Modeling Active Neutron Coincidence Counters with MCNPX

Improved Monte Carlo codes now allow for the direct modeling of coincidence count rates (singles, doubles, and triples) of neutron coincidence counters. The modeling can be used to develop calibration curves and bias correction factors. This modeling has already resulted in a greater understanding of the detector response for the passive multiplicity assay of plutonium items. New analysis algorithms developed with the Monte Carlo codes result in more accurate assay results. A similar study has been undertaken with active neutron coincidence counters. Active neutron counters for the assay of uranium are more complicated than passive multiplicity counting because they require the use of an interrogation neutron source to induce fission in the measurement sample. The latest version of the MCNPX (Monte Carlo N-Particle eXtended) code is capable of modeling the induced fission coincidence rate on uranium from an AmLi interrogation source. The MCNPX code was benchmarked to measurements made with an active well coincidence counter on bulk uranium. In addition, wide range of Monte Carlo calculations on several sets of idealized calibration standards was modeled. The preliminary results of these Monte Carlo calculations will be discussed with respect to the passive calibration and active multiplicity analysis methods.