Integral data assimilation on U235 and U238 nuclear data and impact on fca-ix spectral indices to reassess minor actinides fission cross sections

International audience ; Critical mass calculations of various HEU-fueled reactors show there is space for improvement in current U235 and U238 nuclear data evaluations and covariances in the fast energy range. This work makes use of Bayesian Inference method as implemented in the CONRAD code. Experimental database used includes ICSBEP Uranium based critical experiments and benefits from recent re-analyses of MASURCA and FCA-IX criticality experiments (with Monte-Carlo calculations) and of PROFIL irradiation experiments. These last ones provide very specific information on U235 and U238 capture cross sections. Our integral experiment assimilation work notably suggests a 30percent decrease for U235 capture around 1-2 keV, a 10percent increase in the URR when using JEFF3.1.1 as “a priori” data. The work was focused on JEFF-3.1.1 U235 & U238 evaluations but also on minor actinides fission cross sections. An underestimation of 3-4percent for Np237 fission and 4percent for Am241 in JEFF3.1.1 is suggested by FCA-IX fission chamber results, in agreement with recent differential measurements. For Pu242 fission, FCA-IX C/E highlights a need for re-normalization for some differential measurements, as an overestimation of around 10percent is observed. For Pu238 fission, the dependency on spectrum of the C/E calculated allows us to identify an overestimation of this cross section in JEFF-3.1.1 below the threshold.