Calculation of kinetic parameters in lead-cooled fast reactors

In Lead Fast Reactors (LFRs), some accurate calculations of the kinetic parameters β eff and Λ eff are necessary because of their fast kinetics. The response frequency to a reactivity change exhibits a weak sensitivity to the mean neutron generation time, implying that relatively larger uncertainties in Λ eff can be tolerated during low-reactivity transients. Monte Carlo methods represent the current best-estimate approach for the evaluation of kinetic parameters, thanks to their capability to model complex geometries and detailed neutron physics without homogenization assumptions. Among these, the Iterated Fission Probability (IFP) method enables the direct calculation of β eff and Λ eff within Monte Carlo frameworks. However, the recent implementation of the IFP method in the OpenMC code has not yet been systematically validated for fast reactor systems, which are essential for the reliable deployment of LFR technology. To address this gap, a set of MOX-fuelled experimental benchmarks representative of fast-spectrum systems is selected from the ICSBEP and IRPhEP databases and modelled using OpenMC and Serpent. The objective is not merely to reproduce reference values, but to assess the capability of Monte Carlo IFP implementations to reproduce experimental kinetic parameters within uncertainty bounds. The computed kinetic parameters are compared against reference values from experimental measurements and high-fidelity calculations when measurements are unavailable. Different calculation methods and nuclear data libraries are considered to evaluate associated uncertainties. Additional analyses are performed to assess the impact of temperature, neutron leakage, and fuel inventory evolution under irradiation. Sensitivity and representativity analyses are carried out to quantify nuclear data uncertainties and to support the transposition of validation results from fast-spectrum benchmarks to LFR systems, in accordance with ASN Guide 28. The results show that both OpenMC and Serpent are able to reproduce experimental kinetic parameters within acceptable uncertainty margins. This work provides a targeted validation of the IFP methodology for fast reactor systems and supports its application to the estimation of kinetic parameters in LFR designs under development.