Verification of Plutonium content in Spent Fuel Assemblies Using Neutron Self- Interrogation

The large amounts of plutonium in reactor spent fuel assemblies has led to increased research directed toward the measurement of the plutonium for safeguards verification. The high levels of fission product gamma-ray activity and curium neutron backgrounds have made the plutonium measurement difficult. We have developed a new technique that can directly measure both the 235U concentration and the plutonium fissile concentration using the intrinsic neutron emission from the curium in the fuel assembly. The passive neutron albedo reactivity (PNAR) method has been described previously [1] where the curium neutrons are moderated in the surrounding water and reflect back into the fuel assembly to induce fissions in the fissile material in the assembly. The cadmium (Cd) ratio is used to separate the spontaneous fission source neutrons from the reflected thermal neutron fission reactions. This method can measure the sum of the 235U and the plutonium fissile mass, but not the separate components. Our new differential die-away selfinterrogation method (DDSI) can be used to separate the 235U from the 239Pu. The method has been applied to both fuel rods and full assemblies. For fuel rods, the epi-thermal neutron reflection method filters the reflected neutrons through thin Cd filters so that the reflected neutrons are from the epi-cadmium energy region. The neutron fission energy response in the epi-cadmium region is distinctly different for 235U and 239Pu. We are able to measure the difference between 235U and 239Pu by sampling the neutron induced fission rate as a function of time and multiplicity after the initial fission neutron is detected. We measure the neutron fission rate using list-mode data collection that stores the time correlations between all of the counts. The computer software can select from the data base the time correlations that include singles, doubles, and triples. The die-away time for the doubles distribution is distinctly different for 235U and 239Pu. The 239Pu has a higher fission cross-section in the epi-cadmium neutron region and larger induced fission moments than 235U, so the measured die-away time can provide the relative amounts of 239Pu and 235U. This paper will present the Monte Carlo simulations for the detector and sample configurations for both fuel pins and full fuel assemblies.