higher temperatures. The number of grains per unit area (accounting for multiple occurrences of the same 3-D grain) is higher in deformed samples than undeformed samples, and it increases with strain, suggesting that nucleation is involved in recrystallization. “Core-and-mantle” structures (rings of small grains surrounding big grains) occur in −20 and −30 ∘C experiments, suggesting that subgrain rotation recrystallization is active. At temperatures warmer than −20 ∘C, c axes develop a crystallographic preferred orientation (CPO) characterized by a cone (i.e. small circle) around the compression axis. We suggest the c-axis cone forms via the selective growth of grains in easy slip orientations (i.e. ∼ 45∘ to shortening direction) by GBM. The opening angle of the c-axis cone decreases with strain, suggesting strain-induced GBM is balanced by grain rotation. Furthermore, the opening angle of the c-axis cone decreases with temperature. At −30 ∘C, the c-axis CPO changes from a narrow cone to a cluster, parallel to compression, with increasing strain. This closure of the c-axis cone is interpreted as the result of a more active grain rotation together with a less effective GBM. We suggest that lattice rotation, facilitated by intracrystalline dislocation glide on the basal plane, is the dominant mechanism controlling grain rotation. Low-angle neighbour-pair misorientations, relating to subgrain boundaries, are more extensive and extend to higher misorientation angles at lower temperatures and higher strains supporting a relative increase in the importance of dislocation activity. As the temperature decreases, the overall CPO intensity decreases, primarily because the CPO of small grains is weaker. High-angle grain boundaries between small grains have misorientation axes that have distributed crystallographic orientations. This implies that, in contrast to subgrain boundaries, grain boundary misorientation is not controlled by crystallography. Nucleation during recrystallization cannot be explained by subgrain rotation recrystallization alone. Grain boundary sliding of finer grains or a different nucleation mechanism that generates grains with random orientations could explain the weaker CPO of the fine-grained fraction and the lack of crystallographic control on high-angle grain boundaries.

Mylonites are ubiquitous structural features of dynamic plate boundaries, and are widely assumed to represent the product of localized deformation at high pressure and temperature. There are two features of mylonites that distinguish them from typical host rocks: grain-sizes that may be reduced by orders of magnitude and mineral phases that generally well-mixed. Together, these microstructural characteristics are thought to promote rheological weakening over long geologic intervals, an essential feature of Earth-like plate tectonics. In this contribution we describe experiments that seek to reproduce deformation processes and resulting microstructures that occur during mylonitization. Experiments were conducted at high pressure (1-2 GPa) and temperature (500-750 C) on dense synthetic composites of calcite (Ca) and quartz (Qz), anhydrite (An), or fluorite (Fl). These composites were selected to investigate the influence of viscosity contrast on the phase mixing process. Shear strains of γ > 50 were produced using the Large Volume Torsion Apparatus (LVT) at Washington University in St. Louis. Ex situ microstructural analysis was performed with optical microscopy, SEM, EBSD, and TEM. Experiments are interpreted to have deformed by either viscoplastic (Ca+Fl and Ca+An) or semi-brittle mechanisms (Ca+Qz). We show that the evolution of the protolith towards recrystallized and well-mixed microstructures occurs over a large range of shear strains. The critical strain depends on the mechanism of mixing, the viscosity contrast between the two phases, and the microstructure of the starting material. Phase mixing is determined to be the product of several independent mechanisms, the relative importance of which depends on pressure, stress, strain, composition, viscosity contrast, and the ratio of the initial grain-size to the recrystallized grain size.

Vast quantities of solid CO2 reside in topographic basins of the south polar layered deposits (SPLD) on Mars and exhibit morphological features indicative of glacial flow. Previous experimental studies showed that CO2 ice is 1–2 orders of magnitude weaker than water ice under Martian polar conditions. Here we present data from deformation experiments on pure, fine‐grained CO2 ice, over a broader range of temperatures than previously explored (158–213 K). The experiments confirm previous observations of highly nonlinear power law creep at larger stresses, but also show a transition to a previously unseen linear‐viscous creep regime at lower stresses. We examine the viscosity of CO2 within the SPLD and predict that the CO2‐rich deposits are modestly stronger than previously thought. Nevertheless, CO2 ice flows much more readily than H2O ice, particularly on the steep flanks of SPLD topographic basins, allowing the CO2 to pond as observed.