Fluid-rock interactions play a critical role in Earth’s lithosphere and environmental subsurface systems. In the absence of chemical mass transport, mineral-hydration reactions would be accompanied by a solid-volume increase that may induce differential stresses and associated reaction-induced deformation processes, such as dilatant fracturing to increase fluid permeability. However, the magnitudes of stresses that manifest in natural systems remain poorly constrained. We used optical and electron microscopy to show that one of the simplest hydration reactions in nature [MgO + H2O = Mg(OH)2] can induce stresses of several hundred megapascals, with local stresses of as much as ~1.5 GPa. We demonstrate that these stresses not only cause fracturing but also induce plastic deformation with dislocation densities (1015 m–2) exceeding those typical of tectonically deformed rocks. If these reaction-induced stresses can be transmitted across larger length scales, they may influence the bulk stress state of reacting regions. Moreover, the structural damage induced may be the first step toward catastrophic rock failure, triggering crustal seismicity.

<p>Subsurface activities, such as energy production or geo-storage, affect the natural equilibrium of the reservoir and surrounding geological system. Fluid production from porous sandstones, for example, is often associated with reservoir compaction and induced seismicity, such as seen in the Groningen Gas Field. Production-induced stress changes lead to compaction by elastic and inelastic mechanisms. Partitioning between elastic and inelastic processes control the energy budget available for driving seismogenic events. To predict the amount of inelastic strain, it is key to identify the microscopic mechanisms controlling it. One of the current hypotheses is that micro-strains are accommodated by localized compaction of inter-granular clay films. In contrast to sandstones, claystones offer potential both as source rocks for shale gas and containment for the storage of radioactive waste and CO<sub>2</sub>. It is known that fluid flow in intact and fractured claystones is slow due to pore throats below 10 nm. However, it is unclear whether fractured shales contain a hierarchy of multi-scale highways and byways for fluid transport that is either poorly connected or more easily cross-linked and stable under in-situ conditions. Depending on how fractures change due to in-situ conditions, the shales may have a high potential as barriers in geo-storage systems, or they are of interest in relation to energy production.</p><p>This leads to two widely different research questions:</p><ul><li>How do sandstones compact due to changing stress conditions?</li> <li>How do fractures influence fluid flow in shales under in-situ stress conditions?</li> </ul><p>Despite the distance between these research questions, both can be addressed using in-situ imaging. We have developed a compaction cell and a fluid flow cell to perform experiments at the D50/NeXT beamline of the Institut Laue-Langevin in Grenoble, France. Here, combined X-ray and neutron imaging is possible.</p><p>With the compaction cell, sandstone samples from the Groningen gas field were uniaxially compacted to axial stresses of 45 MPa. At different intervals, 3D neutron and X-ray computed tomography scans were taken. As such, 4D representations (3D volumetric + time) of the in-situ changes were obtained using both neutron and X-ray tomography. The X-ray imaging allows a thorough inspection of the grain-scale deformation of the sample, while the neutron imaging highlights the changes in porosity and gives an indication of the role of clay films.</p><p>With the fluid flow cell, fractured samples of the Whitby mudstone were subjected to fluid flow under different hydrostatic pressures. The flow path evolution within the sample was visualized using neutron radiography, giving an indication of the differences between fracture and matrix permeability.</p><p>In this contribution, we will show preliminary results of four experiments performed at the D50/NeXT beamline in October 2019. We will discuss the applicability of using neutron imaging to study grain-scale processes occurring in compacting sandstone, as well as for understanding the fluid pathways in clay-rich shales, with direct implications for energy production and geo-storage.</p>

<p><strong>Molecular dynamics simulations of diffusive properties of stressed water films in quartz and clay grain contacts</strong></p><p>Floris S.R. Teuling<sup>1</sup>, Marthe G. Guren<sup>2</sup>, François Renard<sup>2</sup>, Martyn R. Drury<sup>1</sup>, Suzanne J.T. Hangx<sup>1</sup>, Helen E. King<sup>1</sup>, Oliver Plümper<sup>1</sup>, Henrik A. Sveinsson<sup>2</sup></p><ol><li>Utrecht University, Department of Earth Sciences, Princetonlaan 8a, 3584 CB Utrecht, the Netherlands</li> <li>University of Oslo, Departments of Geosciences and Physics, The Njord Centre, box 1048, Blindern, 0316 Oslo, Norway</li> </ol><p>Hydrocarbon extraction can increase effective normal stresses in geological reservoirs, potentially inducing deformation and seismicity<sup>1</sup>. The kinetics of time-dependent creep processes that could persist long after production has ended, such as pressure solution and stress corrosion, are poorly quantified. These processes can be limited by diffusion efficiency at stressed grain contacts, which depends strongly on fluid film thickness as well as interfacial and surface energies. The diffusive properties of stressed fluid films between various crystallographic surfaces of the rock forming minerals clay and quartz are critical to predict long term deformation of reservoir. Due to the small length scales of grain contacts, experimental data on these quantities are difficult to acquire. Therefore, we use molecular dynamic simulations to elucidate the physico-chemical behaviour of fluid films at different mineral interfaces.</p><p>We apply large-scale classical molecular dynamics in LAMMPS to numerically resolve fluid film behaviour in grain contacts. The silicate-water system is modelled using a modified ClayFF force field<sup>2</sup>. A β-quartz block was placed within a water-filled nanopore with either hydroxylated  β-quartz or basal illite clay surfaces as walls.  This geometry was built using the software packages Atomic Simulation Environment, Ovito and Packmol. The system was first equilibrated using an NVT thermostat and an NPT barostat for tens of picoseconds under conditions of 8 MPa fluid pressure and a temperature of 100°C. Then, a force was applied on the quartz block, corresponding to 10-200 MPa normal contact stress, such that a thin water film is squeezed at the interface between two grains. Self-diffusion constants were calculated by mean square displacements and velocity autocorrelation in films at steady state thicknesses.</p><p>Simulations reach a steady state after several nanoseconds run time. Under reservoir conditions, fluid film thicknesses are reduced to less than one nanometre. Two to three layers of adsorbed water remain in the grain contact, a result consistent with reported fluid film properties for grain contacts in upper crustal systems. Our results quantify how various juxtaposed quartz surfaces and quartz-clay interfaces influence fluid film thickness, self-diffusion and the dynamics of the water layer, which allows for constraining the kinetics of pressure solution creep in sandstone reservoirs.</p><p>Acknowledgements</p><p>This project received funding from the DeepNL programme</p><ol><li>Pijnenburg, R. P. J., Verberne, B. A., Hangx, S. J. T., & Spiers, C. J. (2019). Intergranular clay films control inelastic deformation in the Groningen gas reservoir: Evidence from split‐cylinder deformation tests. Journal of Geophysical Research: Solid Earth, 124.</li> </ol><p> </p><ol><li>Cygan, R. T., Liang, J. J., & Kalinichev, A. G. (2004). Molecular models of hydroxide, oxyhydroxide, and clay phases and the development of a general force field. The Journal of Physical Chemistry B, 108(4), 1255-1266.</li> </ol>