Numerical validation of the ECCO self-shielding model for LFR cells

Following the Generation IV International Forum, interest in advanced reactor systems based on liquid metal cooling increased significantly in the last years. This has led to a greater demand for deterministic tools that can guarantee reliable results at a low computational cost. The ECCO cell code, adopted by the newcleo company for the design of its LFR units, produces condensed and homogenised cross sections for subsequent core calculations in ERANOS. Although ECCO was extensively validated for sodium fast reactor applications in the past, more limited studies have been carried out so far for lead fast reactors. The same applies to the generation of the input nuclear data libraries for the ECCO code. This thesis aims at numerically validating the ECCO self-shielding module. The module is implemented in the code to perform energy averaging of cross sections within resonance intervals, to simplify their numerical integration. The first part of this work focuses on the analytical description of the methods employed in ECCO and the creation of nuclear data libraries. The second part delves into the numerical validation of the code by comparing LFR cell simulations with a Monte Carlo reference. This work shows that, when excessive discrepancies are found, they are mostly related to the processing of the nuclear data libraries rather than the numerical schemes in the code. Sensitivity and decomposition analyses are employed to evaluate the impact of such discrepancies. Ultimately, the processing scheme for preparing nuclear data libraries for ECCO is implemented and revised, suggesting solutions that could ease the validation process. Hopefully, the findings of this study will contribute to upgrading and modernising the systems of codes for preparing nuclear data libraries for the ECCO code. This would lay the foundation for any subsequent validation activity on the ECCO code for lead coolant systems.