E. Kasparek, H. Völzke, R. Scheidemann, U. Zencker, D. Wolff
Safety assessments of containers for transport and storage of radioactive materials involve drop scenarios pre-scribed in the regulations of the International Atomic Energy Agency – IAEA or potential accidents that might occur in a specific nuclear facility. Hereby, the design of impact limiters and foundation properties of the handling region have a major effect on the mechanical loading of the cask and lid system. Reliable numerical simulations, which very often constitute a relevant part of such safety proofs, require systematic information about the energy absorption potential of the respective materials and its implementation in FEM (finite element) programs. However, when performing drop tests, BAM (BAM Bundesanstalt für Materialfor-schung und –prüfung, Federal Institute for Materials Research and Testing) identified significant differences between numerical and experimental results, which most likely result from numerical methods that do not sufficiently reproduce the behaviour of impact limiting structures. Particularly concerned are container impact limiter made of wood or polyurethane foam as well as damping concrete bricks, which are embedded in foundations of German interim storages.
Hence, at the end of 2008 BAM initiated the research project ENREA (development of numerical methods for analyzing impact limiters subjected to impact or drop scenarios) funded by BMBF. It aims to develop appropriate materials models for simulating impact limiters based on a comprehensive experimental program. Latter consists of static and dynamic compression tests on wood, foam and damping concrete, where test parameters include temperature, support conditions and the material orientation. This contribution focus on results from the first project stage, which comprises displacement-driven tests at constant loading rates varying between 0.02 mm/s and 3,000 mm/s. The data are used to adopt materials models by stepwise optimizing the applicable parameters of FE calculations. The underlying concepts as well as the limitations of available models are presented. Not included is the outcome for wood due to the huge test volume and complexity of applicable numerical techniques. These results are rather given in an independent paper at KONTEC 2011.