Computational Evaluation of Neutron Multiplicity Measurements of Polyethylene-Reflected Plutonium Metal

Simulations of neutron multiplicity measurements of a highly multiplicative plutonium sphere measured with a moderated array of ³He proportional counters have consistently overpredicted the mean and variance of the measured multiplicity distribution. In contrast, identical experiments using a ²⁵²Cf source have been accurately simulated. This paper outlines a sensitivity analysis of several key parameters that could account for the overprediction in the simulation of the plutonium sphere. Parameters that were analyzed include source-detector distance, detector dead time, variations in density and volume of the plutonium, and the value of  for v̅ ²³⁹Pu-induced fission. Of these parameters, the only factor that accounted for the overprediction within reasonable bounds was a change in the value of the ²³⁹Pu  v̅. The sensitivity analysis showed that a small change (1.14% reduction) in the value of v̅ dramatically improved the simulated results.