FEBEX simulation

Fig. 4. Simulations calculated with the PHREEQC geochemical code (Parkhust Appelo 1999) (a) time-dependent evolution of Eh (mV) for a Spanish granite groundwater in contact with FEBEX bentonite (b) time-dependent evolution of pH for a Spanish granite groundwater in contact with FEBEX bentonite (Arcos et al. 2000a).

The FEBEX T-H-M experiment is a valuable and important project which should lead to an improvement in the understanding of the behaviour of the bentonite barrier around heat-emitting Nuclear Fuel Waste(NFW) containers. Such large field experiments should always be undertaken with the simultaneous development of constitutive and computational models to interpret the experiments. The FEBEX bentonite possesses strong nonlinear behaviour in the unsaturated state. In order to simulate that behaviour, we have adopted a nonlinear poro-elastic approach. In this approach, the coefficients of the poroelastic equations are assumed to be functions of suction and the void ratio. These functions are derived from the state-surface equation which has been experimentally obtained from suction-controlled oedometric tests performed by the Spanish research organizations UPC and CIEMAT. 

Abstract Coupled THM simulation of the FEBEX, which is the full-scale in-situ Engineered Barrier System Experiment performed in Grimsel Test Site in Switzerland, is one Task in the international cooperation project DECOVALEX III. In the Task, the simulation of the thermal, hydraulic and mechanical behaviour in the buffer during heating phase is required, e.g. the evolutions and the distributions of stress, relative humidity and temperature at the specified points in bentonite buffer material. 

This report presents our approach for calculation of the Task. Our numerical code THAMES is the three-dimensional finite element simulator of fully coupled processes. First, we defined the input data for THAMES from the supplied properties of FEBEX bentonite. After calibrations of some parameters such as thermal vapour diffusivity, the analysis that treats fully coupled thermal, hydraulic and mechanical processes was carried out. 

The FEBEX is the full-scale in-situ Engineered Barrier System (EBS) Experiment performed in Grimsel Test Site (GTS) in Switzerland (enresa (2000)). The simulation of the coupled thermal, hydraulic and mechanical (THM) behaviour of FEBEX is the task of the DECOVALEX (DEvelopment of COupled models and their VALidation against Experiments) III. 

First, we identified the input parameter for THAMES on properties of FEBEX bentonite, because the fundamental properties of FEBEX bentonite had been obtained by various laboratory tests to identify the input data for the numerical code CODE BRIGHT (enresa (1998)). After calibrations of the all required parameters for THAMES, such as thermal vapour flow diffusivity and intrinsic permeability, the coupled THM simulations were carried out. 

Figure 5 shows the schematic view of the FEBEX. FEBEX has two heaters. Vertical sections, as D, El, E2 and G, in the test tunnel are instrumented sections that are selected for comparison between the prediction and the monitored data for part B. Sections of El and E2 are for relative humidity simulation. Sections of D and G are for temperature simulation. Section of E2 is for total pressure simulation. 

Philip de Vries model can be applied to simulate the unsaturated water movement in not only Japanese bentonite but also FEBEX bentonite. 

These blind predictions of the FEBEX data do not make a strong case that, for this particular geomechanical situation, a coupled analysis is entirely necessary. The granite in this case is sparsely fractured, and most of the inflow occurs at the lamprophyre and other more fractured areas. Also, the rock mass is sufficiently nonporous and saturated that inelastic deformation of the rock matrix is not a significant issue for repository performance. However, the exercise was very valuable for developing rationale for modeling the more complex coupled problems associated with the introduction of the bentonite barrier and the heat of the simulated waste. 

In this analysis, the transient tunnelling process was simulated in a two-dimensional section across the FEBEX tunnel. A coupled HM analysis was conducted using a Biot (1941) model with Young s modulus E = 24.68 GPa, Poisson s ratio v = 0.37, Biot s coefficient b = 1 (Terzaghi assumption), and a Biot s modulus M equal to infinity (the storage phenomena is caused only by skeleton strain). The hydraulic permeability was set to 7xl0 m after model calibration against observed water inflow into the FEBEX tunnel. 

Figure 3 presents the simulated pore pressure versus time at the 6 points described above and the measured pore pressure at the P4 interval of the FEBEX.95002. The pore pressure measured in the P4 interval (0 = 14°) agree well with the numerical results obtained from the point located at 0 = 53°. This means that a good agreement between measured and simulated pressure response at P4 could be achieved merely by changing the orientation of the initial stress. 

The entire one-month TBM drilling of the FEBEX tunnel was simulated. 

Figure 5. Finite-element model for a fully coupled simulation of the one-month TBM drilling for the FEBEX tunnel.

Figure 8. Simulated and measured fluid pressure evolution in interval P4 of borehole FEX 95.002, for adjusted local stress field at the FEBEX tunnel.

The results from two and three-dimensional numerical analyses show that fluid-pressure responses observed in the rock mass during TBM drilling of the FEBEX tunnel could not be captured using current estimates of regional stress. It was shown that the measured pressure responses can be captured in both two and three-dimensional simulations if the stress field is rotated such that contraction (compressive strain rate) and corresponding increases in mean stress occur near borehole FEX 95.002 on the side of the drift. From the results of the two-dimensional analysis, it appears that good agreement between measured and simulated evolution of fluid... 

McKinley, 1., Kickmaier, W., del Olmo, C., Huertas F. 1996. The FEBEX project full-scale simulation of engineered barriers for a HLW repository. NAGRA Bulletin No. 27, NAGRA, Switzerland. 

The model is fitted to a suction experiment for Febex bentonite and applied to the TH simulation of the bentonite buffer of the Febex in situ test, which is considered in the international Decovalex 3 project. The present approach is to describe the essential features of the TH behaviour of the buffer in a simple ID geometry. The results calculated with FEM are compared to the measurements. 

Figure 10 presents simulated and measured evolution of stress normal to the drift wall at two locations (E2G2 and B2G in Figure 9). The simulated stress began to develop as soon as the wetting commenced and increased to about 2 to 2.5 MPa at 1,000 days. The measured stress indicates that the swelling stress might not have begun to develop until several months after heater tum-on. This delay in the development of swelling stress was a common observation at many monitoring points in the bentonite barrier at FEBEX.

The general agreement between simulated and measured THM responses at FEBEX indicates that coupled THM processes are well represented in the ROCMAS code. 

For the prediction of water head changes induced by FEBEX tunnel boring, it was impossible to simulate a transient evolution due to changes in geometry with HM3D. We chose to model tunnel boring in 4 excavation phases and then simulate the steady state corresponding to the end of each phase. 

This paper will focus on the hydraulic properties of the swelling materials, especially on the bentonite that is used in the Full-scale Engineered Barrieres Experiment in Crystalline Rock (FEBEX). The influence of swelling will be shown in a simulated laboratory experiment and in the application on the FEBEX in situ experiment. 

---