List-mode Enabled In-situ Analysis of Optimized Gate Widths for Neutron Based Nondestructive Assay
Measurements to Maximize Throughput for Nuclear Material Accountancy

List mode enables the high-resolution acquisition of data from neutron based nondestructive measurements of nuclear material for control and accounting. The data is stored with complete fidelity in the spatial and temporal dimensions which allows iterative and sophisticated analysis techniques. A series of instruments, data acquisition electronics, software, and analysis are being developed to exploit the additional, complex information contained in list mode data. This work describes a list-mode enabled analysis technique to optimize a neutron instrument’s gate width setting by iteratively reanalyzing the neutron data while varying the parameter until convergence on a minimum measurement uncertainty is achieved. This analysis technique will significantly improve measurement throughput and statistical precision for almost every neutron coincidence or multiplicity measurement performed in plutonium and uranium nuclear facilities around the world. While the technique is complex, it can be implemented in software and automatically performed with no burden on the end user. The physics motivation relates to the fixed gate width commonly used in neutron coincidence and multiplicity counting. The gate width affects the fraction of real coincidences which are counted. Accidental coincidences contribute to statistical uncertainty and are, unavoidably, also counted. The amount of real and accidental coincidences contained in the gate both affect the measurement’s statistical uncertainty, which informs real-world measurement times implemented by equipment operators. Fixed gate widths are used in almost all applications. However, the optimal gate width depends on the accidentals rate, which can vary by a factor of 2,000, and the reals rate. List mode data collection allows the re-analysis of the measurement to identify the gate width that provides the smallest statistical uncertainty. The analysis can be performed for every measurement, so the increased performance is achieved even if the item stream results in a mix of low and high accidental coincidence rates. This capability may allow reduced measurement times for almost every coincidence or multiplicity counting application worldwide.