Uncertainty analysis of the prototype FBR Monju with the JENDL-4.0 nuclear data set

Abstract

This paper discusses the uncertainty analysis of the criticality of the prototype Fast Breeder Reactor Monju with JENDL-4.0 nuclear data. The JENDL-4.0 nuclear data are more physically correct and complete than previous JENDL releases. The well-established ERANOS code was used. Since no cross section libraries based on JENDL-4.0 exist for ERANOS, these libraries were generated as part of the work. To validate the newly made libraries, benchmark calculations were performed. A model of the Monju reactor was prepared, based exclusively on publicly available information. The criticality of May 2010 of Monju was adequately calculated. A perturbation analysis identified the most important differences between JENDL-3.3 and JENDL-4.0. A strong contribution of 241Am was found. In the subsequent uncertainty analysis, it was found that the overall uncertainty is roughly the same between JENDL-3.3 and JENDL-4.0. This result was not as expected. Two main causes have been identified: firstly, in JENDL-4.0 covariance data has been added for isotopes and energy ranges for which there previously was no covariance data. Secondly, for some isotopes the covariances in JENDL-4.0 are larger than in JENDL-3.3. For several main reactions and isotopes, the uncertainty is largely reduced between JENDL-3.3 and JENDL-4.0. But for many isotopes with a non-negligible contribution the uncertainty is increased in JENDL-4.0. The balance is a slight increase in the overall uncertainty when JENDL-4.0 is used.