Natural and lab-induced deformation impact on fluid flow on sedimentary rocks via high-speed neutron tomography

Pure and shear-enhanced compaction bands have been previously identified in natural high porosity outcrops [1 and 2, respectively]. The former were formed normal to the direction of the maximum compressive principal stress, while the latter were formed at 38-53o relative to the maximum compression. Both deformation features are characterised by textural changes due to grain rotation, fragmentation and crushing leading to local variations in porosity and presumed permeability. Thus, both deformation bands impact fluid flow with significant implications for a number of geo-energy applications, such as hydrocarbon production in subsurface geological reservoirs and CO2 storage into acquirers or depleted reservoirs.
At the laboratory scale a range of non-destructive experimental techniques have been applied to describe primarily the mechanical response during the creation and evolution of compaction bands, providing understanding of the spatio-temporal evolution of their properties [3, 4]. The aim of this work is to better understand the impact of compaction bands on fluid flow. To do so, we track the fluid flow in porous sandstones, performing water imbibition tests on specimens with laboratory induced compaction bands, using High Speed Neutron Tomography (HSNT).
The advantage of using neutrons is the high contrast between the rock material and the water, due to the high absorption of neutrons by hydrogen. Neutron radiography has been previously applied to characterise fluid flow in a shear band [5, 6], whereas x-ray CT has been used during capillary imbibition tests to characterise flow in a compaction band [7]. Here we report the first time that in-situ HSNT is used to characterise the fluid flow front during water imbibition tests in porous media such as sandstones containing laboratory induced compaction bands.
Imbibition tests and in-situ HSNT have been performed in the CONRAD Neutron Tomography instrument at the HZB [8]. HSNT was performed acquiring the projections while the rotation stage was moving, with a constant speed, from 0° to 180° and back. The acquisition time for each of the 300 radiographies was 0.2 sec, which results in a total time of 1 min per tomography. Specimens, wrapped with Teflon tape and placed in a fluorated membrane, were settled in a cup sealed with silicon and fixed at the rotation stage. A small reservoir was connected to the cup through an electro-operated valve so that the supply of water to the cup could be controlled during the experiments. Both image reconstructions and flow front detection and tracking have been achieved by using in-house software (based on ASTRA libraries in the case of reconstructions).
Our results show that HSNT can successfully capture the 4D evolution of the water front during imbibition tests in porous sandstone specimens with laboratory induced shear-enhanced compaction bands. When the water front is approximately at the level of these bands, shear-enhanced compaction bands are visualised by HSNT, whereas when the saturation of the sample increases, this is not feasible anymore. Moreover, it appears that the flow front is affected by the network of these deformation features. In particular, local increase in flow speed is observed inside these bands. Shear-enhanced compaction bands are characterised by grain fragmentation, compaction and porosity loss, presumably leading to local increase in capillary pressure, thus saturation accelerates locally.