Validation of the SCALE System for PWR Spent Fuel Isotopic Composition Analyses

The validity of the computation of pressurized-water-reactor (PWR) spent fuel isotopic
composition by the SCALE system depletion analysis was assessed using data presented in the report.
Radiochemical measurements and SCALE/SAS2H computations of depleted fuel isotopics were
compared with 19 benchmark-problem samples from Calvert Cliffs Unit 1, H. B. Robinson Unit 2,
and Obrigheim PWRs. Even though not exhaustive in scope, the validation included comparison of
predicted and measured concentrations for 14 actinides and 37 fission and activation products.
The basic method by which the SAS2H control module applies the neutron transport treatment
and point-depletion methods of SCALE functional modules (XSDRNPM-S, NITAWL-II, BONAMI,
and ORIGEN-S) is described in the report. Also, the reactor fuel design data, the operating histories,
and the isotopic measurements for all cases are included in detail. The underlying radiochemical
assays were conducted by the Materials Characterization Center at Pacific Northwest Laboratory as
part of the Approved Testing Material program and by four different laboratories in Europe on
samples processed at the Karlsruhe Reprocessing Plant.
Comparisons are given in terms of percentage differences of computed minus measured
compositions of the fuel. The SCALE depletion analyses for all cases applied two different crosssection
libraries. One was the 27-energy-group SCALE-4 library, which has light-element and
actinide data processed from ENDF/B-IV and fission-product data processed from ENDF/B-V. The
second library applied was a 44-energy-group library derived entirely from the latest ENDF/B-V files
with the exception that data for 16O, 154Eu, and 155Eu were taken from ENDF/B-VI. Almost all the
total average percentage differences for the actinide isotopes and for the isobaric mass values 154 and
155 were significantly better in cases using the latter library, whereas other fission-product
comparisons with measurements were essentially the same in using either library. The final fuel rod
and batch average differences for nuclides 235U and 239Pu were –3.1 and 5.4% for the former library
and –2.2 and –0.4% for the latter library, respectively. Except for 237Np, only 242Cm and 244Cm had
excessive differences (>15%); these isotopes had similarly large uncertainties in their measurements.
The percentage differences exceeded 15% for fission products 99Tc, 126Sn, 148Sm, 149Sm, 151Sm +
151Eu, 152Sm, 154Sm + 154Eu + 154Gd (for the former library only) and 155Eu + 155Gd. Although the
spread of the percentage differences for all cases of each nuclide were large (particularly for
comparisons of pellet sample results), reasons why the pellet-location-dependence of reactor flux
affect significant nuclide reaction rates were briefly investigated and discussed. It was concluded by
the authors that the SCALE depletion analysis properly qualifies as a basic tool for predicting isotopic
compositions of spent fuel from PWR power plants. Also, it was evident that cross sections of
actinide isotopes processed from ENDF/B-V were superior to those taken from ENDF/B-IV.