J.M. Osborn, S.S. Chirayath “Assessment of Inverse Techniques Used in Forensics Attribution Analysis of Trace Nuclide Ratios in Weapons-Grade Plutonium”, 9th Annual DNDO Academic Research Initiative (ARI) Grantees Review Meeting, Atlanta, GA, USA, July 11-14, 2016.
A study is currently in progress at Texas A&M University on the computational and experimental methods for reliably predicting and measuring unique physical characteristics of separated weapons-grade plutonium produced by certain reactor types, specifically a Fast Breeder Reactor (FBR) and a Pressurized Heavy Water Reactor (PHWR). These reactor types will likely be operating in a non-safeguarded manner in some countries. Both the FBR and PHWR fuels produce weapons-grade plutonium when discharged at a low burnup of about 1 MWd/kg. We anticipated that the differences in neutron energy spectra as well as fission yield curves between the two reactor types would result in variations in isotopes of plutonium and fission products. The computational part of the project utilizes the MCNPX-2.7 radiation transport code to model the reactor cores, perform burnup cycles and estimate the resulting isotopics of the discharged fuel. The specific fission product and actinide ratios selected for analyses were chosen based upon several parameters including, the amount of isotope production, half-life, and probability of detection. A comparison of the PHWR results and FBR results was done to identify isotopic ratios with reactor dependencies. A suite of isotopic ratios were then selected which would be useful for accurate nuclear forensics attribution of the source reactor for interdicted weapons-grade plutonium. The second half of this study comprises the experimental analysis of irradiated fuel samples for comparison and verification of the computational results. Depleted uranium fuel samples were irradiated with a replicated fast neutron spectrum at the Oak Ridge National Lab-High Flux Isotope Reactor (ORNL-HFIR), and natural uranium fuel samples will be irradiated at the University of Missouri Research Reactor (MURR). The irradiated fuel samples are then subject to a lab scale PUREX process, and the product measured via NDA and DA methods. The current work assesses inverse analysis techniques for the attribution of a source reactor as well as reactor characteristics from the measured isotopic ratio data. The proposed techniques for source reactor attribution include a maximum likelihood calculation from a basic library comparison of parameters such as isotopic ratios, burnup, irradiation time, cooling time, etc., for various reactors, in addition to a methodology which works back to reactor parameters from resulting isotopics.
We would like to acknowledge the funding support from NSF and DHS joint Academic Research Initiative program (NSF Grant No. ECCS-1140018 and DHS contract number 2012-DN-077-ARI057-02&03) to carry out this research work.