Impact of Missing Resonance Levels on Benchmark Calculations

Nuclear data evaluations available in existing nuclear data libraries are derived based on experimental differential nuclear data, often measured at room temperature, that is, 293.6 K. Very seldom are experimental data available for temperatures below or above room temperatures. Nuclear data measurements and evaluation needs have been driven mainly for reactor, shielding, and criticality safety applications. As a result, most of the data evaluations in the nuclear data libraries are expected to be used for temperatures from room temperature and above. Recently, there has been a demand for nuclear data for low temperatures, that is, below room temperature, for criticality safety applications. The approach used to address the need for data at temperatures other than those for which the data were evaluated is to perform calculations based on extrapolating the existing room temperature data to the desired temperatures. Although this is an acceptable practice, however, care should be taken to understand whether the validity of the nuclear data can be extended to temperatures other than those for which the data were evaluated. The question can be linked to the experimental data resolution as to whether resonances in a neutron-nucleus interaction can be fully identified. Therefore, the subject related to the impact of loss of resonance needs to be evaluated for practical applications. For instance, are the energy resonance self-shielding effects well addressed? How much a system multiplication factor, keff, will be affected due to missing resonances? The objective of this work is to investigate the issue in connection to the impact of resonance missing levels in benchmark calculations, that is, the impact on keff results. Simulated resonance parameters in the resolved resonance region were generated for ²³⁵U based on resonance parameters statistical distributions. A resonance parameter set including all resonances in the energy region from thermal to 2250 keV was generated. This set of resonance parameters can be regarded as the true resonance with no missing levels. The next step consisted of generating a resonance parameter set with fewer resonances (missing levels) that reproduced the characteristics of the complete resonance parameter by fitting of simulated “experimental” differential data generated with the true resonance parameters. The benchmark testings have been done based on benchmark included in the International Criticality Safety Benchmark Evaluation Project (ICSBEP) Handbook. Four thermal benchmarks, named as HEU-SOL-THERM-013 (Cases 1-4), and four intermediate energy benchmarks named HEU-MET-INTER-006 have been considered. Both sets of resonance parameters the true set and that from the fitting of the simulated “experimental” data were used in the benchmark calculations, and the corresponding results are presented.