Michael C. Hamel

Dual-particle imager for standoff detection of special nuclear material

An advanced dual-particle imaging system is being developed for standoff, passive detection of special nuclear material. This system consists of three detector planes and will be capable of imaging both photons and fast neutrons. The ability of the system to detect fast neutrons makes it more difficult to effectively shield a threat source. This feature has an advantage over the commonly used Compton-camera systems, which are only sensitive to photons. Additionally, the detection of fast neutrons will allow for increased performance in regions with high levels of photon background radiation. The first two planes of the system consist of EJ-309 liquid scintillators and the third plane consists of NaI scintillators. This detector/plane combination allows image reconstruction using both neutrons and photons. In the liquid scintillators, neutron interactions are distinguished from photon interactions using an optimized pulse shape discrimination technique. The Monte Carlo transport code MCNPX-PoliMi has been used for the initial studies of this system due to its ability to track detailed information on interactions of interest and time-correlated particle production. This information has been used to optimize system parameters and has also allowed for investigation of image reconstruction techniques including simple backprojection and maximum likelihood expectation maximization (MLEM). A small-scale prototype is being developed for testing and validation of the simulations. This paper will analyze preliminary measurements and will also discuss simulations of several shielded source scenarios.

Active neutron and gamma-ray imaging of highly enriched uranium for treaty verification

The detection and characterization of highly enriched uranium (HEU) presents a large challenge in the non-proliferation field. HEU has a low neutron emission rate and most gamma rays are low energy and easily shielded. To address this challenge, an instrument known as the dual-particle imager (DPI) was used with a portable deuterium-tritium (DT) neutron generator to detect neutrons and gamma rays from induced fission in HEU. We evaluated system response using a 13.7-kg HEU sphere in several configurations with no moderation, high-density polyethylene (HDPE) moderation, and tungsten moderation. A hollow tungsten sphere was interrogated to evaluate the response to a possible hoax item. First, localization capabilities were demonstrated by reconstructing neutron and gamma-ray images. Once localized, additional properties such as fast neutron energy spectra and time-dependent neutron count rates were attributed to the items. For the interrogated configurations containing HEU, the reconstructed neutron spectra resembled Watt spectra, which gave confidence that the interrogated items were undergoing induced fission. The time-dependent neutron count rate was also compared for each configuration and shown to be dependent on the neutron multiplication of the item. This result showed that the DPI is a viable tool for localizing and confirming fissile mass and multiplication.

Digital data acquisition and processing for a neutron-gamma-ray imaging system

A digital, data-acquisition system for use with a large number of detectors was set up at the University of Michigan. Fast waveform digitizers from CAEN Technologies with 8 channels were synchronized to create a fully scalable system, with a current set-up of 32 channels. While some of the systems limitations are still being investigated, the excellent time resolution of the system enabled accurate time-of-flight measurements at the Los Alamos Neutron Science Center (LANSCE).

Image reconstruction of shielded mixed-oxide fuel using a dual-particle imaging system

The dual-particle imaging system being developed at the University of Michigan was used at the Joint Research Centre in Ispra, Italy for measurements on samples of special nuclear material. A 1,150-g mixed-oxide (MOX) fuel sample was measured with various shielding configurations to determine how the presence of lead and/or polyethylene shielding degrades the systems ability to localize a source by simultaneous neutron and photon imaging. Three two-hour measurements were taken with the source shielded by an: a) 8-mm lead sheath and 5.1-cm lead bricks, b) 8-mm lead sheath and 6.5-cm polyethylene bricks, and c) 8-mm lead sheath, 5.1-cm lead bricks, and 6.5-cm polyethylene bricks. The 8-mm lead sheath was used in all cases to reduce the measured photon count rate. A “bare” measurement was also made by using only the 8-mm lead sheath, but the unexpected presence of additional sources has rendered the measurement unsuitable for comparison. The resulting images show that the dual-particle imaging system is able to accurately localize the MOX canister in the presence of intervening material.

Time-of-flight neutron spectrum unfolding for mixed-oxide nuclear fuel and plutonium metal using a dual-particle imager

A dual-particle imager with sensitivity to both neutrons and photons has been developed for the detection of special nuclear material. This system has the capability to provide spectral information on detected sources that is useful for safeguard applications. In nuclear facilities with high photon background rates, neutron detection can aid in identification and verification capabilities. The dual-particle imager creates a reconstructed neutron spectrum by correlating neutron counts in pairs of liquid scintillators. The reconstructed spectrum, a combination of energy deposited and time-of-flight, contains resolution effects that blur the true, emitted source-spectrum. To reduce the impact of resolution effects, the reconstructed spectrum can be unfolded to create an estimated source spectrum. A statistical technique, maximum-likelihood expectation-maximization, has been shown to produce an unfolded result for Cf-252 and Am-Be. This technique requires a system response-matrix that contains reconstructed spectra for single neutron energies. Since measuring mono-energetic neutrons is difficult, the system-response-matrix is simulated. Monte Carlo modeling with MCNPX-PoliMi has been show to produce an accurate system response. This work expands on previous neutron unfolding results by showing results for mixed-oxide fuel and plutonium metal measured with the dual-particle imager. These results also demonstrate the ability of the dual-particle imager to discrimination between spontaneous fission neutrons and (α,n) reaction neutrons.

Large-scale Compton-camera simulations, validation experiments, and image reconstruction

Recent efforts in nonproliferation and homeland security areas have focused on designing systems to accurately detect and locate radioactive material. This task is especially challenging when trying to locate material at large distances. One solution that has been under investigation is the Compton camera. In this work, the performance of a large-scale, two-plane Compton camera is investigated using different imaging reconstruction methods on simulated data. The simulation methodology is being validated by measurements conducted using a small-scale prototype of the system in the laboratory environment. The geometry consists of two planar arrays of scintillation detectors.

Image reconstruction using a three-plane, dual-particle imager for standoff detection of special nuclear material

An advanced, dual-particle, imaging system is being developed for standoff, passive detection of special nuclear material. This system consists of three planes of detectors and will be capable of imaging photons and fast neutrons simultaneously. The first two planes of the system consist of EJ309 liquid scintillators, while the third plane consists of NaI detectors. The EJ-309 detectors have excellent pulse shape discrimination capabilities, which allows the imaging system to distinguish incident photons from neutrons. Preliminary investigations of the dual-particle imager focused primarily on the use of simple-backprojection methods for image reconstruction. However, efforts have been made toward implementing the use of a maximum-likelihood expectation-maximization algorithm for image reconstruction. MCNPX-PoliMi has been used to simulate system response matrices for the dual-particle imager. Results have been promising for image reconstruction using measurement-specific system response matrices. However, field use requires the ability to detect and locate unknown sources in unknown shielding configurations. This has resulted in the development of an algorithm capable of generating a measurement-specific system response matrix on the fly.

Author Correction: Active neutron and gamma-ray imaging of highly enriched uranium for treaty verification

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

Stochastic image reconstruction for non-proliferation applications

The stochastic origin ensembles (SOE) method has been implemented for the reconstruction of neutron and gamma-ray images from a dual-particle imager (DPI). SOE image reconstruction is an iterative stochastic process that creates images using Markov-chain Monte-Carlo sampling with event origins from the reconstructed backprojection cones. This method has been shown to produce image quality comparable to that of the widely implemented maximum-likelihood expectation-maximization (MLEM) algorithm. SOE can be advantageous because the method is simple to implement and thorough definition of spatial efficiency or resolution is not required. The SOE algorithm also reaches a steady-state solution after iterating, as opposed to MLEM solutions, which can degrade with over-iteration. This paper shows photon and neutron images images reconstructed using the SOE algorithm for an experiment with multiple neutron and gamma-ray sources.

Simulations of the cosmic-ray-induced neutron background

A method for simulating the cosmic-ray-induced neutron background at ground level has been developed. This method was used to obtain the expected pulse height distributions, net neutron rates, and angular distribution using EJ-309 liquid scintillators, and has been validated using outdoor measurements. The simulations were carried out using the Monte Carlo transport code MCNPX-PoliMi combined with the source subroutine CRY. The absolute count rates for neutrons were then obtained from the simulations using an empirical scaling formula from literature. Preliminary measurements of the angular distribution of background neutrons were done using a collimated EJ-309 liquid scintillator, showing fair agreement with simulations.