SUMMARY

In order to improve the efficiency of dry used nuclear fuel (UNF) storage, reduced reactivity methods are being developed for various reactor types and operating conditions. Sub-criticality must be maintained in the storage configuration and conservative computer simulations are used as the primary basis for loading the storage casks. Methodologies are now being developed to reduce the amount of modeling and computation in order to make conservative assessments of how densely fuel can be packed. The SCALE/TRITON (Standardized Computer Analyses for Licensing Evaluation) code system has been used to simulate boiling water reactor (BWR) operating conditions in order to investigate nuclide densities in UNF and how the use of control rod blades affect nuclide densities found in UNF. Rodded and unrodded operating cases for a fuel assembly have been used as bounding cases and are used as reference solutions in a piece wise data approximation methodology (PDA method). A variety of control blade insertion patterns have been used with the PDA method and simulated in order to observe trends in nuclide densities with varying control blade use. The PDA method is compared with TRITON simulated data in order to evaluate the validity and accuracy of the methodology. The PDA method gives very accurate results for fissile nuclides but is insufficient in treating densities as a function of burnup for fission products and fertile nuclides. Predicting nuclide densities from high frequency control blade insertion and withdrawal patterns is also a strength of the PDA method. The PDA method, however, is not capable of properly accounting for neutron spectral shifts and the behavior in nuclide densities brought about by the spectral shift or nuclide density saturation. Observing the causes for the shortcomings in the PDA method, a more robust methodology can be developed.