Intergranular pressure solution creep is an important deformation mechanism in the Earth’s crust. The phenomenon has been frequently studied and several analytical models have been proposed that describe its constitutive behavior. These models require assumptions regarding the geometry of the aggregate and the grain size distribution in order to solve for the contact stresses, and often neglect shear tractions. Furthermore, analytical models tend to overestimate experimental compaction rates at low porosities, an observation for which the underlying mechanisms remain to be elucidated.
Here we present a conceptually simple, 3D Discrete Element Method (DEM) approach for simulating intergranular pressure solution creep that explicitly models individual grains, relaxing many of the assumptions that are required by analytical models. The DEM model is validated against experiments by direct comparison of macroscopic sample compaction rates. Furthermore, the sensitivity of the overall DEM compaction rate to the grain size and applied stress is tested. The effects of the interparticle friction and of a distributed grain size on macroscopic strain rates are subsequently investigated.
Overall, we find that the DEM model is capable of reproducing realistic compaction behavior, and that the strain rates produced by the model are in good agreement with uniaxial compaction experiments. Characteristic features, such as the dependence of the strain rate on grain size and applied stress, as predicted by analytical models, are also observed in the simulations. DEM results show that interparticle friction and a distributed grain size affect the compaction rates by less than half an order of magnitude.
The zip-file Van-den-Ende_2017.018.zip contains several folders with raw data from the laboratory experiments, output data from Discrete Element Method simulations, and Python 2.7 script files that read and process these data. All data are stored in ASCII format.
We investigated the frictional properties of simulated fault gouges derived from the main lithologies present in the seismogenic Groningen gas field (NE Netherlands), employing in-situ P-T conditions and varying pore fluid salinity. Direct shear experiments were performed on gouges prepared from the Carboniferous Shale/Siltstone underburden, the Upper Rotliegend Slochteren Sandstone reservoir, the overlying Ten Boer Claystone, and the Basal Zechstein anhydrite-carbonate caprock, at 100 ºC, 40 MPa effective normal stress, and sliding velocities of 0.1-10 µm/s. As pore fluids, we used pure water, 0.5-6.2 M NaCl solutions, and a 6.9 M mixed chloride brine mimicking the formation water. Our results show a mechanical stratigraphy, with a maximum friction coefficient (µ) of ~0.65 for the Basal Zechstein, a minimum of ~0.37 for the Ten Boer claystone, ~0.6 for the reservoir sandstone, ~0.5 for the Carboniferous, and µ-values between the end-members for mixed gouges. Pore fluid salinity had no effect on frictional strength. Most gouges showed velocity-strengthening behavior, with little effect of pore fluid salinity on (a-b). However, Basal Zechstein gouge showed velocity-weakening at low salinities and/or sliding velocities, as did 50:50 mixtures with sandstone gouges, tested with the 6.9 M reservoir brine. From a Rate-and-State-Friction viewpoint, our results imply that faults incorporating Basal Zechstein anhydrite-carbonate material at the top of the reservoir are the most prone to accelerating slip, i.e. have the highest seismogenic potential. The results are equally relevant to other Dutch Rotliegend fields and to similar sequences globally.
The data is provided in a .zip folder with 29 subfolders for 29 experiments/samples. Detailed information about the files in these subfolders as well as information on how the data is processed is given in the explanatory file Hunfeld-et-al-2017-Data-Description.pdf