WP 3.3:Simulation test / Input for benchmark

Workpackage number


Start date or starting event:

Month 3

Activity type

Simulation test / Input for THMC - RTD

Participant id




Person-months per participant:





Within Workpackage 3.3 laboratory tests will be performed simulating as close as possible the conditions in a real repository. This workpackage will also provide the experimental basis for validating the constitutive models that will be mainly built on the basis of the results of Workpackage 3.1.

Precisely, the tests performed in WP3.3 aim to study in laboratory the fracturing and sealing processes that develop in the Excavation-Damaged Zone around galleries in clayey formations and the impact of a thermal phase on their evolution. For this purpose, simulation tests will be performed on hollow cylinder samples with mechanical and thermal loadings similar to the evolution that will be encountered around disposal galleries for heat emitting radioactive waste. In order to examine the possible scale effect and to enable extrapolation of the lab results obtained on normally-sized samples to the repository scale, tests will be performed on different sizes of hollow cylinder samples. The results of these tests will be used for modelling benchmark exercises in task 1 of WP 5.2. The hydraulic transfer properties through the tunnel face and through fractures will also be characterised in order to optimise the hydraulic boundary conditions in the benchmark exercises.

Description of work

The simulation tests will be jointly performed by GRS and EPFL. Although the testing device and procedure adopted by two partners are somewhat different, the experimental conditions will be discussed and adjusted by the partners during the project in order to achieve comparable results.

Middle-scale hollow cylinder tests

The tests will be performed on thick-walled hollow cylinder samples of 80 mm external diameter and 160 mm height. The mechanical and thermal loadings will be defined to simulate as well as possible the conditions and stages of repository galleries, i.e. excavation, drainage of gallery, heating and cooling cycles. A particular attention will be paid, before the thermal cycle, to induce in the tested clayey specimens a Damage Zone around the central hole, similarly to the Excavation-Damaged Zone observed around galleries (e.g. the pairs of conjugate shear surfaces in Mol URL).

EPFL will conduct the tests on Boom and Opalinus Clay samples. Improvements to the LEGEP testing device and procedure developed in the SELFRAC project will be necessary in order to generate a Damaged Zone around the central hole of the specimens and to apply the thermal loading. What concerns the testing procedure, after preparation and installation in the LEGEP cell, the thick-walled hollow cylinder samples will be subjected to the following stages:

  1. saturation, loading up to the in situ stress conditions and consolidation of the samples;
  2. external confining pressure held constant and axial plane-strain condition imposed, while the borehole pressure is reduced. The aim of this undrained unloading is to model the extension stress path that develops during the construction of disposal galleries and to generate around the central hole a Damaged Zone consisting of conjugate pairs of shear surfaces.
  3. Heating and cooling cycle to model the thermal load that will occur in a real repository and check whether it induces an evolution of the Damaged Zone extent and/or flow properties. For this purpose, a heating system will be installed in the central hole to simulate the heat generated by the waste.

To study the evolution and characterise the properties of the Damaged Zone that develops around the central hole, different techniques could be used:

  1. Permeability measurements before and after the mechanical and thermal loadings to check their effect on the hydraulic conductivity;
  2. visual inspection of the specimens after their removal from the cell;
  3. laboratory determination of water content and porosity (Mercury Intrusion Porosimetry MIP);
  4. X-Ray tomography XRCT (CHUV Lausanne).

GRS will perform the simulation experiments on hollow cylinders of Callovo-Oxfordian Argilitte:

  1. The tests will be performed in triaxial cells under consideration of relevant in-situ conditions, i.e., excavation, backfilling, heating, cooling, and re-hydration;
  2. At different steps of the tests, measurements of gas permeability in the radial direction will be performed as an indicator for the development or the sealing of the EDZ. A particular attention will be paid to quantify the effects of the heating and cooling cycle on its possible extent and additional damage.
  3. Specific investigations will also aim to develop relationships between gas/water permeability and porosity as well as to evaluate the anisotropy of permeability depending on the bedding plane and the major stress direction.

Large-scale hollow cylinder tests

In order to examine the possible scale effect and to enable extrapolation of the lab results obtained on normally-sized samples to the repository scale, simulation experiments are foreseen on large hollow cylinder samples (specimens up to 60 cm high and 30 cm in diameter). These tests will require specimens of the appropriate size from the Underground Research Laboratories (Mol, Mont Terri, Bure) and will be performed in the same way as the middle scale tests. GRS will test the Callovo-Oxfordian argillite samples in its large-scale THM triaxial apparatus while EPFL intends to use its large triaxial press (TRIROC) for the testing of the Boom and Opalinus Clay specimens. To achieve some specific objectives of the TIMODAZ project, more specifically the effect on the hydraulic conductivity of mechanical and thermal loadings, some changes to the testing device and procedure will be necessary. Moreover, owing to the very low permeability of the tested clays and the large dimensions of the specimens, extremely long testing time is expected (probably more than one year).

Interface/skin hydraulic behaviour

To asses correctly the HM behaviour during the open-drift phase, the hydraulic transfer properties through the tunnel face and throug fractures need further investigations. Numerical modelling has proved that the transfer coefficient has a significant influence on the pore pressure field. To give some quantitative values about that parameter, clay samples will be worked in order to obtain (on one face only) different surface states: samples will be cut, sawed or adjusted, rupture surfaces will be obtained through Brazilian test (tension mode I fracture) or by strain localisation in triaxial or simple shear device (shear mode II fracture) with different surface displacements. Other surfaces will be paraffin waxed. Then the sample will be placed in desiccators with controlled suction by saline solution. Sample weight evolution will be measured. The results will be interpreted by inverse analysis with special emphasis on surface transfer properties. The surface state could also be visualised e.g. with an environmental MEB in cooperation with other partners equipped with  such a technique. This work will be mainly realised by ULg (partner 8).


D7 Report on the results of the simulation tests

Milestones and expected result

The results of the proposed experimental will be used for modelling benchmark exercises. Their constitute therefore a milestone for task 1 of W5.2

Protocol and objectives of the tests performed  as input for the first end-users group review …………Month 5

Prior approval by the GB of the enduser review reports to proceed with the next related tests - Months 8,26,44

Results of simulation laboratory experiments.........................................................................................Month 36