Variations in AmLi source spectra and their estimation utilizing the 5 Ring Multiplicity Counter

Active-mode assay systems are widely used for the safeguards of uranium items to verify compliance with the Non-Proliferation Treaty. Systems such as the Active-Well Coincidence Counter (AWCC) and the Uranium Neutron Coincidence Collar (UNCL) use americium-lithium (AmLi) neutron sources to induce fissions which are measured to determine the sample mass. These systems have historically relied on calibrations derived from well-defined standards. Recently, restricted access to standards or more difficult measurements have resulted in a reliance on modeling and simulation for the calibration of systems, which introduces potential simulation biases. The AmLi source energy spectra commonly used in the safeguards community do not accurately represent measurement results and the spectrum uncertainty can represent a large contribution to the total modeling uncertainty in active-mode systems.

The 5-Ring Multiplicity Counter (5RMC) has been used to measure 17 AmLi sources. The measurements showed a significant spectral variation between different sources. Utilization of a spectrum that is specific to an individual source or a series of sources will give improved results over historical general spectra when modeling AmLi sources. Candidate AmLi neutron spectra were calculated in MCNP and SOURCES4C for a range of physical AmLi characteristics. The measurement and simulation data were used to fit reliable and accurate AmLi spectra for use in the simulation of active-mode systems. Spectra were created for average Gammatron C, Gammatron N, and MRC series sources, and for individual sources. The systematic uncertainty introduced by physical aspects of the AmLi source were characterized through simulations. The accuracy of spectra from the literature was compared.