Nuclear Materials Identification Research Articles

Investigation of neutron/gamma-ray distribution in SiPM-based pulse shape discrimination using EJ-276 plastic scintillators

The distribution of neutrons and gamma-rays in mixed fields is a main interest in pulse shape discrimination (PSD) studies, and becomes more significant when it involves techniques such as machine learning which requires true data. One of the surest methods to label neutrons and gamma-rays is neutron Time-of-Flight. This can be applied to the identification of special nuclear materials and by extension to nuclear non-proliferation. Moreover, the integration of silicon photomultiplier (SiPM) technology can make the neutron Time-of-Flight system more compact. Two EJ-276 plastic scintillators, which are capable of PSD, were coupled to SiPMs. The actual portion of neutrons from a 252Cf source intruding the gamma-rays in PSD ratio distribution, labeled by neutron Time-of-Flight, was calculated and then compared with gamma-rejection ratio (GRR) for each threshold in SiPM-based pulse shape discrimination.

Networked Sensing for Radiation Detection, Localization, and Tracking

The detection, identification, and localization of illicit radiological and nuclear material continue to be key components of nuclear non-proliferation and nuclear security efforts around the world. Networks of radiation detectors deployed at strategic locations in urban environments have the potential to provide continuous radiological/nuclear (R/N) surveillance and provide high probabilities of intercepting threat sources. The integration of contextual information from sensors such as video, Lidar, and meteorological sensors can provide significantly enhanced situational awareness, and improved detection and localization performance through the fusion of the radiological and contextual data. In this work, we present details of our work to establish a city-scale multi-sensor network testbed for intelligent, adaptive R/N detection in urban environments, and develop new techniques that enable city-scale source detection, localization, and tracking.

ISOTOPIC IDENTIFICATION OF PHOTOFISSED NUCLEAR MATERIALS IN STAINLESS STEEL CONTAINERS USING DELAYED GAMMA-RAYS

The results of isotope identification of nuclear materials (232Th, 235U, 238U, 239Pu) packed in hermetic stainless steel containers by their stimulated delayed gamma radiation are presented. The ratios of the values of the time dependences of the intensity of gamma radiation of the light and heavy product of their photofission were used for the analysis. Gamma radiation from photofission products 89Rb (1031.9; 1248.1 keV) and 138Cs (1009.8; 1436.8; 2218.0; 2639.6 keV) was used to differentiate nuclear materials. The photofission reaction the samples of nuclear materials was stimulated on an electron accelerator, an M-30 microtron, at maximum bremsstrahlung energy of 12.5 MeV. The possibility of carrying out isotopic identification of nuclear materials using the following sets of ratios of the values of the time dependences of the gamma-line intensity has been demonstrated: X(1031.9)/X(2639.6), X(1248.1)/X(1009.8), X(1248.1)/X(1436.8), X(1248.1)/X(2218.0), X(1248.1)/X(2639.6).

Explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping

Radionuclide identification is an important part of the nuclear material identification system. The development of artificial intelligence and machine learning has made nuclide identification rapid and automatic. However, many methods directly use existing deep learning models to analyze the gamma-ray spectrum, which lacks interpretability for researchers. This study proposes an explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping. This method shows the area of interest of the neural network on the gamma-ray spectrum by generating a class activation map. We analyzed the class activation map of the gamma-ray spectrum of different types, different gross counts, and different signal-to-noise ratios. The results show that the convolutional neural network attempted to learn the relationship between the input gamma-ray spectrum and the nuclide type, and could identify the nuclide based on the photoelectric peak and Compton edge. Furthermore, the results explain why the neural network could identify gamma-ray spectra with low counts and low signal-to-noise ratios. Thus, the findings improve researchers’ confidence in the ability of neural networks to identify nuclides and promote the application of artificial intelligence methods in the field of nuclide identification.

Development of a CsI(Tl) scintillator based gamma probe for the identification of nuclear materials in unknown areas

In case of nuclear accident or an act of terrorism, such as planting a dirty bomb, the nuclear and radioactive materials could be spread over a large area. The environmental radioactivity information has to be obtained rather fast in order to secure the safety of any activity in these potentially high-level radiation contaminated areas. In addition, a lack of both social infrastructure and environmental radiation information may be expected in this area. Therefore, the sampling process must be carried out in a timely and efficient manner. In this study, a probe operated via a distribution method was developed to improve the efficiency of the sampling process by identifying the location and nuclides of leaked materials. For distribution type operation, CsI(Tl) and SiPM were applied to the sensor part of the probe to operate at low voltage operation and for durability. In addition, the power consumption of the data acquisition module and the weight of the probe were optimized for distribution through drones, battery operation, and wireless communication. The energy resolution of this probe was 10.84% at 662 keV and it showed adequate characteristics in the spectrum of nuclear materials. In addition, the practical operability of this probe was verified through determination of the minimum detectable dose rate condition and the difference in the measured values based on the landing orientation.

High-Performance Remote Radioactive Material Identification of Mixtures

Detecting radioactive materials in mixtures is challenging due to low concentration, environmental factors, sensor noise, and others. This paper presents new results on nuclear material identification and mixing ratio estimation for mixtures of materials in which there are multiple isotopes present. Conventional and deep learning-based machine learning algorithms were compared. Both simulated and actual experimental data were used in the comparative studies. It was observed that some newly developed machine learning and deep learning methods have better performance than a conventional method based on region of interest (ROI). Extensive experiments also demonstrated that trained models using low signal-to-background ratio (SBR) data can better generalize to other SBR conditions.

New Results on Radioactive Mixture Identification and Relative Count Contribution Estimation.

Detecting nuclear materials in mixtures is challenging due to low concentration, environmental factors, sensor noise, source-detector distance variations, and others. This paper presents new results on nuclear material identification and relative count contribution (also known as mixing ratio) estimation for mixtures of materials in which there are multiple isotopes present. Conventional and deep-learning-based machine learning algorithms were compared. Realistic simulated data using Gamma Detector Response and Analysis Software (GADRAS) were used in our comparative studies. It was observed that a deep learning approach is highly promising.

Qualification Test System for Radiation Detection Devices QuTeSt

Measurement equipment for the detection and identification of radioactive and nuclear (RN) material has a wide application area. The main application aspects are monitoring, search, and identification. A common goal is to gain reliable measurement results. In the past, the only way to assess the performance of a measuring device was to rely on the data given by the manufacturer of the device itself. Reliable test results from an independent third party are more than welcome. These tests can be performed against consensus standards in order to have reproducible test results, independent of the testing location and the performing laboratory. Fraunhofer INT has conceived and built a test environment to perform dynamic and static test measurements using neutron and gamma sources. Tests can be performed in accordance with the IEC and ANSI standards as well as the ITRAP+10 test procedures. This includes qualification tests of truck portal monitors with the dynamic test system. Generally, the effects of one test parameter on other test parameters are not considered in the test procedures. For example, the accuracy of the dose rate may depend on the energy range of the radioactive source used. Besides the overview of the test systems the paper will address restrictions, problems and limitations of the possible qualification measurements as well as potential limitations arising from the given test procedures themselves.

Data for training and testing radiation detection algorithms in an urban environment

The detection, identification, and localization of illicit nuclear materials in urban environments is of utmost importance for national security. Most often, the process of performing these operations consists of a team of trained individuals equipped with radiation detection devices that have built-in algorithms to alert the user to the presence nuclear material and, if possible, to identify the type of nuclear material present. To encourage the development of new detection, radioisotope identification, and source localization algorithms, a dataset consisting of realistic Monte Carlo–simulated radiation detection data from a 2 in. × 4 in. × 16 in. NaI(Tl) scintillation detector moving through a simulated urban environment based on Knoxville, Tennessee, was developed and made public in the form of a Topcoder competition. The methodology used to create this dataset has been verified using experimental data collected at the Fort Indiantown Gap National Guard facility. Realistic signals from special nuclear material and industrial and medical sources are included in the data for developing and testing algorithms in a dynamic real-world background.

Identification of unknown nuclear material

The identification of spent PWR nuclear fuel in terms of its initial enrichment and final burnup is demonstrated. Spent UO2 fuel from a PWR power station was used as the nuclear material of supposed unknown irradiation history. The identification procedure was based on determining the U and Pu isotopic composition of the fuel by chemical analyses, simulation calculations of fuel evolution and statistical analysis. The procedures followed and associated limitations are discussed.

Concealed nuclear material identification via combined fast-neutron/γ-ray computed tomography (FNGCT): a Monte Carlo study

The potential of a combined and simultaneous fast-neutron/γ-ray computed tomography technique using Monte Carlo simulations is described. This technique is applied on the basis of a hypothetical tomography system comprising an isotopic radiation source (americium-beryllium) and a number (13) of organic scintillation detectors for the production and detection of both fast neutrons and γ rays, respectively. Via a combination of γ-ray and fast neutron tomography the potential is demonstrated to discern nuclear materials, such as compounds comprising plutonium and uranium, from substances that are used widely for neutron moderation and shielding. This discrimination is achieved on the basis of the difference in the attenuation characteristics of these substances. Discrimination of a variety of nuclear material compounds from shielding/moderating substances (the latter comprising lead or polyethylene for example) is shown to be challenging when using either γ-ray or neutron tomography in isolation of one another. Much-improved contrast is obtained for a combination of these tomographic modalities. This method has potential applications for in-situ, non-destructive assessments in nuclear security, safeguards, waste management and related requirements in the nuclear industry.

Pillar-structured neutron detector based multiplicity system

This work demonstrates the potential of silicon pillars filled with boron-10 as a sensor technology for a compact and portable neutron multiplicity system. Solid-state, semiconductor based neutron detectors may enable completely new detector form factors, offer an alternate approach to helium-3 based systems, and reduce detector weight and volume requirements. Thirty-two pillar-structured neutron detectors were assembled into a system with an active area of over 20 cm2 and were used in this work to demonstrate the feasibility of this sensor technology as a potential replacement for helium-3 based gas detectors. Multiplicity measurements were successfully carried out using a californium-252 neutron source, in which the source mass, system efficiency, and die-away time were determined. This demonstration shows that these solid-state detectors could allow for a more compact and portable system that could be used for special nuclear material identification in the field.

Extension of the Dytlewski-style dead time correction formalism for neutron multiplicity counting to any order

Neutron multiplicity counting using shift-register calculus is an established technique in the science of international nuclear safeguards for the identification, verification, and assay of special nuclear materials. Typically passive counting is used for Pu and mixed Pu-U items and active methods are used for U materials. Three measured counting rates, singles, doubles and triples are measured and, in combination with a simple analytical point-model, are used to calculate characteristics of the measurement item in terms of known detector and nuclear parameters. However, the measurement problem usually involves more than three quantities of interest, but even in cases where the next higher order count rate, quads, is statistically viable, it is not quantitatively applied because corrections for dead time losses are currently not available in the predominant analysis paradigm. In this work we overcome this limitation by extending the commonly used dead time correction method, developed by Dytlewski, to quads. We also give results for pents, which may be of interest for certain special investigations. Extension to still higher orders, may be accomplished by inspection based on the sequence presented. We discuss the foundations of the Dytlewski method, give limiting cases, and highlight the opportunities and implications that these new results expose. In particular there exist a number of ways in which the new results may be combined with other approaches to extract the correlated rates, and this leads to various practical implementations.

Identification and Imaging of Special Nuclear Materials and Contraband using Active x-ray Interrogation

A x-ray inspection system utilizing a continuous-wave 9 MeV rhodotron x-ray source for scanning cargo containers is presented. This system scans for contraband, anomalies, stowaway passengers, and nuclear threats for trucks and towed cargo containers. A transmission image is generated concurrently with a 3D image of the cargo, the latter presenting material information in the form of atomic number and density. Neutrons from photofission are also detected during each scan. In addition, nuclear resonance fluorescence detectors are capable of identifying specific isotopes. This system has recently been deployed at the Port of Boston.

Application of Artificial Neural Networks to Reliable Nuclear Data for Nonproliferation Modeling and Simulation

Detection and identification of special nuclear materials (SNMs) are an essential part of the US nonproliferation effort. Modern cutting-edge SNM detection methodologies rely more and more on modeling and simulation techniques. Experiments with radiological samples in realistic configurations, is the ultimate tool that establishes the minimum detection limits of SNMs in a host of different geometries. Modern modeling and simulation approaches have the potential to significantly reduce the number of experiments with radioactive sources needed to determine these detection limits and reduce the financial barrier to SNM detection. Unreliable nuclear data is one of the principal causes of uncertainty in modeling and simulating nuclear systems. In particular, nuclear cross sections introduce a significant uncertainty in the nuclear data. The goal of this research is to develop a methodology that will autonomously extract the correct nuclear resonance characteristics of experimental data in a reliable way, a task previously left to expert judgement. Accurate nuclear data will in turn allow contemporary modeling and simulation to become far more reliable, de-escalating the extent of experimental testing. Consequently, modeling and simulation techniques reduce the use and distribution of radiological sources, while at the same time increase the reliability of the currently used methods for the detection and identification of SNMs.

Simulation study into the identification of nuclear materials in cargo containers using cosmic rays

Muon tomography represents a new type of imaging technique that can be used in detecting high-Z materials. Monte Carlo simulations for muon scattering in different types of target materials are presented. The dependence of the detector capability to identify high-Z targets on spatial resolution has been studied. Muon tracks are reconstructed using a basic point of closest approach (PoCA) algorithm. In this article we report the development of a secondary analysis algorithm that is applied to the reconstructed PoCA points. This algorithm efficiently ascertains clusters of voxels with high average scattering angles to identify `areas of interest' within the inspected volume. Using this approach the effect of other parameters, such as the distance between detectors and the number of detectors per set, on material identification is also presented. Finally, false positive and false negative rates for detecting shielded HEU in realistic scenarios with low-Z clutter are presented.

Recent Fast Neutron Imaging Measurements with the Fieldable Nuclear Materials Identification System

This paper describes recent fast-neutron imaging measurements of the fieldable nuclear materials identification system (FNMIS) under development by Oak Ridge National Laboratory with National Nuclear Security Administration (NNSA-NA-22) support for possible future use in arms control and nonproliferation applications. This paper presents initial imaging measurements performed at Oak Ridge National Laboratory with a Thermo Fisher API 120 DT generator and the fast-neutron imaging module of the FNMIS.1This manuscript has been authored by Oak Ridge National Laboratory, managed by UT-Battelle LLC under contract no. DE-AC05-00OR22725 with the US Department of Energy. The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes.

Multiplicity correlation between neutrons and gamma-rays emitted from SNM and non-SNM sources

The challenge in detection and identification of Special Nuclear Materials (SNM) is to discriminate between the time-correlated neutrons and gamma-rays emitted from SNM and those originating from non-correlated or differently-correlated environmental non-SNM sources. Time-correlated neutron and gamma-ray bursts can be generated by penetrating components of cosmic radiation. The characteristic features or attributes of correlated signatures can be revealed by analyzing the joint probability density functions (JPDFs) of various parameters of neutrons and gamma-rays. Monte Carlo simulations of SNM and cosmic-ray (non-SNM) sources of neutrons and gamma-rays are performed. For both SNM and non-SNM sources, energy-lifetime JPDF of neutrons, energy-lifetime JPDF of gamma-rays, and JPDFs of neutron-gamma-ray multiplicity are evaluated. Mean values, standard deviations, covariance and correlation are estimated. It is found that the number (multiplicity) of neutrons and gamma-rays emitted from an SNM source is moderately correlated (∼0.48). The multiplicity of neutrons and gamma-rays generated by cosmic-ray showers at sea level is only weakly correlated (∼−0.046). The exploitation of neutron-gamma-ray multiplicity correlation in detectors can provide a tool to discriminate non-SNM sources.

Potential of a laser-driven source of characteristic γ-rays for fast detection and identification of nuclear materials

Active interrogation of special nuclear materials (SNM) represents the most promising detection technique in combating their illicit trafficking and diverted use. Among the many interrogation approaches under investigation, the one based on photoinduced fission relies on large prompt and delayed radiation signatures. Pulsed induction (tens of ns) of photofission, in particular, enables to distinguish between prompt and delayed radiation emission, providing additional signatures to identify fissile materials. As the probing radiation, 6- to 7-MeV characteristic γ-rays produced by the 19F(p,αγ)16O reaction are of particular interest; in fact, they have energy sufficient to induce fission in SNM, but insufficient to create significant radioactivity via photonuclear activation, especially in lighter elements. In this study, the feasibility of a pulsed source of characteristic γ-rays based on the interaction of laser-accelerated protons with a fluorine-rich target is assessed. For practical applications, proton energies above 2 MeV, a bunch population of the order of 1014 particles or higher, and a laser repetition rate of the order of 10 Hz are of interest. The development of ultrahigh-power table-top lasers could greatly contribute to the deployment of this kind of interrogation source for a variety of operations.

GEANT4 simulation of a scintillating-fibre tracker for the cosmic-ray muon tomography of legacy nuclear waste containers

Cosmic-ray muons are highly penetrative charged particles that are observed at the sea level with a flux of approximately one per square centimetre per minute. They interact with matter primarily through Coulomb scattering, which is exploited in the field of muon tomography to image shielded objects in a wide range of applications. In this paper, simulation studies are presented that assess the feasibility of a scintillating-fibre tracker system for use in the identification and characterisation of nuclear materials stored within industrial legacy waste containers. A system consisting of a pair of tracking modules above and a pair below the volume to be assayed is simulated within the GEANT4 framework using a range of potential fibre pitches and module separations. Each module comprises two orthogonal planes of fibres that allow the reconstruction of the initial and Coulomb-scattered muon trajectories. A likelihood-based image reconstruction algorithm has been developed that allows the container content to be determined with respect to the scattering density λ, a parameter which is related to the atomic number Z of the scattering material. Images reconstructed from this simulation are presented for a range of anticipated scenarios that highlight the expected image resolution and the potential of this system for the identification of high-Z materials within a shielded, concrete-filled container. First results from a constructed prototype system are presented in comparison with those from a detailed simulation. Excellent agreement between experimental data and simulation is observed showing clear discrimination between the different materials assayed throughout.