BN-600 fully MOX fuelled core benchmark analyses (Phase 4). Draft synthesis report - Revision 1

A benchmark analysis of a BN-600 fully mixed oxide (MOX) fuelled core design with sodium plenum above the core has been performed as an extension to the study of the BN-600 hybrid uranium oxide (UOX)/MOX fuelled core carried out during 1999-2001. This work was carried out within the the IAEA sponsored Co-ordinated Research Project (CRP) on 'Updated Codes and Methods to Reduce the Calculational Uncertainties of the LMFR Reactivity Effects'. This benchmark analysis retains the general objective of the CRP which is to validate, verify and improve methodologies and computer codes used for the calculation of reactivity coefficients in fast reactors aiming at enhancing the utilization of plutonium and minor actinides. The scope of the benchmark is to reduce the uncertainties of safety relevant reactor physics parameter calculations of MOX fuelled fast reactors and hence to validate and improve data and methods involved in such analyses. In previous benchmark analyses of the BN-600 hybrid core that closely conforms to a traditional configuration, the comparative analyses showed that sufficient accuracy is achieved using the diffusion theory approximation, widely applied in fast reactor physics calculations. With the purpose of investigating a core configuration of full MOX fuel loading, a core model of the BN-600 type reactor, designed to reduce the sodium void effect by installing a sodium plenum above the core, was newly defined for the next benchmark study. The specifications and input data for the benchmark neutronics calculations were prepared by EPPE (Russia). The specifications given for the benchmark describe only a preliminary core model variant and represent only one conceptual approach to BN-600 full MOX core designs. The organizations participating in the BN-600 fully MOX fuelled core benchmark analysis are: ANL from the USA, CEA and SA from EU (France and the UK, respectively), CIAE from China, FZK/IKET from Germany, IGCAR from India, JNC from Japan, KAERI from the Republic of Korea, IPPE from the Russian Federation. The participants applied their own state-of-the-art basic data, computer codes and methods to the benchmark analysis. Within the scope of these core benchmark analyses, they have validated their efforts to update basic nuclear data, and to improve methodologies and computer codes for calculating safety relevant reactor physics parameters. This report first addresses the benchmark definitions and specifications given for Phase 4 and briefly introduces the basic data, computer codes, and methodologies applied to the benchmark analysis by various participants. Then, the results for integral and local reactivity coefficient values obtained by the participants are inter-compared in terms of calculational uncertainty resulting from different data and method approximations along with their effects on the ULOF and UTOP transient behaviours. In addition, the results are evaluated in comparison with the results for the hybrid core. This benchmark concludes that the safety analysis of such a design would require the use of the most advanced computational tools (transport theory and three-dimensional Hex-Z modeling are required) which should be checked by representative experiments. A side conclusion to this benchmark is that the use of MOX fuel in the BN-600 core could be envisaged given with a design penalty associated to the larger reactivity coefficient uncertainties