Nuclear data evaluation for decay heat analysis of spent nuclear fuel over 1–100 k year timescale

Abstract

Accurate nuclear data are essential in the evaluation of decay heat from spent nuclear fuel (SNF). The accuracy of such data was assessed using an approach that compares values reported in different evaluated libraries and determines whether discrepancies reflect inaccuracies in primary data. A short list of 43 isotopes which are most significant to SNF decay heat calculations over 1–100 k years was produced by combining generic reactor inventory code with decay heat analysis for undifferentiated SNF. Decay properties (half-lives and decay energies) and neutron interactions (cross section and fission yields) were compared from 6 evaluated libraries. Fission product (FP) discrepancies identified are 90Sr half-life, where inclusion of a single measurement significantly reduces the evaluated value; 95mNb beta energy, where DDEP evaluation omits the decay to the 95Mo ground state; 99Tc beta energy, where evaluations differ by approximately 10% with a variety of shape factors used; 126Sb/126mSb beta (JEF2.2/3.1.1/3.3) and electron energies (JEFF3.1.1), where intensity differences are reported; and 137Cs beta energy, where ENDF/B-VIII.0 and JEF3.3 evaluations use incorrect shape factors. For actinides, the major discrepancies identified were 237Np alpha energy (JEF2.2/3.1.1) and 225Ac electron energies (ENDF/B-VIII.0) but overall show less discrepancies during long-term disposal (0.1–100 ky) compared to FP’s during interim storage (1–100 years). Further assessments of the 90Sr half-life and the best shape factor for the 99Tc beta decay are needed to improve future decay heat analyses, which are important for designing future stores and evaluating schemes for possible heat recovery.

Assay data of spent nuclear fuel: the lab-work behind the numbers

“…All decay and in-growth calculations for the timespan between measurement date and EOL reference date make use of nuclear data. For consistency, users of the assay data should check whether their set of nuclear data correspond to the nuclear data used for the EOL calculations as different libraries may contain different nuclear data (Doran et al, 2022). Just to give an example: the half-life of Cm-244 in the ENDF/B-VIII.0 library (18.11 ± 0.03 years) is 0.6% higher than the one in the JEFF3.3 library (18.00 ± 0.10 years).…”

Note on the potential to increase the accuracy of source term calculations for spent nuclear fuel

“…This perspective conservatively assumes that the observed C/E spread is indicative of any future, possible variation between a calculated and real nuclide concentration, and that unaccounted sources of error exist. For example, in (Doran et al, 2022) it was concluded that the 137 Cs mean beta energies reported by JEFF3.3., and ENDF/ B-VIII.0 are in error. Also, newer evaluations of nuclear data do not necessarily lead to higher quality.…”

An absolute measurement of the neutron production rate of a spent nuclear fuel sample used for depletion code validation

“…A maximum difference of almost 1.5% is noticed for the results obtained with JEFF-3.1.2. Differences in nuclear data relevant for the estimation of the decay heat of SNF between the main libraries (JEFF, ENDF/B) and recommended decay libraries (ENSDF, DDEP) are also discussed by Doran et al (2022). 4 Neutron detection system for SNF sample measurements…”