MODELING AND VALIDATION OF NEUTRON ACTIVATION AND GAMMA-RAY SPECTROSCOPY MEASUREMENTS AS AN EXPLORATORY TOOL FOR NUCLEAR FORENSIC ANALYSIS

Abstract

The continued success of nuclear forensic analysis relies on the development of new material and process signatures. However, the unique safety hazards and strict controls concerning nuclear materials and operations limit the practicality of experimental scenarios. To bypass these limitations, the nuclear science community is increasingly reliant on simulation-based tools. In this dissertation, neutron activation and gamma-ray spectroscopy measurements are simulated to explore the activation network of stainless steel and its components using two neutron sources. The goal is to identify nuclides or ratios that are indicative of the neutron source and test their measurability in complex samples. The neutron sources are a critical assembly, providing fission spectrum neutrons, and a beryllium (Be) neutron converter, producing neutrons through various deuteron induced reactions. Simulated neutron energy distributions are calculated using the Monte Carlo N-Particle (MCNP) radiation transport code.

Neutron activation has an inherent neutron energy dependence, making nuclide production rates contingent on the neutron energy distribution. Activation calculations performed by hand and with the FISPACT-II code are compared against experiments to validate the neutron energy distributions and assess available reaction cross-section data. Additionally, ratios of activation products common to both neutron sources are investigated to determine if they are indicative of the neutron source.

Gamma-ray spectroscopy with high-purity germanium (HPGe) detectors is the leading passive assay technique for radioactive samples, providing detailed qualitative and quantitative information while preserving sample integrity. A simple HPGe detector is modeled using MCNP to assess the measurability of different activation product ratios. The HPGe model is validated against its real counterpart to determine if the level of complexity is sufficient for this work.

Activation calculations were able to validate the critical assembly neutron energy distribution but showed significant errors in the Be converter model. Additionally, validation of activation calculations identified shortcomings in the 60Ni(n,p)60Co reaction cross section. Absent interferences, HPGe simulation performance was equivalent to the real detector. The HPGe model also showed that decay time can affect measurement accuracy when significant interferences are present. Activation product ratios identified in this work that are indicative of the neutron source are 57Co/54Mn, 51Cr/54Mn, 57Co/59Fe, and 51Cr/59Fe.