Partners : Andra, Nagra
The design option of a gas permeable seal for the L/ILW repository concept was studied in the framework of SGT – Phase 1, aimed at increasing the gas transport capacity of the backfilled underground structures without compromising the radionuclide retention capacity of the engineered barrier system (Nagra, 2008). The design option is called "engineered gas transport system (EGTS)" (Fig. 1). It involves specially designed backfill and sealing materials such as high porosity mortars as backfill materials for the emplacement caverns and sand/bentonite (S/B) mixtures with a bentonite content of 20 – 30 % for backfilling other underground structures and for the seals.
Figure 1: Schematic picture of engineered gas transport system (EGTS) with detailed design drawing of repository seal (V4).
Preliminary experimental studies confirmed the high gas transport capacity of the S/B mixtures. These experiments have shown as well the ability to design S/B mixtures with specific target permeabilities for water and gas flow. First in-situ experiences were gained through the Gas Migration Test (GMT) at Grimsel Test Site (GTS), but with in-silo emplacement. A comprehensive laboratory programme on combined gas/water transport in S/B mixtures has been started as part of the EU project FORGE.
Numerical simulations of the entire repository were conducted which demonstrate the effective functioning of the EGTS concept for a range of repository configurations and parameter variants. The model calculations show clearly an engineered gas transport system can release the gas that is produced in emplacement caverns very efficiently from the underground structures to the backfilled access tunnel into the adjacent rock formations.
More recent higher resolution modelling studies of the repository plug have revealed a rather complex gas and water flow pattern in the repository seal in the first 10000 years (Fig. 2). In the early stage after sealing the repository water from the adjacent formations will be pushed into the S/B seal under hydrostatic pressure (few 10s of bars). After about 1000 years, the gas pressure in the repository is expected to become higher than the water pressure and gas begins to be expulsed from the repository through the S/B seal into the neighbouring rock formation. Provided the modelling concept and the model assumptions underlying these results are correct, this would mean that the V5 plugs would, at no time, fully saturate during the gas production phase (about 100 ky) and thus would always remain highly conductive during the gas production phase.
Figure 2: Saturation conditions in and around the repository seal after 1, 100, 500, 1000, 1700, 10000 and 20000 years.
The geological/hydrogeological conditions at GTS (practically non-existing EDZ at certain tunnels), the technical infrastructure and the possibility to establish hydraulic/gas pressures similar to a future L/ILW repository provide ideal conditions for the performance of this large-scale demonstration and validation experiment. Of particular importance in this context is the high water injection pressure of more than 5 MPa which can be realised at the GTS to simulate in a realistic way the resaturation process of the seal (note that such high injection pressures cannot be applied at Mont Terri Test Site without impairing the integrity of the surrounding rock). A realistic resaturation history is an indispensible element for the ‘proof of concept’ for the EGTS (note that the consolidation behaviour of the host rock is of minor importance to prove the concept). The experiment complements to the HG-A experiment performed at Mont Terri URL, which focuses on the behaviour of the EDZ.
The GAST experiment focuses on the specific issue of seal behaviour during saturation and later gas invasion phase, intentionally leaving the more complicated boundary conditions (i.e. plastic behaviour) of clay host rock away. Looking only at this subsystem enables providing high-quality datasets at true scale to validate and, if necessary, improve the existing models in order to make reliable long-term predictions of the EGTS system behaviour.
Figure 3: Schematic picture of the GAST experiment layout with the 8-10m long sand/bentonite plug in between two gravel packs (for water and gas injection/circulation) and a concrete plug for reinforcement. Water injection will be either one-sided from the cement plug or double-sided, whereas gas injection will only be performed from the far end of the tunnel.
Gas-Permeable Seal Test (GAST)