The NEA has studied multiple recycling issues associated with various reactor systems fuelled with mixed-oxide (MOX) and published a series of computational physics benchmarks. This has led to improvements in the nuclear data libraries and calculation methods. Several benchmarks were completed comparing those findings with data from experiments. Previous benchmarks have concentrated mainly on PWRs, BWRs, VVER-1000s and FRs. The present benchmark concerns a pebble bed modular reactor (PBMR) fuelled with reactor-grade plutonium.
Although the benchmark has been specifically designed to provide intercomparisons for plutonium and thorium fuels, phases of calculations for uranium fuel have also been included. The purpose of these phases is to identify any increased uncertainties, relative to uranium fuel, that are associated with plutonium and thorium fuel.
This report provides an analysis of the twelve sets of results supplied by seven experts from five countries. Participants have used nuclear data from three different evaluations having applied both Monte Carlo and deterministic methods of analysis. Participants using the same nuclear data report similar results, although some differences have been noted, particularly in relation to the fuel temperature coefficients and the whole-core xenon fission product poisoning effect. There is also evidence of good agreement between Monte Carlo and deterministic solutions for some of the participants despite the difficult nature of the problem with stochastic geometry.
The report will be of interest to reactor physicists and designers.