Impact of nuclear data uncertainties on criticality calculations of the very high temperature reactor critical assembly benchmark

Within the framework of the IAEA Coordinated Research Project on HTGR Uncertainty Analysis in Modeling, uncertainty and sensitivity analyses of the Very High Temperature Critical Assembly experiment were performed as the validation reference to the prismatic MHTGR-350 lattice calculations. Criticality measurements performed at several temperature points at this Japanese graphite-moderated facility were recently included in the International Handbook of Evaluated Reactor Physics Benchmark Experiments, and represent one of the few data sets available for the validation of HTGR lattice physics. Nominal multi-group criticality calculations with the KENO-VI Monte Carlo code of the SCALE 6.1.2 code package were compared with the continuous-energy Monte Carlo Code Serpent. Good agreement with a maximum difference of 150 pcm was obtained. The experimental data set, however, differs by several hundred pcm to the results of both codes when using the ENDF-VII.O library. Uncertainties of the VHTRC multiplication factors due to uncertainties in nuclear data were determined with the GRS code XSUSA in combination with KENO-VI. For all core states, the obtained uncertainty was found to be 0.58%. The top contributor to this uncertainty is the average number of neutrons per fission event of U-235. When considering these uncertainties, the XSUSA/KENO-VI calculations show overlapping uncertainty intervals with the experiment.