Swinney, Mathew Wayne (2015-12). Experimental and Computational Assessment of Trace Nuclide Ratios in Weapons Grade Plutonium for Nuclear Forensics Analysis. Doctoral Dissertation.
Thesis
A terrorist attack using an improvised nuclear device is one of the most serious dangers facing the United States. The work presented here is an effort to improve nuclear deterrence by developing a methodology to attribute weapons-grade plutonium to a source reactor by measuring the intrinsic physical characteristics of the interdicted plutonium. The reactor source attribution methodology attempted here used measurements and analysis of plutonium samples (along with the fission-product contaminants) produced from depleted uranium dioxide samples irradiated in a fast neutron environment. In order to replicate the neutron flux in a fast-spectrum reactor and obtain experimental samples emulating weapons-grade plutonium produced in the blanket of a Fast Breeder Reactor (FBR), depleted uranium dioxide (DUO2) samples were placed in a gadolinium sheath and irradiated in the High-Flux Isotope Reactor (HFIR) at Oak Ridge National Laboratory (ORNL). Previous computational work on this topic identified several fission products that could be used to distinguish between reactor types, specifically; 137^Cs, 134^Cs, 154^Eu, 125^Sb, 144^Ce, 85^Rb, 147^Pm, and 150^Sm along with the Pu-vector. Simulations of the fast neutron irradiation of the DUO2 fuel samples in the HFIR were carried out using the radiation transport code, MCNPX. Comparisons of the predicted values of plutonium and fission-product concentrations to destructive and non-destructive assay measurements of neutron irradiated DUO2 samples are presented. The agreement between the predictions and the gamma spectroscopic measurements were within ~12% for 134^Cs, 137^Cs, 154^Eu and 144^Ce. Additional experimental results (mass spectroscopy) agreed to within 5% of the Monte-Carlo simulations for the following isotopes: 85^Rb, 147^Pm, 150^Sm, 154^Eu, 148^Nd, 144^Ce and 239^Pu. Results obtained from simulations of an Indian Prototype FBR and a Pressurized Heavy Water Reactor (PHWR) are also compared with the results from the HFIR simulation and experiment to demonstrate a methodology implementing a straightforward maximum likelihood calculation for attributing plutonium to a source reactor.
