This data set provides two series of experiments from ring-shear tests (RST) on glass beads that are in use at the Helmholtz Laboratory for Tectonic Modelling (HelTec) at the GFZ German Research Centre for Geosciences in Potsdam. The main experimental series contains shear experiments to analyse the slip behaviour of the granular material under analogue experiment conditions. Additionally, a series of slide-hold-slide (SHS) tests was used to determine the rate and state friction properties. A basic characterisation and average friction coefficients of the glass beads are found in Pohlenz et al. (2020).

The glass beads show a slip behaviour that is depending on loading rate, normal stress and apparatus stiffness which were varied systematically for this study. The apparatus was modified with springs resulting in 4 different stiffnesses. For each stiffness a set of 4 experiments with different normal stresses (5, 10, 15 and 20 kPa) were performed. During each experiment loading rate was decreased from 0.02 to 0.0008 mm/s resulting in 9 subsets of constant velocity for each experiment. We observe a large variety of slip modes that ranges from pure stick-slip to steady state creep. The main characteristics of these slip modes are the slip velocity and the ratio of slip event duration compared to no slip phases.

We find that high loading rates promote stable slip, while low loading rates lead to stick-slip cycles. Lowering the normal stress leads to a larger amount of creep which changes the overall shape of a stick-slip curve and extends the time between slip events. Changing stiffness leads to an overall change in slip behaviour switching from simple stick-slip to more complex patterns of slip modes including oscillations and bimodal slip events with large and small events.

The SHS tests were done at maximum stiffness and higher loading rates (>0.05 mm/s) but at the same normal stress intervals as the main series. Using various techniques, we estimate the rate-and-state constitutive parameters. The peak stress after a certain amount of holding increases with a healing rate of b=0.0057±0.0005. From the increase in peak stress compared to the loading rate in slide-hold-slide tests we compute a direct effect a=-0.0076±0.0005 which leads to (a-b)=-0.0130±0.0006. Using a specific subset of the SHS tests, which have an equal ratio of hold time to reloading rate, we estimate (a-b)=-0.0087±0.0029. Both approaches show that the material is velocity weakening with a reduction in friction of 1.30 to 0.87 % per e-fold increase in loading rate. Additionally, the critical slip distance Dc is estimated to be in the range of 200 µm. With these parameters the theoretical critical stiffness kc is estimated and applied to the slip modes found in the main series.

We find that the changes in slip mode are in good agreement with the estimated critical stiffness and thus confirm the findings from the SHS tests.