High-burn up 10 × 10 100%MOX ABWR core physics analysis with APOLLO2.8 and TRIPOLI-4.5 codes

Within the frame of several extensive experimental core physics programs led between 1996 and 2008 between CEA and Japan Nuclear Energy Safety Organization (JNES), the FUBILA experiment has been conducted in the French EOLE Facility between 2005 and 2006 to obtain valuable data for the validation of core analysis methods related to full MOX advanced BWR and high-burn up BWR cores. During this experimental campaign, a particular FUBILA 10 × 10 Advanced BWR configuration devoted to the validation of high-burn up 100%MOX BWR bundles was built. It is characterized by an assembly average total Pu enrichment of 10.6 wt.% and in-channel void of 40%, representative of hot full power conditions at core mid-plane and average discharge burnup of 65 GWd/t. This paper details the validation work led on the TRIPOLI-4.5 Continuous Energy Monte Carlo code and APOLLO2.8/CEA2005V4 deterministic code package for the interpretation of this 10 × 10 high-burn up configuration. The APOLLO2.8/CEA2005V4 package relies on the deterministic lattice transport code APOLLO2.8 based on the Method of Characteristics (MOC), and its new CEA2005v4 multigroup library based on the latest JEFF-3.1.1 nuclear data file, processed also for the TRIPOLI-4.5 code. The results obtained on critical mass and radial pin-by-pin power distributions are presented. For critical mass, the calculation-to-experiment C–E on the keff spreads from 300 pcm for TRIPOLI to 600 pcm for APOLLO2.8 in its Optimized BWR Scheme (OBS) in 26 groups. For pin-by-pin radial power distributions, all codes give acceptable results, with maximum discrepancies on C/E − 1 of the order of 3–4% for very heterogeneous bundles where Pmax/Pmin reaches 4, 2. These values are within 2 standard deviations of the experimental uncertainty. Those results demonstrate the capability of both codes and schemes to accurately predict Advanced High burnup 100%-MOX BWR key-neutron parameters.