Analysis of Test D1.1 of the LIFUS5/Mod3 facility for In-box LOCA in WCLL-BB

The in-box Loss of Coolant Accident (LOCA) scenario is considered as one of the most affecting safety concerns for the Water-Cooled Lead Lithium Breeding Blanket (WCLL-BB) modules of the DEMOnstration (DEMO) reactor, which is sequentially followed by a multi-phase multi-component physical and chemical interaction. Therefore, the transient behavior of such accidents has to be carefully investigated during the design phase of the plant, to evaluate the consequences and to adopt the necessary mitigating countermeasures. This also requires a numerical predictive tool, which is capable to model such transients and predict the relevant phenomena under an operational condition and the connected safety parameters i.e. system pressure, temperature, chemical products mass, and volume fractions of all the existing components. Consequently, the SIMMER-III code was firstly improved at the University of Pisa by implementing the chemical reaction between PbLi eutectic alloy and water. In addition to this, an experimental campaign and a test-matrix have been recently designed according to the LIFUS5/Mod3 facility to perform a series of experiments and code post-test analyses. In the present work, the experimental data of the first LIFUS5/Mod3 test is used for the validation of the chemical model implemented in SIMMER-III through a comprehensive sensitivity study. The applied methodology for the code validation is based on a three-step procedure including qualitative analysis, quantitative analysis and the results from sensitivity analyses. The qualitative accuracy evaluation is performed through a systematic comparison between experimental and calculated time trends based on the engineering analysis, the resulting sequence of main events and the identification of phenomenological windows and of relevant thermo-hydraulic aspects. Afterwards, the accuracy of the code prediction is evaluated from a quantitative point of view by means of selected, widely used, figures of merit. Finally, the results from the sensitivity cases are analysed and quantified, to determine the effects of the most influencing code input options and transient parameters. Furthermore, the analysis is followed by applying the Fast Fourier Transform Method (FFTM) to the experimental signals and all the sensitivity calculations. The comparison shows a very good agreement for pressure transient between the experimental and numerical data, while for the temperature and the hydrogen production the results fall into acceptable criteria, which means that the code is reliable in capturing and predicting the transient values but not perfectly match with the experimental signals.