Confidence building in the prediction of crushed salt as long-term barrier

The safety concept of a repository in rock salt is based on a multi-barrier system in that barrier’s sealing effects are time dependent. In short-term of the post-closure phase, the safe containment of radionuclides is provided by the waste canisters and geotechnical sealing elements (e.g., drift and shaft seals). In long-term, the sealing effect is provided by the geological barrier together with the crushed salt backfill. The sealing effectiveness of crushed salt evolves with ongoing compaction and therefore reduction of porosity/permeability. For the long-term safety assessment, the period in which crushed salt reaches barrier properties is a crucial information.

Numerical simulations must be used to predict the long-term compaction behaviour of crushed salt, thus it needs confidence in the results. Confidence can be built by extensive testing, verification and validation of individual simulation codes and the underlying constitutive models and by comparing constitutive models with different approaches/backgrounds.

Within the MEASURES project, a systematic approach was chosen to build trust in numerical simulations. A calculation task is defined including the analysis and calibration of constitutive models against systematic experimental observations and the subsequent calculation of a generic backfilled drift providing the opportunity to directly compare the model performance and strikingly visualize the differences. The aim is to achieve a convergence of the simulation results by systematically improving the material models and adapting the parameters to the growing experimental observations.

The process of calibration and model comparison is an iterative process and performed several times and the continuous progress is shown by the simulation of a generic backfilled drift. Each time a calibration process is finished, the generic backfilled drift is modelled and the evolution of porosity (identified as most important process variable) over time is evaluated.

The importance of employing different simulation codes/constitutive models and comparing their results is based on confidence building in the numerical predictions, quality assurance of the models and giving robustness to the results. Further, there is no complete consensus on formulating constitutive models. In MEASURES, the models are divided into phenomenological based and microstructural based formulations (Table 1) (Friedenberg et al., 2024). Uncertainties are quantified by using different model formulations due to the possibility to interpret the diversity and complexity of many physical processes involved and due to a general comparison of different methods.

Table 1. Overview of constitutive models and simulation codes

Organization	Constitutive model	Simulation code	Basis
BGE-TEC	Hein-Korthaus	FLAC 3D	Phenomenological
BGR	BGR-CS	JIFE	Microstructural
GRS	CODE_BRIGHT model	CODE_BRIGHT	Microstructural
IfG	Modified C-WIPP model*	FLAC 3D	Phenomenological
Sandia	Callahan model*	Sierra/Solid Mechanics	Microstructural
TUC	EXPO-COM	FLAC 3D	Phenomenological

 

 Acknowledgements

Thanks go to the MEASURES family for the constant support.

The project partner GRS, BGE-TEC, IfG and TUC acknowledge the project funding received by the German Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection (BMUV), represented by the Project Management Agency Karlsruhe (PTKA) (FKZ 02 E 12214 A-D).

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

References

Friedenberg, L., et al. (2024). Kompass-II: Compaction of crushed salt for safe containment - phase 2 (GRS-751).