Active interrogation of cargo containers employing monoenergetic photons offers an expeditious low-dose approach in pursuit of shielded special nuclear materials. Cherenkov detectors can be used for a variety of purposes, including imaging of the cargo contingent upon gamma ray energies used in interrogation. If the gamma ray energies are sufficiently well separated, as the case in $^{11}$B(d,n-$\gamma$)$^{12}$C reaction resulting in 4.4 MeV and 15.1 MeV photons, spectral analysis using Cherenkov detectors is possible. Applying an array of custom designed Cherenkov detectors reduce undesired low energy background, both natural and scatter from the application environment, while producing the capability of high contrast transmission imaging. Spectral analysis of the image can be used to confirm the presence of high-Z materials by analyzing the relative transmission of the two main energies emitted by exploiting the large difference in Compton Scatter and pair production cross sections. These detectors require a special approach to design and energy calibration due to the lack of resolution in order for spectral analysis to take place. This thesis addresses the design and utility of Cherenkov detectors for active interrogation with monoenergetic photons as well as the results of computational and experimental studies of their energy calibration and application to imaging with material identification.

Detection of special nuclear material (SNM) being smuggled into the US through ports of entry has been identified as a crucial capability for ensuring the safety and security of the US from radiological threats. Programs such as the NNSA's Second Line of Defense aim to deploy detection systems, both domestically and abroad, in an attempt to interdict the SNM before it reaches its destination. Active interrogation (AI) is a technique that relies on the detection of emitted particles which are produced when SNM is bombarded with a source of high energy photons or neutrons. This work presents a general framework that allows for fast radiation transport modeling of AI scenarios by generating families of response functions which depict neutron, gamma, or electron radiation exiting various regions within the problem, per unit source of radiation entering the region. The solution for a given scenario, typically the detector count rate, is computed by injecting a source term into the first region and applying the appropriate response functions, in sequence, for each subsequent region. For the AI systems modeled in this work, the source is an electron beam in a linear accelerator. Subsequent response functions create and transport bremsstrahlung photons into the SNM, and transport neutrons born in the problem to a detector. The computed solution is comparable to that of a full Monte Carlo simulation, but is assembled in orders of magnitude less time from pre-computed response function libraries. The ability to rapidly compute detector spectra for complicated AI scenarios opens up research and analysis possibilities not previously possible, including conducting parametric studies of scenarios spanning a large portion of the threat space and generating detector spectra used for conditioning and testing of alarm algorithms. / text

<div>Detecting hidden/smuggled special nuclear materials (SNM) is one of the unsolved problems in the safeguards industry. The biggest challenge is to quantify and track SNM and prevent the use of these materials for illicit purposes. The goal is to detect smallest quantity of SNM in large cargo containers, at the ports of entry, in the shortest amount of measurement time. Currently passive detection techniques, which is based on long-lived isotopes, are used to detect hidden SNM. This technique is not very reliable, as appropriate shielding of the SNM can reduce detection signals from these long-lived isotopes. Accelerator based active interrogation methods are proposed to solve the SNM problem. Besides SNM, another challenge in the nuclear industry is to meet the demand and supply of medical radioisotopes, particularly Tc-99m (half-life 6 hours). Mo-99, which decays to Tc-99m, is one of the fission products found in nuclear reactors. Because of short half-life of 66 hours, Mo-99 cannot be stockpiled. The shutdown of various research reactors globally disrupted the supply of Mo-99. Because of the financial and regulatory burden on the nuclear reactors, accelerator-based systems can be used to produce Mo-99.</div><div><br></div><div>With the aim to solve these two major challenges, a preliminary study is done to understand the neutrons behavior on milliseconds (or shorter) time steps in an accelerator driven subcritical system. A pulsed external neutron source, i.e. Deuterium-Deuterium (DD) generator, drives the assembly. Using first principles, the transient equations are derived and the neutron population at different time stamps is calculated. The Li-6 detector’s response to the neutron population is predicted. Experiments are performed to compare the predicted behavior with the observed behavior. The model is extended further to investigate the case of no uranium fuel inside the system. Transient measurements, in the absence of the uranium fuel, are made and the neutron die-away time is determined. This die-away time is compared with the predicted time.</div>

This thesis explores the inverse problem of optical tomography applied to two-dimensional heterogeneous domains. The neutral particle transport equation was used as the forward model to simulate how neutral particles stream through and interact within these heterogeneous domains. A constrained optimization technique that uses Newton's method served as the basis of the inverse problem. The capabilities and limitations of the presented method were explored through various two-dimensional domains. The major factors that influenced the ability of the optimization method to reconstruct the cross sections of these domains included the locations of the sources used to illuminate the domains, the number of separate experiments used in the reconstruction, the locations where measurements were collected, the optical thickness of the domain, the amount of signal noise and signal bias applied to the measurements, and the initial guess for the cross section distribution. All of these factors were explored for problems with and without scattering. Increasing the number of sources, measurements and experiments used in the reconstruction generally produced more successful reconstructions with less error. Using more sources, experiments and measurements also allowed for optically thicker domains to be reconstructed. The maximum optical thickness that could be reconstructed with this method was ten mean free paths for pure absorber domains and two mean free paths for domains with scattering. Applying signal noise and signal bias to the measured fluxes produced more error in the reconstructed image. Generally, Newton's method was more successful at reconstructing domains from an initial guess for the cross sections that was greater in magnitude than their true values than from an initial guess that was lower in magnitude.

No description available.