Propagation of Nuclear Data Uncertainties for ELECTRA Burn-up Calculations

Abstract

The European Lead-Cooled Training Reactor (ELECTRA) has been proposed as a training reactor for fast systems within the Swedish nuclear program. It is a low-power fast reactor cooled by pure liquid lead. In this work, we propagate the uncertainties in 239Pu transport data to uncertainties in the fuel inventory of ELECTRA during the reactor lifetime using the Total Monte Carlo approach (TMC). Within the TENDL project, nuclear models input parameters were randomized within their uncertainties and 740 239Pu nuclear data libraries were generated. These libraries are used as inputs to reactor codes, in our case SERPENT, to perform uncertainty analysis of nuclear reactor inventory during burn-up. The uncertainty in the inventory determines uncertainties in: the long-term radio-toxicity, the decay heat, the evolution of reactivity parameters, gas pressure and volatile fission product content. In this work, a methodology called fast TMC is utilized, which reduces the overall calculation time. The uncertainty of some minor actinides were observed to be rather large and therefore their impact on multiple recycling should be investigated further. It was also found that, criticality benchmarks can be used to reduce inventory uncertainties due to nuclear data. Further studies are needed to include fission yield uncertainties, more isotopes, and a larger set of benchmarks.