Transmission nuclear resonance fluorescence measurements of 238U in thick targets

https://doi.org/10.1016/j.nimb.2011.02.081Get rights and content

Abstract

Transmission nuclear resonance fluorescence measurements were made on targets consisting of Pb and depleted U with total areal densities near 86g/cm2. The 238U content in the targets varied from 0% to 8.5% (atom fraction). The experiment demonstrates the capability of using transmission measurements as a non-destructive technique to identify and quantify the presence of an isotope in samples with thicknesses comparable to the average thickness of a nuclear fuel assembly. The experimental data also appear to demonstrate the process of notch refilling with a predictable intensity. Comparison of measured spectra to previous backscatter 238U measurements indicates general agreement in observed excited states. Evidence of two new 238U excited states and possibly a third state have also been observed.

Introduction

Nuclear resonance fluorescence (NRF) has been a known phenomenon for many years. The original interest was primarily devoted to nuclear structure studies [1], however the process has also been used for non-destructive isotopic measurements of 13C for diamond formation studies [2], and more recently, has been identified as a potential technology for cargo screening [3], [4], [5], [6] and nuclear safeguards [6], [7], [8].

NRF signatures are most commonly measured by either a backscatter or a transmission measurement [9], [10]. In both measurement techniques, a source photon beam is used to induce nuclear excitations in a target. The rate at which NRF occurs in the target is proportional to the amount of the corresponding isotope contained therein. Hence, measurement of the amount of an isotope present in a target is accomplished by determining the rate at which the isotope undergoes NRF. In the backscatter method, this rate is determined by measuring the fluorescence γ rays using radiation detectors positioned at backwards locations, relative to the beam incident upon the target. In a transmission measurement, the rate at which NRF occurs in the target is determined by the attenuation of resonant-energy photons, which causes a reduced NRF rate in a sheet of the same isotope located further along the beam trajectory. This sheet is herein referred to as the transmission detection sheet.

In this paper, we describe transmission NRF measurements of 238U in thick targets using bremsstrahlung. The target dimensions were selected to have an areal density and attenuation properties similar to a nuclear fuel assembly so that the applicability of the transmission method as a non-destructive measurement technique to quantify actinide content in nuclear fuel could be tested. In this experiment, Pb was used as a surrogate for the UO2 matrix in spent fuel and the 238U in depleted uranium (DU) was used as a surrogate for 239Pu or any other actinide that would be measured in spent fuel. The DU sheets were supplied by Manufacturing Sciences Corporation and are 99.8% 238U by mass. The remaining portion of the DU is almost entirely 235U, which has been examined for NRF signatures within the energy range examined here [5], [11]. The amount of 238U used in the experiment represents significantly higher concentrations than those of actinides in spent fuel. These amounts were selected to demonstrate the transmission attenuation effect in a timely manner using readily available radiation detectors and photon sources. A similar measurement has been made using thinner targets and a quasi-monoenergetic photon source [4]. This measurement reported a null result for the observation of the notch refill phenomenon, whereas the data presented here indicate notch refill has occurred. The process of notch refill will be discussed in Section 3.

The transmission measurement also provided information regarding 238U states that undergo NRF. γ rays due to 238U NRF that had previously been reported by Heil et al. [12] were again observed. Six additional γ rays have also been observed. These data are presented in Section 7.

Section snippets

Nuclear resonance fluorescence signatures

States that undergo NRF are described by the total width, Γ, which is the sum of partial widths for different de-excitation modes, Γ=Γi. De-excitation to the ground state by emission of a γ ray is described with a width, Γ0, and likewise, de-excitation to the first-excited state, by Γ1. The cross section for a photon of energy, E, to excite a nucleus to a resonance with centroid energy, EC, is given by the Breit–Wigner distribution.σNRF(E)=πg(c)2E2ΓΓ0(E-Ec)2+(Γ/2)2where g is a statistical

Notch refill

The term notch refill is used to describe the process by which photons incident upon the assay geometry down-scatter to the energy of a resonance and subsequently interact in the transmission detection sheet. The process results in less observed resonant attenuation than would be predicted by consideration of simple exponential attenuation, and therefore neglect of the notch refill phenomenon results in NRF transmission measurements that systematically under-predict the areal density of the

Experimental setup

The experiment was conducted at the High Voltage Research Laboratory at Massachusetts Institute of Technology. Electrons accelerated to 2.60 ± 0.03 MeV by a Van de Graaff accelerator are transported through a beamline, bent 90°, and enter the bottom of the experimental geometry shown in Fig. 1. They then impinge upon a converter target consisting of a 102 μm-thick Au layer on a 1 cm-thick water-cooled Cu backing. The electron current incident upon the converter was approximately 65 μA throughout the

Data analysis

Spectra were collected during four day-long irradiation shifts and background spectra, with no electron beam incident upon the bremsstrahlung converter, were collected overnight. The radioactivity of the DU and ambient 40K provided lines for energy-calibrating the detectors. The γ-ray energies used for energy calibration are shown in Table 3. After calibrating the four detectors, their spectra were re-binned to a common energy grid and summed to provide a single spectrum for each target. An

Areal density measurement

The areal density of an isotope in an irradiated target can be determined if it is related to the measured attenuation of resonant energy photons, as observed by a reduction in the relative rate that radiation detectors measure NRF γ rays emitted from the transmission detection sheet. For each 238U resonance, a function was derived (see Appendix A) that relates the relative rate at which NRF γ rays are emitted from the transmission detection sheet to the areal density of 238U in the assay

Resonance parameters

This experiment not only demonstrates a transmission NRF measurement of 238U and relates this to the 238U areal density in thick targets, but is also an independent measurement of many 238U resonance parameters. Comparing the measured intensities of γ rays due to de-excitation to the ground state and first-excited states, the data provide Γ1W1(θ)/Γ0W0(θ) for each resonance, shown in Table 6. These values were collected for each of the four runs, and remained constant within statistical

Discussion

Performing a transmission measurement to relate the areal density of 238U in the target to the excess attenuation of resonant photons warrants explicit consideration of the sources of errors that enter into the measurement. The foremost source of error is that due to counting statistics. The largest NRF γ-ray lines resulted in approximately 2500 full-energy events measured by the germanium detectors during a measurement, indicating the smallest statistical error allowed by the data is 2%.

Conclusion

The determination of 238U areal densities ranging between 1.7 and 8.5g/cm2 in an approximately 86g/cm2 target by observation of attenuation of resonant-energy photons has been accomplished. While previous transmission measurements using quasi-monoenergetic photon sources have indicated null results for the observation of notch refill [4], the data obtained in this experiment, using thick targets and a bremsstrahlung beam have exhibited a trend indicative of notch refill that could increase the

Acknowledgements

The authors would like to thank Chatham Cooke for his operation of the Van de Graaff accelerator at the High Voltage Research Laboratory (HVRL) at MIT. This work was supported by the MPACT campaign of the FCR&D program of the Office of Nuclear Energy and by the Office of Proliferation Detection, National Nuclear Security Administration, US Department of Energy under Contract No. DE-AC02-05CH11231.

References (21)

There are more references available in the full text version of this article.

Cited by (0)

View full text