Synthetic gouges were prepared by mixing 50 wt % calcite and 50 wt % dolomite, as previously described in . The calcite–dolomite ratio in the experimental mixtures is similar to that found in natural fault gouges and cataclasites from the Vado di Corno Fault Zone VCFZ, Italian Central Apennines; and used in previous experimental studies . The calcite gouge was derived by crushing Carrara marble with a modal composition of 98.8 wt % calcite and < 1 wt % dolomite and muscovite (see Supplement). The dolomite gouge was derived by crushing dolomitised portions of the Calcare Massiccio Formation from the VCFZ . The crushed gouges were passed through a 250 µm sieve and then mixed together by slow tumbling for ca. 30 min. Two batches of gouge were prepared (CDM1, calcite = 47.2 wt % and dolomite = 52.8 wt %; CDM2, calcite = 42.9 wt % and dolomite = 57.1 wt %).

A total of 19 experiments were performed at slip rates from 30 µm s-1 to 1 m s-1 with SHIVA (Slow- to HIgh-Velocity rotary-shear friction Apparatus) at the Istituto Nazionale di Geofisica e Vulcanologia in Rome (Table ). The gouges were deformed inside a metal holder specifically designed for incohesive materials Fig. ;. The thickness of the gouge layers at the start of the experiments was ca. 3 mm. Horizontal displacements of the axial column were sampled at 2.5 Hz–25 kHz, and measured using a direct current differential transformer (DCDT, 50 mm range and ca. 50 µm resolution) and a linear variable differential transformer (LVDT, 3 mm range and ca. 0.03 µm resolution). Further details of the data acquisition system and the location and calibration of the load cells, detectors, and devices are found in and .

Measurements of room humidity and room temperature were collected at a distance of < 1 cm from the gouge holder before and during the experiments (Fig. a). Temperature variations during deformation were measured at an acquisition rate of 2.5 Hz using four K-type thermocouples (nickel–alumel) installed on the stationary side of the gouge holder Fig. a–b;. One thermocouple was positioned ca. 200 µm from the gouge layer (Fig. b). The other three thermocouples were located in the sample holder and stationary column to detect temperature variations due to heat conduction through the gouge holder and apparatus (Fig. a). CO2 emissions were monitored using an OmniStar™ GSD 301 O mass spectrometer designed for gas analysis at atmospheric pressure.

Experiments were performed under both room-humidity and water-dampened conditions at a constant normal stress of 17.5 ± 0.1 MPa (Table ), the same normal stress used in some previous SHIVA experiments conducted on calcite, dolomite, and calcite–dolomite gouges . Room humidity varied between 41 % and 62 %, and room temperature varied between 19 and 22 ∘C. In water-dampened conditions, ca. 2 mL of deionised water was added to the top of the gouge layer using a pipette before the gouge holder was positioned in the apparatus. Experiment s1327 was performed using a specially designed water bath that ensured saturation within the gouge layer during this long-duration experiment (Supplement). The gouge holder used in the experiments did not allow us to measure pore-fluid pressure during water-dampened experiments. Experiments were performed at target slip rates ranging from 30 µm s-1 to 1 m s-1, with acceleration and deceleration of 6 m s-2. Total displacements ranged from 0.05 to 0.4 m. Two compaction experiments were performed by applying a normal stress of 17.5 MPa for 300 s (i.e. static load experiments in Table ) and used as references for the microstructure of the starting materials.

After each experiment, the entire gouge layer was recovered and impregnated in low-viscosity epoxy (Araldite 2020) for microstructural analysis. Polished thin sections were cut perpendicular to the gouge layer and sub-parallel (i.e. tangential cut) to the slip direction (Fig. d). Microstructural analysis was performed with a Zeiss Sigma VP field-emission scanning electron microscope (SEM) at the Otago Micro and Nanoscale Imaging facility (OMNI; University of Otago. Acquisition conditions for backscattered electron images: accelerating voltage 15 kV, working distance 6–7 mm). Energy-dispersive X-ray spectroscopy (EDS) in the SEM was used to produce element maps showing the distribution of calcium and magnesium, which highlights the distribution of calcite and dolomite. Crystallographic orientation data from calcite were acquired by electron backscatter diffraction (EBSD) on SYTON-polished thin sections. Data were collected with a Nordlys EBSD camera from Oxford Instruments and processed using AZtec software (Oxford Instruments). Mineralogical changes that occurred during the experiments were determined by semi-quantitative X-ray powder diffraction (XRPD) conducted in the Department of Geoscience, University of Padua. The XRPD analyses were performed on both the bulk gouges and on small intact chips of the localised slip surfaces that formed in the experiments.

The evolution of the effective friction coefficient (μ) with slip and slip rate was influenced by the availability of water during deformation (Figs. , ). In room-humidity conditions and slip rates of ≤ 0.01 m s-1, the calcite–dolomite mixtures showed a progressive increase of μ (slip-strengthening behaviour) up to 0.75–0.80 (measured between 0.15 and 0.35 m of slip) following an initial peak fiction (μpeak) of 0.64–0.71 (Figs. , ). At a slip rate of 0.1 m s-1, a substantial decrease of μ was observed (slip weakening behaviour to steady-state μss of 0.55 ± 0.01) following a prolonged initial strengthening phase (ca. 0.062 m) that reached μpeak of 0.68 (Fig. b). Significant dynamic weakening was observed at a slip rate of 1 m s-1, following a short initial strengthening phase (lasting ca. 0.005 m) that was followed by a steady state of μss of 0.28 (Figs. a, b). A re-strengthening phase (final μ up to ca. 0.56) was observed during deceleration of the rotary column.

In water-dampened conditions, the gouge mixtures showed a similar evolution of friction at slip rates ≤ 0.1 m s-1, characterised by slight slip strengthening to slip neutral behaviour (Fig. c). Notably, μpeak and μss were lower than in room-humidity experiments, with μpeak = 0.61–0.64 and μss = 0.62–0.70, respectively (Fig. ). At a slip rate of 1 m s-1, the initial strengthening phase was much shorter than in room-humidity conditions (ca. 0.003 m), and dynamic weakening resulted in μss of 0.31 ± 0.02. Re-strengthening was also observed during deceleration, with an increase in μ up to 0.57.

No significant gouge loss was observed during the experiments, with the exception of those performed at V=0.1 m s-1 discussed below. Therefore, the evolution of axial displacement is interpreted to result from changes in gouge layer thickness due to dilation and compaction. In room-humidity conditions, the evolution of gouge layer thickness depends on slip rate (Fig. a). At V≤0.001 m s-1, the gouge layers show a three-stage evolution: (i) initial compaction of ca. 90–120 µm at the onset of sliding, (ii) dilation of ca. 50–70 µm during the slip strengthening phase, and (iii) approximately constant thickness once the steady-state friction coefficient is reached. Overall compaction of ca. 30–60 µm is recorded. At V=0.01 m s-1, after a short lived dilatancy phase (inset in Fig. a), initial compaction of 100 µm is followed by approximately constant thickness (Fig. a). At higher slip rates (V≥0.1 m s-1), continuous compaction was observed throughout the experiments (up to ca. 300 µm of axial shortening at V=1 m s-1), and compaction rate increased with slip rate (Fig. a).

Under water-dampened conditions, the gouge mixtures exhibit a similar evolution of thickness irrespective of slip rate (Fig. b). Compaction was initially rapid in the first few centimetres of sliding and then reached an approximately constant compaction rate that was similar in all experiments. Total compaction of ca. 200–250 µm was recorded (Fig. b).

Figure a shows maximum temperatures measured by the thermocouple located closest to the gouge layers Fig. b;described temperature evolution with slip. The maximum temperature (621 ∘C) was achieved in experiment s1221 performed under room-humidity conditions at V=1 m s-1 (Fig. a). For the same slip rate and normal stress but in water-dampened conditions, the maximum temperature was 210 ∘C (Fig. a). Temperature increases were detected in all experiments at slip rates ≥0.01 m s-1, and the maximum temperature increased with increasing slip rate (Fig. a).

Because the OmniStar mass spectrometer acquisition system used to detect CO2 emissions is independent from the control system on SHIVA, CO2 emissions are plotted simply against the time (in seconds) at which the mass spectrometer was started, which is variable for each experiment (Fig. b). CO2 emissions above ambient levels were only detected in experiments at slip rates ≥0.1 m s-1 (Fig. b). Because the mass spectrometer was not calibrated and the sample holder was open to the laboratory, the data can only be used in a qualitative way. In room-humidity conditions, the intensity of the CO2 peak was significantly higher at 1 m s-1 than at 0.1 m s-1. In water-dampened conditions, the CO2 peaks were substantially smaller than at equivalent room-humidity conditions.

Compared to the starting materials, no mineralogical changes were detected in any of the deformed bulk gouges (see Supplement). In room-humidity experiment s1210 (30 µm s-1), a slight broadening of the main peak for calcite was observed (Fig. a), and this was also observed to a lesser degree for dolomite. XRPD analysis of cohesive chips recovered from the slip surface of water-dampened experiment s1214 (V=30 µm s-1) indicates the presence of aragonite (Fig. b). At V=1 m s-1 and room-humidity conditions (experiment s1221), the recovered slip surface was composed of dolomite, Mg calcite, and periclase (MgO) (Fig. c). Mg calcite and periclase are two of the main products of dolomite decarbonation that starts at ca. 550 ∘C MgCa(CO3)2→MgO+(Ca,Mg)CO3+CO2;.

The mechanical behaviour and microstructural evolution of calcite–dolomite gouges show substantial differences based on the availability of water during deformation (Fig. ). In room-humidity conditions, slip strengthening at slip rates ≤ 0.01 m s-1 is associated with (i) initial compaction followed by dilation (Fig. a) and (ii) the development of a > 500 µm thick slip zone composed of a fine-grained (ca. 1 µm) calcite–dolomite mixture cut by Y-, R-, and R1-shear bands (Fig. ). These observations have previously been interpreted to relate to the development and broadening of a distributed zone of deformation during strain hardening e.g.. This is also supported by widening of the main peaks for calcite and dolomite in XRPD analysis (Fig. a), which is interpreted to result from a decrease in the mean crystallite size or deformation-induced microstrain within the crystallites e.g.. A much shorter initial period of dilatancy is observed in experiments performed at V≥0.1 m s-1 (Fig. a), and this correlates with (i) a transition from slip hardening to slip weakening (Figs. –) and (ii) the development of a well-defined, localised principal slip zone that accommodates most of the strain after it forms (Figs. e–f, and ) see also. The switch to slip weakening and a higher degree of strain localisation is also associated with a significant temperature rise generated within the principal slip zone (Fig. a), CO2 emissions (Fig. b), and the formation of Mg calcite and periclase in samples collected from the principal slip zone (Fig. c). Collectively, these observations suggest that the temperature rise at relatively high slip velocities triggered dynamic weakening and decarbonation of dolomite (and possibly calcite).

At high slip rates (V≥0.1 m s-1), the onset of dynamic weakening in carbonate gouges deformed at room humidity has been interpreted as a consequence of local heating along incipient slip surfaces, which eventually coalesce into a localised and through-going shear band . Further slip then increases the bulk temperature due to continued frictional heating in the principal slip zone and dissipation of heat in to the bulk gouge, resulting in local gouge recrystallisation . Under these conditions, high strain rates can be accommodated by temperature- and grain-size-dependent deformation mechanisms leading to “viscous” flow . performed transmission Kikuchi diffraction (TKD) analysis on electron-transparent samples of the low porosity, fine-grained principal slip zone of experiment s1221 (V=1 m s-1 under room-humidity conditions; Fig. c) to investigate the deformation mechanisms active during coseismic sliding in calcite–dolomite mixtures. Their results show that the principal slip zone is composed of a nanogranular aggregate made of two grain populations: (i) nanograins 100–300 nm in size exhibiting low internal lattice distortion, compatible with deformation by grain-size-sensitive creep, and (ii) nanograins > 800 nm in size showing development of subgrains, suggesting deformation by grain-size-insensitive creep . Although the maximum temperature measured during deformation in the present experiments was 621 ∘C (Fig. a), accommodation of the calculated strain rates (γ˙=6×103 s-1) could be explained by a significant decrease of the activation energy for creep mechanisms (and also decarbonation reactions) due to the nanogranular nature of the particles . Similar observations were documented by in experimental nanogranular principal slip zones in pure calcite gouges deformed at coseismic slip rates. Alternatively, the temperatures achieved in the slip zones of high-velocity (V=1 m s-1) but short-duration experiments (< 0.5 s) could be higher than those measured with thermocouples and estimated using numerical models. The thermocouples used here were located a few millimetres from the edge of the slipping zone (Fig. b). Additionally, they have large thermal inertia and low real acquisition rates because the electric potential developing in response to temperature changes is very slow (0.1–0.5 s) compared to the duration of the experiments (ca. 0.5 s) . Recent studies in which the temperature during experimental seismic slip was measured with optical fibres located inside the slip zone (in situ measurements at acquisition rates of 1 kHz) detected temperatures 300–400 ∘C higher than those measured with thermocouples . Temperatures in the slipping zone substantially higher than 621 ∘C would render grain-size- and temperature-dependent deformation mechanisms more efficient. Instead, in the case of experiment s1218 performed at V=0.1 m s-1, the moderate dynamic weakening (μss=0.55) can be related to more limited frictional heating within the principal slip zone both in time (max temperature measured of 190 ∘C, Fig. a) and space (patchy recrystallised areas in Fig. e–f). However, the temperature increase was at least locally sufficiently large to decompose dolomite (i.e. ca. 550 ∘C), as testified by the clear CO2 peak measured during shearing at this velocity (Fig. b).

In water-dampened conditions, the mechanical behaviour of the calcite–dolomite mixtures is similar (slight slip strengthening to slip neutral) at all slip rates up to 0.1 m s-1 (Fig. c). The thickness of the principal slip zone decreases log-linearly with increasing slip rate, indicating a progressively higher degree of localisation (Figs.  and ). However, this has no obvious effect on the steady-state friction coefficient (Fig. b), possibly suggesting that the steady state is controlled by strain and that strain is kept constant by microstructural reorganisation distributed within the slip zone. The principal slip zone is composed of a very fine-grained (≪1 µm) matrix of calcite and dolomite that includes a few well-rounded dolomite clasts up to 20–30 µm in size (Fig. ). The similarity in microstructure at all investigated slip rates suggests that water has a major role in promoting faster grain size reduction at the onset of slip, possibly by decreasing the surface energy and yield stress of calcite and dolomite . XRPD analysis of the slip surface of experiment s1214 (30 µm s-1) showed the formation of aragonite (Fig. b). Given that the starting materials were composed of calcite and dolomite only, the aragonite must have formed during deformation. documented polymorphic transformation of calcite into aragonite due to mechanical grinding in a dry (i.e. room humidity) environment. Our observations therefore suggest that relatively dry patches could develop in the gouge layer during slip (or were present at the onset of slip) or that such transformation is also possible under water-dampened conditions.

In water-dampened experiments at 1 m s-1, abrupt dynamic weakening preceded by a very short-lived strengthening phase has previously been documented in experiments on calcite gouges and calcite marbles . In gouges, suggested that the rapid onset of dynamic weakening could be related to faster grain size reduction in the presence of water, leading to an early switch from brittle deformation to grain-size-sensitive creep in the principal slip zone, analogous to the process suggested to occur in dry gouges . However, there is an apparent discrepancy between the relatively low maximum temperature measured close to the principal slip zone in water-dampened experiments (200 ∘C at V=1 m s-1; Fig. a), and the observed CO2 production (Fig. b) combined with microstructural evidence for recrystallisation during deformation (Fig. f–g). As previously discussed, this could be due to an underestimate of the peak temperature . Alternatively, vaporisation of water during coseismic sliding could buffer the temperature due to the endothermic nature of the phase transition . If the pore pressure increase is sufficiently large, then fluids (and vapour) could pressurise the gouge layers and play an important role during dynamic weakening. Finally, proposed that mechanical liming see along natural faults could be a possible slip weakening mechanism that does not necessarily involve a macroscopic temperature increase of > 500–600 ∘C.