Michael Streicher

Special Nuclear Material Characterization Using Digital 3-D Position Sensitive CdZnTe Detectors and High Purity Germanium Spectrometers

Special nuclear material (SNM) monitoring often requires high resolution gamma-ray spectroscopy for material characterization. Portable systems and rapid deployment are also highly valued for some applications. Deployable gamma-ray imaging spectrometers using pixelated CdZnTe semiconductor detectors have become commercially available within the past three years to meet these requirements. CdZnTe systems have demonstrated room-temperature energy resolutions below 0.5% FWHM at 662 keV using single pixel events and below 0.7% FWHM at 662 keV using all events. The systems are able to go from storage to measurement in less than two minutes. Special nuclear materials were measured at the Y-12 National Security Complex and at the Device Assembly Facility at the Nevada National Security Site. The use of a CdZnTe system to measure uranium enrichment was demonstrated and uranium spectral features were compared to a commercially available high purity germanium (HPGe) spectrometer. The use of passive gamma-ray spectroscopy techniques to estimate plutonium grade using CdZnTe detectors was demonstrated for the first time.

A portable 2 × 2 digital 3D CZT imaging spectrometer system

A new portable gamma-ray imaging spectroscopy system using three dimensional position-sensitive CdTeZe has been built at the University of Michigan. This system uses a first generation digital VAD_UM ASIC jointly developed by Integrated Detector Electronics AS (IDEAS) and the University of Michigan and allows four 20 × 20 × 15 mm3 CdZnTe detectors to be simultaneously read out in either a full readout mode (where all channels from the triggered ASIC are read out to a personal computer) or in sparse readout mode (only the triggered channel and its neighbors are read out). The performance of the new system in terms of noise, spectroscopy, and isotope identification will be reported. The portable system is able to use standard AC power and is contained in a pelican case 54 × 30 × 32 cm3.

1-D Fast Neutron Source Localization Using Digital Pixelated 3-D Position-Sensitive CdZnTe Detectors

Recoil of constituent nuclei from neutron elastic scatter in pixelated, 3-D CdZnTe gamma-ray detectors is detectable given current low energy thresholds. Fast neutrons are attenuated by CdZnTe detectors via outscatter and measured gradients in neutron interaction rates across detector pixels that enables 1-D fast neutron source localization through a maximum likelihood estimator. Experimental results using an MP320 deuterium–deuterium neutron generator with the four detector crystal Orion prototype successfully localize four different source locations across a 1-D field of view to within absolute measurement errors between 2.5° and 14.0°.

Fast Neutron Detection Using Pixelated CdZnTe Spectrometers

Fast neutrons are an important signature of special nuclear materials (SNMs). They have a low natural background rate and readily penetrate high atomic number materials that easily shield gamma-ray signatures. Therefore, they provide a complementary signal to gamma rays for detecting shielded SNM. Scattering kinematics dictate that a large nucleus (such as Cd or Te) will recoil with small kinetic energy after an elastic collision with a fast neutron. Charge carrier recombination and quenching further reduce the recorded energy deposited. Thus, the energy threshold of CdZnTe detectors must be very low in order to sense the small signals from these recoils. In this paper, the threshold was reduced to less than 5 keVee to demonstrate that the 5.9-keV X-ray line from 55Fe could be separated from electronic noise. Elastic scattering neutron interactions were observed as small energy depositions (less than 20 keVee) using digitally sampled pulse waveforms from pixelated CdZnTe detectors. Characteristic gamma-ray lines from inelastic neutron scattering were also observed.

A Method to Estimate the Atomic Number and Mass Thickness of Intervening Materials in Uranium and Plutonium Gamma-Ray Spectroscopy Measurements

To accurately characterize shielded special nuclear materials (SNM) using passive gamma-ray spectroscopy measurement techniques, the effective atomic number and the thickness of shielding materials must be measured. Intervening materials between the source and detector may affect the estimated source isotopics (uranium enrichment and plutonium grade) for techniques which rely on raw count rates or photopeak ratios of gamma-ray lines separated in energy. Furthermore, knowledge of the surrounding materials can provide insight regarding the configuration of a device containing SNM. The described method was developed using spectra recorded using high energy resolution CdZnTe detectors, but can be expanded to any gamma-ray spectrometers with energy resolution of better than 1% FWHM at 662 keV. The effective atomic number, Z, and mass thickness of the intervening shielding material are identified by comparing the relative attenuation of different gamma-ray lines and estimating the proportion of Compton scattering interactions to photoelectric absorptions within the shield. While characteristic Kα x-rays can be used to identify shielding materials made of high Z elements, this method can be applied to all shielding materials. This algorithm has adequately estimated the effective atomic number for shields made of iron, aluminum, and polyethylene surrounding uranium samples using experimental data. The mass thicknesses of shielding materials have been estimated with a standard error of less than 1.3 g/cm2 for iron shields up to 2.5 cm thick. The effective atomic number was accurately estimated to 26 ± 5 for all iron thicknesses.

Unbiased Filtered Back-Projection in

In Compton imaging, iterative methods provide good performance but are usually computationally intensive. Thus, direct inverse algorithms such as filtered back-projection (FBP) are preferable when computation time is limited. The conventional FBP method assumes that the point spread function of a back-projection image is isotropic and invariant to incident direction, as required by frequency spectrum analysis. However, most of the time this assumption is not true because the detector geometry is rarely uniform. Therefore the conventional FBP reconstructed image is biased. To solve the geometry non-uniformity problem, this paper proposes an unbiased FBP algorithm that groups Compton events having the same Compton rings using a system matrix decomposition strategy. The proposed method produces more isotropic resolution, and preserves the capability to use frequency spectrum analysis. The algorithm has been applied to data from a 3D position-sensitive detector array with 4 crystals and a digital readout system. The resulting images had more isotropic resolution than standard FBP.

4\pi

The mass thickness and atomic number of materials shielding radioactive sources emitting multiple resolvable gamma-ray energies can be characterized by measuring the attenuation and Compton scatter of emitted gamma rays in recorded spectra against estimated values for a suite of materials and thicknesses. Compton imaging using a maximum-likelihood expectation maximization-based reconstruction can be used to separate angular spectra allowing simultaneous characterization of multiple shielded sources. Using the described algorithm on experimental 133Ba data, we demonstrate estimation of mass thickness and atomic number for iron, tin, and lead shields with another bare source in the field of view with average standard error of 0.6 g/cm2 and 1.5, respectively, while an aluminum shield is reconstructed with ambiguous atomic number but correct thickness.

Compton Imaging With 3D Position Sensitive Detectors

X-ray fluorescence CT (XFCT) imaging is a promising molecular imaging modality with high spatial resolution and non-radioactive high atomic number tracers for medical imaging. While dozens of bench-top experiments proofed the concept of the approach, scalability for clinical systems with high molecular sensitivity at acceptable doses is not yet accomplished. This study compares a spectroscopic CdTe detector which was used for XFCT measurements beforehand to a CZT detector with scalable active area. The results show that the CZT detector is able to provide better image quality due to higher detection efficiency even with a significantly reduced energy resolution and absorbing enclosure.

Identification of Intervening Materials in Gamma-Ray Spectroscopy Measurements Using Angularly Deconvolved Spectra With Multiple Sources in the Field of View

CdZnTe imaging gamma-ray spectrometers using digital pulse processing techniques have shown great promise in recent years. The energy resolution of 15 mm thick CdZnTe spectrometers has improved to around 0.4 % FWHM at 662 keV for single pixel events and near 0.6 % FWHM at 662 keV using all events, regardless of the number of pixels triggered. Additionally, the position resolution through subpixel position sensing has improved to less than 300 μm FWHM for 662 keV gamma-rays. This technical progress has encouraged the use of CdZnTe in challenging environments with high gamma ray fluxes such as nuclear power plants. CdZnTe is also being studied for use in medical imaging applications as well as nuclear safeguards and treaty verification where high count rate environments may be encountered. Signal pulse waveforms generated from charge motion in CdZnTe were studied at high count rates to learn how the material responds in high gamma-ray fluxes. The relationship between average preamplifier decay slope and dose rate is shown to be monotonic. This relationship could be used in a real system to estimate the dose rate on the detector surface. Minimal energy resolution degradation was observed up to 50 mR/hr. Beyond 50 mR/hr, the energy resolution degrades more substantially for multiple pixel events due to chance coincidence Compton scattering events. Changes in the electric field due to positive space charge accumulation were observed which likely contributes to the energy resolution degradation. Preamplifier instability is the other major contributor to energy resolution degradation.

Comparison of a large area CZT detector to a spectroscopic CdTe detector for X-ray fluorescence computed tomography

Nonlinearity in the output signals for the energy deposited by gamma-ray interactions can seriously degrade the energy resolution of a pixellated detector when the signals from more than one pixel are summed to get the total deposited energy. Such nonlinearity can be calibrated on a pixel-by-pixel basis by using multiple gamma-ray check sources of various known energy photopeaks for gamma-ray energy below 1.5 MeV. However such method becomes less practical above 1.5 MeV since only a few commercially available check sources provide a few prominent photopeaks between 1.5 MeV and 3.0 MeV and it takes a long time to get statistically enough counts in these high-energy peaks for each pixel. The nonlinearity in the response of a detector system consists of two parts - one from the nonlinearity of the readout electronics (pre-amp, shaping-amp, peak-hold, etc), the other from the nonlinearity of the detector (charge generation, charge transportation and charge collection). This paper will characterize the nonlinearity of 3-D position-sensitive CdZnTe detectors by using on-chip test pulse injection to separate the nonlinearity of the readout electronics and the detector. The nonlinearity of the detector can be much better modeled after deducting the nonlinearity of the electronics.