MINIMUM CRITICAL MASSES AND THEIR UNCERTAINTIES FOR SEVERAL FISSILE NUCLIDES AND REFLECTORS

摘要

A systematic effort was undertaken to reevaluate the minimum critical mass for several fissile nuclides and reflectors. This was done using MCNP6 in combination with several modern nuclear data libraries, viz. ENDF/B-VII. 1, JENDL-4.0 and JEFF-3.1.1. This combination was benchmarked extensively against criticality safety benchmarks. The procedure to extract the minimum critical mass from the MCNP6 calculations was carefully implemented, so that it would have a negligible impact on the accuracy of the end results. All parts of the calculation, i.e. nuclear data processing, MCNP6 models, and post-processing, were applied consistently for all cases, involving varying fissile nuclides and reflectors. Results for the minimum critical mass were generated for ~(233,235)U, ~(236,239,241)Pu, ~(242m)Am, ~(243,245,247)Cm, and ~(249,251)Cf, for the case of water moderation and water reflection. For ~(235)U, the minimum critical mass was also calculated for cases with other reflectors, such as beryllium, heavy water, graphite, lead, and concrete. The results for the water reflection cases are in good agreement with most of the literature values, but for the cases with reflectors such as graphite, lead, or concrete, there are sizable differences. The accuracy of the results was assessed in two ways. First, a comparison was made with criticality safety benchmark results, by selecting benchmarks that resemble the minimum mass systems the most. It was clear that for~(233,235) U and~(239) Pu with water reflection, the differences in minimum critical masses between the libraries are consistent with the differences obtained for benchmark cases. The difference between the benchmark calculations and the benchmark values is then an indication of the accuracy with which one can simulate comparable systems. Secondly, sensitivity runs were performed with the new MCNP6 sensitivity option, based on ENDF/B-VII. 1 nuclear data. For each fissile nuclide, the sensitivity of keff to cross sections such as capture, fission, elastic, inelastic was calculated for the minimum critical mass case. The same was done for the ~1H and ~(16)0 cross sections in the moderator and reflector. The uncertainty in keff was calculated by combining the sensitivity profiles with the appropriate ENDF/B-VII. 1 covariance matrix, using the SUSD code. As a final step, the uncertainty in k_(eff) can be translated to an uncertainty in the minimum critical mass.