• ASTEC
• dust
• Fission reactors
• Fusion reactors
• LOCA
• MELCOR
• Severe Accident

During the latest years, different updates were introduced inside the ASTEC and the MELCOR severe accident codes. These modifications were mainly aimed to improve the modelling aspects of specific phenomena occurring during a severe accident in a LWR, and to extend their capabilities in the prediction of accidental scenario in the future fusion reactors. Most of the improvements introduced for LWRs focused on the reduction of the uncertainties affecting the fission products behavior in the containment system. In turn, new models were implemented to cope with specific phenomena potentially occurring in two of the main fusion reactors under development: ITER and DEMO. ITER is under construction in France, while different concepts are under development for DEMO reactor. Both the ASTEC and MELCOR codes can treat the different phenomena occurring in ITER, but only the MELCOR code implements specific models to treat peculiar phenomena occurring in some DEMO concepts.
In any case, the improved models introduced by the code developers for LWRs must be validated, such as the new models implemented for ITER and DEMO. Therefore, this PhD thesis is aimed to support this validation, and to extend the capabilities of the codes for specific phenomena still not treated with sufficient detail. For this reason, it can be subdivided into four parts.
The first two parts refer to the analysis of some experimental tests investigating the fission products behavior inside the LWR containment system: the THAI Iod-11 and Iod-12 tests, and the Phébus FPT tests. The THAI tests were aimed to the investigation of the iodine-steel reaction in containment, under high relative humidity conditions with and without condensation onto the containment walls. In turn, the Phébus FPT tests were aimed to the analysis of the progression of a severe accident, from the damaging of the core to the fission products behavior in containment. These tests were investigated i to validate some specific ASTEC and MELCOR models, and to try to reduce the uncertainties shown during precedent international benchmark exercises. The objectives of these activities have been fulfilled, and both codes showed an overall behavior closer to the experimental data than in the past analysis. Furthermore, the use of similar spatial discretizations between the two codes also demonstrated that they are both capable to achieve similar thermal-hydraulics results also if different correlations and models are implemented. A good agreement for the aerosol behavior in atmosphere was also found, but important limitations still affect the modelling of Iodine behavior in the pool for MELCOR.
The third part of the work refers to the application of the ASTEC code against the several experimental tests performed in the ICE facility, aimed to investigate the behaviour of the ITER pressure suppression system. In the present thesis, these tests were utilized to expand the ASTEC validation matrix for the ITER reactor safety analysis. A deep insight on the ASTEC capabilities was achieved during this part of the activities, especially regarding its behavior employing fluids at very low pressure. Different results were obtained basing on the investigated ICE test, meaning that the code still presents some minor uncertainties that have to be addressed in the future. This activity was carried out during a stage at the Institut de Radioprotection et de Sûreté Nucléaire (IRSN) in Cadarache (France).
The last part refers to the application of the ASTEC and the MELCOR codes for the safety analysis of the DEMO reactor and specifically for the Helium Cooled Pebble Bed (HCPB) blanket concept. These analyses were carried out to underline the improvements needed for the ASTEC code to simulate this DEMO concept, and to validate the MELCOR code in its application to DEMO. A secondary aim of this final part of the work was also to support the development of this specific DEMO/HCPB concept. The results obtained from these activities showed that ASTEC is not capable to treat such kind of plant/blanket concept, but this is not a surprising result since their evaluation is still not part of the current code’s objectives. In turn MELCOR showed a greater maturity, but still some phenomena are not yet properly treated, as the dust resuspension phenomenology in case of an incidental transient. To treat this phenomenon, a simple model has been created, implemented and validated during the thesis. This model was found suitable to give a conservative evaluation of the resuspended dust mass. These last activities on the DEMO reactor were carried out during two stages at the Karlsruhe Institute of Technology (KIT) in Karlsruhe (Germany).