Quantification of the Uncertainty on the Mass Estimation in NMC Due to the Uncertainty on the Fission Multiplicity Distribution

Neutron multiplicity counting, aimed to measure the mass of spontaneous fissile material in the sample by measuring the spontaneous neutron yield is becoming a standard tool in nuclear materials control and accountability. In response, we see growing volume of studies, both academic and technological, aimed to quantify the uncertainty of neutron multiplicity counting. In the present study, we fully analyze the uncertainty in the mass estimation due to the uncertainty on the factorial moments of neutron fission yield multiplicity number in 240Pu and 239Pu. In standard practice of neutron multiplicity counting, the first three moments of the detection count distribution are used in an inversion formula to compute the magnitude of the three neutron sources at hand: the spontaneous fission source, the (alpha,n) source and the leakage multiplication factor (which accounts for induced fissions in the sample). Since the factorial moments of the fission neutrons multiplicity distribution appears in the inversion formula, any uncertainty on the neutron multiplicity moments is bound to create a systematic uncertainty on the mass estimation. However, since the inversion formula is non-linear, the error propagation from the multiplicity numbers to the mass is also non-linear, and might have a very strong dependence on the sample parameters. In the study, we first formulate the mathematical formulas that describe the error propagation from the factorial moments of the multiplicity numbers to the mass, and then use the formulas to quantify the uncertainty on the mass in terms of the sample characteristics. For validation, the computational results are then compared with experimental results.