Analysis of deficiencies in fast reactor blanket physics predictions
Abstract
This analysis addresses a deviation between experimental measurements and fast reactor blanket physics predictions. A review of worldwide results reveals that reaction rates in the blanket are underpredicted with the discrepancy increasing with penetration into the blanket. The analysis of this discrepancy involves two parts: quantifying possible error reductions using the most advanced methods and investigating deficiencies in current methodology. The source of these discrepancies was investigated by application of "state-of-the-art" group constant generation and flux prediction methodology to flux calculations for the Purdue University Fast Breeder Blanket Facility (FBBF). Refined group constant generation methods yielded a significant reduction in the blanket deviations; however, only about half of the discrepancy can be accounted for in this manner. Transport theory calculations were used to predict the blanket neutron transmission problem. The surprising result is that transport theory predictions utilizing diffusion theory group constants did not improve the blanket results. Transport theory predictions exhibited blanket underpredictions similar to the diffusion theory results. The residual blanket discrepancies not explained using advanced methods require a refinement of the theory. For this purpose an analysis of deficiencies in current methodology was performed. The major deficiency investigated is the treatment of transitional resonance spectra. To accurately account for the spatial streaming in resonance interference dips, an integral transport theory formalism was developed to analyze spatially dependent isolated resonance spectra. This theory was utilized to analyze transitional resonance spectra for isolated actinide resonances in the FBBF problem. Application has not been performed in a comprehensive manner. However, selective results indicate that deviations between the transitional integral transport theory treatment and the traditional approach are of a magnitude that could account for these discrepancies.
Degree
Ph.D.
Advisors
Clikeman, Purdue University.
Subject Area
Nuclear physics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.