Coupled processes across a claystone-concrete interface: results of a combined X-ray CT and PET transport experiment

Interfaces between dense clay materials and cementitious materials are studied in the context of deep disposal of radioactive waste for one main reason: mineral reactions due to contrasting chemistries will modify the pore network and affect transport of water, solutes and gas. Substantial research efforts were directed towards mineralogical and physical characterisation of interface regions (e.g. Mäder et al. 2018) but little evidence exists on direct observations of transport behaviour across such skins. This study aims at providing evidence on how mineralogical-physical changes at such an interface affect transport of water and solutes, and linking mineralogical-physical characterisation, X-ray computed tomography and positron emission tomography (PET).

We developed an X-ray transparent core infiltration apparatus whereby a sample core subject to a hydraulic confining pressure can be tested with a hydraulic gradient (Mäder, 2018, for details of method and design in steel and titanium). This compact apparatus uses a carbon fibre tube as pressure vessel and various polymer plastics for other components. Several small pressure tanks integrated into the apparatus allow for self-contained operation for several days, and switching of the percolating fluid. A further extension in form of an integrated lead-shielded pressure container allows also for using radioactive tracers such that the equipment can be used for positron-emission tomography (PET). PET is a superb method to directly image the mobile phase in 3D, and its time evolution (Kuhlenkampff et al., 2017).

A 14 year-old sample core was recovered by stabilized drilling from a long term in situ experiment (CI) at the Mont Terri rock laboratory (Mäder et al., 2018), containing a physically preserved interface between Opalinus Clay and OPC concrete. This larger sample (101 mm DM) was sub-sampled and a 50x50 mm core was stabilised and cored from it. The clay part shows pre-existing bedding-parallel weak jointing that can also be seen in high resolution X-ray CT. The aged interface shows mineral transformations at the mm scale with complex mineral alteration patterns in both clay and cement matrix at a sub-mm scale, including porosity re-distribution and net reduction. The OPC concrete contains aggregate and gas pores. The compound sample may represent a repository situation of a claystone somewhat disturbed by excavation, in contact with a concrete liner, with pore water transport from clay across concrete.

A long-term transport experiment was set up by injecting a synthetic claystone pore water into the core sample on the clay-side, and force advection/diffusion across the interface and out of the cement-side. The fluid is traced with deuterium as water tracer, and periodically sampled for chemical and isotopic analysis. The sample was monitored frequently by high resolution X-ray CT during the first few days, and then regularly for the first 4 months. The running experiment was then transported to Leipzig and prepared for PET. 124I was used as PET tracer, and the chosen dose allowed for continuous PET scanning during two weeks, initially every 3 hrs.

A very large data set of 2D interface characterisation (SEM/EDX mapping, etc.) and time-resolved 3D CT and PET is presently being evaluated, enhanced, imaged and interpreted. Preliminary results document an initial self-sealing effect of the joint system in the Opalinus Clay, permeation into the diffusion-controlled pore network in claystone and cement matrix, and partial filling of gas pores. PET captures some preferential flow across claystone along some remaining joints, a spreading of the tracer plume at the clay/cement interface, and some moderate preferential flow across OPC.

This approach provides much more detailed information of coupled processes in complex porous media by imaging both the stationary and the mobile phase. Compared to summation parameters, such as tracer breakthrough, there is infinitely more information obtained about the localisation of flow and the nature of the pore network and its temporal evolution.

The research leading to these results has also received funding from the European Union's European Atomic Energy Community's (Euratom) Horizon 2020 Programme (NFRP-2014/2015) under grant agreement, 662147 – Cebama. Uni Bern acknowledges funding contributions by Nagra and the Mont Terri Consortium (CI Experiment).

Kulenkampff, J., Gründig, M., Zakhnini, A., Lippmann-Pipke, J. (2016). Geoscientific process monitoring with positron emission tomography (GeoPET). Solid Earth 7, 1207-2015.
U. Mäder (2018). Advective Displacement Method for the Characterisation of Pore Water Chemistry and Transport Properties in Claystone, Geofluids, 2018.
U. Mäder, A. Jenni, C. Lerouge, S. Gaboreau, S. Miyoshi, Y. Kimura, V. Cloet, M. Fukaya, F. Claret, T. Otake, M. Shibata, B. Lothenbach (2017). 5-year chemico-physical evolution of concrete-claystone interfaces, Swiss Journal of Geosciences, 110, 307-327.