- Email: [email protected]
SINRD validation experiments at the time-of-flight facility GELINA
Annals of Nuclear Energy 102 (2017) 368–375
Contents lists available at ScienceDirect
Annals of Nuclear Energy journal homepage: www.elsevier.com/locate/anucene
SINRD validation experiments at the time-of-flight facility GELINA Riccardo Rossa a,b, Gery Alaerts c, Alessandro Borella a,⇑, Jan Heyse c, Stefan Kopecky c, Pierre-Etienne Labeau b, Carlos Paradela c, Nicolas Pauly b, Peter Schillebeeckx c, Klaas van der Meer a, Ruud Wynants c a b c
SCKCEN, Belgian Nuclear Research Centre, Boeretang, 200, B2400 Mol, Belgium Université libre de Bruxelles, Ecole polytechnique de Bruxelles, Service de Métrologie Nucléaire (CP165/84), Avenue F.D. Roosevelt, 50, B1050 Brussels, Belgium European Commission Joint Research Centre, Directorate G, Retieseweg 111, B-2440 Geel, Belgium
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 11 October 2016 Accepted 24 November 2016 Available online 13 January 2017 Keywords: Self-Interrogation Neutron Resonance Densitometry SINRD Spent fuel measurements Safeguards Non-destructive assay 239 Pu Time-of-flight measurements GELINA
a b s t r a c t Self-interrogation neutron resonance densitometry (SINRD) is a non-destructive analysis technique that can be used to quantify the amount of 239Pu in spent nuclear fuel. It is a passive method that relies on the detection of neutrons, which are emitted by the fuel. The amount of 239Pu is estimated from the ratio of the neutron intensity in the fast energy region and in a region close to the 0.296 eV resonance of 239Pu. The neutron intensity in the resonance region is obtained from a detection system with a high sensitivity to 0.296 eV neutrons. This can be realized by using two neutron detectors with 239Pu as convertor material. One of the detectors is covered by a thin Gd foil and the other by a thin Cd foil. The Gd and Cd foils are referred to as SINRD filters. An approach based on the measurement of a fuel assembly in air and surrounded by a slab of polyethylene was developed at SCKCEN. This approach foresees the insertion of small neutron detectors in the guide tubes of the assembly, and optimisation studies of SINRD were based on Monte Carlo simulations. Experiments to support the results of such simulations were carried out at the time-of-flight facility GELINA of the Joint Research Centre (JRC) in Geel (Belgium). Transmission measurements were performed to verify the quality of the nuclear data that are used to define the optimum thickness of the SINRD filters. Results of self-indication measurements were used to confirm the basic principle of SINRD, that is, that the best results are obtained with a detector that has a high sensitivity to neutrons with energy close to the energy of a strong resonance of the material under investigation. The results of these experiments are presented in this work. Ó 2017 Elsevier Ltd. All rights reserved.
1. Basic principles of SINRD Self-Interrogation Neutron Resonance Densitometry (SINRD) (LaFleur, 2011; LaFleur et al., 2012a,b; Hu et al., 2012; Hu et al., 2013; LaFleur et al., 2013; LaFleur et al., 2015) is a passive neutron technique to determine the amount of fissile material in spent fuel. It originates from a technique proposed in 1968 (Menlove et al., 1969) which was based on active interrogation with an external neutron source. This technique, referred to as Self-Indication Neutron Resonance Absorption Densitometry, was inspired by the basic principles of self-indication measurements. The latter is a well-known technique to study cross section data in the resonance region (Fröhner et al., 1966). Applying SINRD to spent fuel, the interrogation relies on prompt fission neutrons from spontaneous ⇑ Corresponding author. E-mail address: [email protected] (A. Borella). http://dx.doi.org/10.1016/j.anucene.2016.11.037 0306-4549/Ó 2017 Elsevier Ltd. All rights reserved.
fission of 244Cm present in the fuel without the use of an external neutron source. Therefore, the term self-interrogation was introduced by LaFleur (2011). The total microscopic cross-section for 239Pu is reported in Fig. 1 for neutron energy close to the resonance at 0.296 eV. In addition, the transmitted neutron fluxes calculated in air through two homogeneous samples containing 238U and different percentages of 239Pu are shown. The samples had a density of 10.4 g/cm3 and thickness equal to the radius of a fuel pin in a PWR 17 17 fuel assembly, i.e. 0.4025 mm. The figure shows the corresponding attenuation of the neutron flux at the energy of the 239Pu resonance, and indicates that the attenuation is more evident for the sample with higher quantity of 239Pu. The attenuation of the neutron flux observed in Fig. 1 can be related to the 239Pu content in the spent fuel. Such attenuation is measured with SINRD by calculating the SINRD signature (RSI) according to Formula (1).
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
material both in the detector and in the sample attenuating the flux.
4
1.0 3
0.6
2
10
0.4 1
10
239
239 0
10
2.1. Setup of the transmission experiments
0.8
1% Pu 239 4% Pu
Pu
Transmission
Total cross-section (barn)
10
10
0.2 0.0
0.1
1
Neutron energy (eV) Fig. 1. Total cross-section of 239Pu (left axis) and transmitted neutron fluxes through 239Pu samples (right axis) as a function of the neutron energy.
RSI ¼
CF C Gd C Cd
369
ð1Þ
The SINRD signature was defined in (LaFleur, 2011) as the ratio between the neutron intensity in the fast energy region and in a region close to the 239Pu resonance. The use of SINRD for spent fuel measurement was first proposed by (LaFleur, 2011) by measuring spent fuel under water, and by placing a set of 235U fission chambers on one side of the fuel assembly. This differs significantly from the SINRD approach proposed by SCKCEN (Rossa et al., 2014; Rossa et al., 2015a,b) that focuses on measurements of PWR spent fuel assemblies in dry conditions. Neutrons emitted from the fuel are measured with small detectors which are inserted in the guide tubes of a nuclear fuel assembly. According to our approach (Rossa et al., 2015a), the neutron intensity in the fast region is derived from the response of a 238U fission chamber (CF), while the neutron intensity in the resonance region is taken as the difference between the neutron counts of two 239 Pu fission chambers covered with Gd and Cd foils, denoted by CGd and CCd, respectively. The choice a 239Pu fission chambers allows the implementation of the self-indication principle (Fröhner et al., 1966), with an increased sensitivity to the neutron flux around the energy region close to the 239Pu resonance. According to the self-indication principle, one utilizes the same sample material, 239Pu in this case, both in the detector and in the sample attenuating the flux being measured. 2. Validation experiments at GELINA Most of the optimization studies of SINRD, e.g. (LaFleur et al., 2012a; Hu et al., 2013; Rossa et al., 2015a,b) are based on results of Monte Carlo simulations. The optimum characteristics of the SINRD filters were defined by Rossa et al. (2015b) based on theoretical calculations. These characteristics differ from those proposed by LaFleur et al. (2012a). To validate the results of Rossa et al. (2015b), transmission experiments were carried out at the time-of-flight (TOF) facility GELINA. A detailed description of this facility, which is operated by the Joint Research Centre, Geel, Belgium, can be found in Mondelaers et al. (2006). Transmission factors through well-characterized Cd and Gd samples of different thickness were obtained and compared with the obtained ones when using the data libraries used for the optimization (Rossa et al., 2015b). In addition, self-indication experiments were carried out to demonstrate the basic principles of SINRD, and an increased sensitivity to the neutron flux around a given energy region can be achieved by utilizing the same sample
Transmission measurements can be used for the determination of total neutron cross-section data (Fröhner et al., 1966; Massimi et al., 2011; Schillebeeckx et al., 2014), and the schematic view of the measurement setup is shown in Fig. 2. The experiments were carried out to measure the transmitted neutron flux (utr) through several samples of Gd and Cd as a function of the time-of-flight. Transmission measurements were carried out at a 10 m transmission station of GELINA (Paradela et al., 2015), with the accelerator operating at 50 Hz. The moderated neutron spectra originated from GELINA can be approximated by a Maxwellian distribution in the thermal region, a 1/E0.85 dependence in the epithermal region, and a Watt function in the fast region (Schillebeeckx et al., 2014). A detailed view of the experimental setup is shown in Fig. 3. An automatic sample changer is positioned at 7.7 m from the neutron producing target, allowing an automated alternation of sample-in and sample-out measurements. A second sample changer, which is placed close to the sample position, is used for anti-overlap and background filters. Neutrons passing through the sample and the filters are detected by a 6.35 mm 76 mm 76 mm Li-glass scintillator, placed at a 11 m distance from the neutron producing target. The detector, which is enriched to 95% in 6Li, is directly viewed by one photomultiplier tube. A good transmission geometry is realized by proper collimation of the neutron beam. A set of Li, B, Cu, Ni and Pb collimators with decreasing diameter are placed between the neutron producing target and the sample; a similar sequence of collimators with increasing diameter are placed between the sample and the detector. Experiments in good transmission geometry are required to assure that all detected neutrons traverse the sample and scattered neutrons do not reach the detector (Schillebeeckx et al., 2012). Transmission measurements using Gd and Cd metal foils or discs of different thicknesses were carried out. The characteristics of the samples are summarized in Table 1. The beam at the sample position had a 10 mm diameter. To reduce the influence of the cray flash in the detector a permanent Pb filter was installed in the beam line. The experimental transmission, Texp, was obtained from the ratio of a sample-in measurement Ctr and a sample-out measurement C0, both corrected for their background contributions Btr and B0, respectively:
T exp ¼
C tr B0 C tr B0
ð2Þ
The TOF spectra (Ctr, Btr, C0, B0) in Eq. (2) were corrected for losses due to the dead time in the detector and electronics chain, and all spectra were normalized to the same neutron intensity and TOF-bin width. The background, which is a sum of TOF
Fig. 2. Schematic view of a transmission experiment.
370
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 3. Experimental setup of the transmission measurements.
Table 1 Characteristics of the samples used for the transmission measurements. All samples were in the form of a metal foil or disc. Sample-ID
Element
Nominal thickness (mm)
Weight (g)
Area (cm2)
Areal density (104 atom/barn)
TP-NP 07-32 SN3S-2015-01-04 Gd-disc NS06001A NS06001C NS06001D
Gd Gd Gd Cd Cd Cd
0.030 0.100 0.200 0.500 1.000 1.000
1.0982 ± 0.0001 2.093 ± 0.001 8.242 ± 0.001 8.6474 ± 0.0001 17.1120 ± 0.0001 17.1446 ± 0.0001
50.3859 ± 0.0007 25.0455 ± 0.0022 50.2084 ± 0.0003 19.6309 ± 0.0005 19.6433 ± 0.0017 19.6397 ± 0.0030
0.8347 ± 0.001 3.2003 ± 0.0016 6.2865 ± 0.0008 23.598 ± 0.001 46.668 ± 0.004 46.765 ± 0.007
independent and dependent components, was determined by applying the black resonance technique (Schillebeeckx et al., 2012). The resonance dips resulting from Co and Na filters placed permanently in the beam were used to account for the impact of the presence of the sample on the background. The average background (Bin, Bout) below 1 eV is about 4% of the sample-in and sample-out measurements (Cin, Cout). The AGL and AGS codes were then used to process the data obtained during the transmission experiments (Becker et al., 2012, 2014). The AGL code was used in the first step of data reduction to sort the list mode data files and to calculate raw TOF spectra. The output files generated with the AGL code were then introduced in the AGS package to calculate the transmission for each SINRD filter. The AGS code carries out the most important spectra manipulations, such as: dead time correction, background fitting and subtraction. The package performs a full propagation of uncertainties, starting from the uncorrelated uncertainties due to counting statistics. The final transmission includes a complete covariance matrix accounting for both uncorrelated and correlated uncertainty components.
Fig. 4. Transmission through different Gd and Cd foils. The experimental transmission is compared with a theoretical transmission based on Eq. (3). The theoretical transmission does not account for the response of the TOF-spectrometer. The total cross-section of 239Pu and 113Cd are also reported.
2.2. Results from the transmission measurements The results of the transmission measurements obtained with the different Gd and Cd foils are shown in Fig. 4. The experimental transmission obtained from Eq. (3) is compared with its theoretical estimate Tan. The latter is derived from the expression:
T an ¼ e
P
nk rtot;k
ð3Þ
where rtot,k is the Doppler broadened total cross section and nk the areal number density of nuclide k present in the sample. This equation is valid for a parallel neutron beam which is perpendicular to a homogeneous sample. For these calculations the Gd cross sections were taken from the ENDF/B-VII.0 nuclear data library (Chadwick et al., 2006), while those for Cd were taken from JEFF-3.2 (Koning et al., 2010). The latter are based on an evaluation that was reported in (Volev et al., 2013). The cross sections were obtained from the JANIS software (Soppera et al., 2014) by using a logarithmic interpolation from 109 to 20 MeV and taking 100 points per decade. In the calculations of the theoretical estimate, the response of the TOFspectrometer was not included. In the low energy region, the influence of this response is small compared to the broadening of the resonance profiles due to the Doppler effect and the total width, i.e. sum of neutron and radiation width, of the resonance. The total cross-section values for 239Pu and 113Cd are also shown in Fig. 4, and the values were taken from the JEFF-3.2 and ENDF/B-VII.0 data library, respectively. The spectra in Fig. 4 illustrate that all Gd foils have a cut-off energy below the Cd resonance at 0.178 eV (Kopecky et al., 2009), whereas the cut-off energy for Cd is slightly above 0.3 eV. Evidently the cut-off energy increases with increasing filter thickness. For a clear absorption of neutrons with energy below about 0.5 eV, a Cd filter with a minimum thickness of 1.0 mm is required. The results in Fig. 4 show a good agreement between the experimental and theoretical transmissions. In addition, the theoretical transmissions were calculated for the Gd and Cd filters with the ENDF/B-VII.0 (Chadwick et al., 2006) and ENDF/B-VII.1 (Chadwick et al., 2011), JEFF-3.2 (Koning et al., 2010) and JENDL-4.0 (Shibata et al., 2011) data libraries to check whether the choice of the data library influences the selection of the optimal combination of SINRD filters. However, similar values of the transmission were obtained for all data libraries. The small differences that are observed might be important for nuclear safety calculations, but
371
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
they do not have an impact on the calculation of the optimum filter thickness. Hence, the total cross sections for Gd and Cd in these libraries can be used for SINRD optimisation studies. To derive the amount of 239Pu from the absorption properties of the 0.296 eV resonance the optimum thickness for a Gd filter is 0.13 mm and 1.0 mm for a Cd filter. This confirms the results reported in (Rossa et al., 2015b). This combination was selected because Gd and Cd foils with these thicknesses have a cut-off energy slightly below and above the resonance of 239Pu at 0.296 eV. 2.3. Setup of the self-indication experiments Self-indication experiments were carried out to demonstrate the basic principles of SINRD. The measurements were performed at a station of GELINA that is mostly used for capture cross section measurements (Lampoudis et al., 2013) (Massimi et al., 2014). This station was already used for self-indication measurements to
Fig. 5. Schematic representation of the self-indication experiments carried out at GELINA.
Fig. 6. Experimental setup for the self-indication experiments. The 0.027 mm Cd capture sample is placed in the neutron beam and is surrounded by 4 C6D6 scintillator detectors.
determine the spin of low energy resonance of 197Au (Massimi et al., 2011). A schematic representation of the set-up is shown in Fig. 5. The incoming neutron flux (u0) interacts with a transmission sample under investigation and with the so-called SINRD filter to obtain the transmitted flux (utr). The sensitivity to the sample is enhanced by combining measurements with different SINRD filters and a detector sensitive to a specific resonance of a material of interest in the sample. Self-indication measurements without the SINRD filter in the neutron beam were also carried out. As indicated in Section 1, 239Pu should be present both in the transmission sample and detector, when applying SINRD to spent fuel. During our measurement campaign, however, due to safety regulations and limitations of available material, measurements on samples with different amounts of 239Pu could not be carried out. Therefore, natural Cd-samples with different thicknesses were used to mimic the presence of 239Pu. The neutron cross section of 113 Cd has a strong resonance at 0.178 eV, which is relatively close in energy to the 0.296 eV resonance of 239Pu. The areal density of the Cd transmission sample that is placed in the beam can be derived from the attenuation of the neutron beam due to the 0.178 eV resonance. The disadvantage of choosing Cd to mimic 239 Pu is that Cd is present both in the transmission sample and the SINRD filter; therefore it is expected that the measurements with the Cd SINRD filter show a response (CCd) that is not sensitive to the Cd transmission sample. Nevertheless, the sensitivity to the Cd transmission sample is present in the measurement with the Gd SINRD filter and therefore, due to the self-indication principle, a correlation with Cd amount will be observed. To construct a detector with a high sensitivity to the 0.178 eV resonance, a thin Cd capture sample (0.027 mm thick) was surrounded by 4 C6D6 liquid scintillators detecting the prompt crays emitted after (n,c) reaction. This detection system was placed at a 13.2 m distance from the neutron producing target, and Fig. 6 shows the experimental setup of the detection system during the self-indication measurements. To enhance the sensitivity to the 0.178 eV resonance, measurements were performed with a Gd and Cd SINRD filter in the beam. The thickness of these filters was optimized for the detection of a 0.178 eV neutron following the criterion used in the previous section for the definition of the SINRD filters thickness for spent fuel measurements. The filter thicknesses were therefore 0.03 mm for the Gd filter and 1.0 mm for the Cd filter for self-indication measurements with Cd as material of interest. The change in thickness of the Gd filters compared to the choice for spent fuel measurement is due to the lower energy of the Cd resonance compared to the one of 239Pu as visible in Fig. 4. The SINRD filters were placed far from the detectors to avoid background contribution of neutron capture reactions in the filters. Measurements were performed using cadmium transmission samples with different thicknesses to mimic spent fuel with different 239Pu contents. The SINRD filters and the transmission samples were placed at about 6.5 m from the neutron producing target. The characteristics of these samples are summarized in Table 2. The results obtained with an ideal self-indication geometry were compared with results derived from measurements with a detector based on the 10B(n,a)7Li and 235U(n,f) conversion reactions. These reactions do not show an enhanced efficiency at the
Table 2 Characteristics of the transmission samples used for the self-indication measurements at GELINA. All samples were in the form of a metal disc with a 80 mm diameter. Sample-ID
Element
Nominal thickness (mm)
Weight (g)
Area (cm2)
Areal density (104 atom/barn)
TP-NP 07-14 TP-NP 07-13 TP-NP 07-12
Cd Cd Cd
0.030 0.050 0.075
1.2814 ± 0.0001 2.0993 ± 0.0001 3.1837 ± 0.0001
50.4334 ± 0.0011 50.4699 ± 0.0018 50.3927 ± 0.0012
1.3611 ± 0.0001 2.2283 ± 0.0001 3.3845 ± 0.0001
372
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
6 RSI,1 RSI,2
5
RSI
4 3 2 1 0 0.0000
0.0001
0.0002
0.0003
0.0004
0.0005
Areal density (at/b) Fig. 7. Spectra of the self-indication experiments with Cd transmission samples in the beam. The detector consists of a 0.027 mm thick Cd capture sample surrounded by 4 C6D6 liquid scintillators. The spectrum obtained with the detector only, i.e. no transmission sample in the beam, is reported for comparison together with the background contribution estimated by Eq. (4). All spectra were normalized to the same beam intensity.
energy of the resonance of interest, i.e. the 0.178 eV resonance. The detectors, which consist of a thin layer of 10B or 235U mounted in a Frisch-gridded ionization chamber, were positioned at a 7.65 m distance from the neutron production target.
Fig. 9. Experimental SINRD signature RSI,1 (Eq. (5)) and RSI,2 (Eq. (6)) as a function of the areal density of the Cd transmission sample placed in the beam. The results are normalized to have RSI = 1 for the measurements without Cd sample in the beam.
Table 3 Neutron counts obtained with the self-indication measurements for different Cd transmission samples. The ideal case refers to the integration of the Time-of-Flight spectra in the region 0.08–0.4 eV whereas the cases with the SINRD filters are the integral value over the full energy range. The values are in 104 counts and are normalized to the same beam intensity. Ideal case
2.4. Results from the self-indication measurements Self-indication experiments using a neutron detector consisting of a 0.027 mm thin Cd capture sample combined with 4 C6D6 detectors were carried out with and without Gd and Cd SINRD filters in the beam. Measurements were performed with 0.03 mm, 0.05 mm, 0.075 mm and 0.105 mm Cd transmission samples placed in the neutron beam. The results with and without SINRD filters are shown in Figs. 7 and 8, respectively. All spectra were normalized to the same neutron intensity using the total counts of a BF3 proportional counter which was installed in the concrete ceiling of the GELINA target hall. The spectrum obtained without a Cd transmission sample is also shown. It is evident that the detection system enhances the detection efficiency in the energy region around 0.178 eV. The figure also reveals the reduction in detector response due to the presence of the Cd transmission samples for the measurements with the Gd SINRD filter. Fig. 8 shows that the
No Cd sample Cd 0.03 mm Cd 0.05 mm Cd 0.075 mm Cd 0.105 mm
18.3 ± < 0.1 9.9 ± 0.1 7.1 ± <0.1 4.7 ± <0.1 3.4 ± <0.1
SINRD filters Gd 0.03 mm
Cd 1.0 mm
22.1 ± <0.1 15.1 ± 0.1 12.0 ± 0.1 10.8 ± <0.1 9.6 ± 0.1
7.0 ± 0.1 7.5 ± 0.1 6.3 ± 0.1 6.8 ± 0.1 6.7 ± 0.1
spectra resulting from the measurements with the 1.0 mm thick Cd SINRD filter are practically insensitive to the presence of the Cd transmission samples. Since almost all neutrons with energy below 1 eV have been absorbed, this part of the spectrum reflects the background contribution. This background is a sum of timeindependent and time-dependent contributions, as discussed in detail in (Schillebeeckx et al., 2012). The total background below 5 eV was approximated by the function:
BðEÞ ¼ a0 þ a1 ðE=1 eVÞa2
ð4Þ
Fig. 8. Spectra of the self-indication experiments with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filter in the beam. The detector consists of a 0.027 mm thick Cd capture sample surrounded by 4 C6D6 liquid scintillators. The spectrum obtained with the detector only, i.e. no transmission sample and SINRD filter in the beam, is reported for comparison together with the background contribution estimated by Eq. (4).
373
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 10. Spectra obtained for the 235U fission chamber with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filter in the beam. Moreover, several Cd transmission samples were used with the Gd SINRD filter to simulate the neutron absorption by fuel pins containing 239Pu.
Fig. 11. Spectra obtained for the 10B ionization chamber with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filters in the beam. Moreover, several Cd transmission samples were used with the Gd SINRD filter to simulate the neutron absorption by fuel pins containing 239Pu.
Table 4 Neutron counts obtained with the 235U fission chamber and 10B ionization chamber for different Cd transmission samples. The values refer to the integral value of the neutron counts with the SINRD filter mentioned in each column. The values are in 104 counts and are normalized to the same beam intensity. 235
10
U fission chamber
No Cd sample Cd 0.03 mm Cd 0.05 mm Cd 0.075 mm Cd 0.105 mm
Cd 1.0 mm
Gd 0.03 mm
Cd 1.0 mm
37.1 ± 0.1 30.5 ± 0.1 27.7 ± 0.1 25.6 ± 0.1 23.9 ± 0.1
19.8 ± <0.1 – – – –
38.4 ± <0.1 30.7 ± <0.1 27.6 ± <0.1 24.8 ± <0.1 22.6 ± <0.1
16.2 ± <0.1 – – – –
The free parameters in Eq. (4) were adjusted by a fit to the data in the energy regions between 0.0022 eV and 0.004 eV, 0.008 eV and 0.009 eV, 1 eV and 1.3 eV, and between 3 eV and 4 eV. The result of such an adjustment is shown in Figs. 7 and 8. From the data in Fig. 7, an ideal observable reflecting a SINRD measurement can be derived by integrating the spectrum C(E) corrected for its background contribution B(E) between 0.08 eV and 0.4 eV. Since all spectra are normalized to the total neutron beam intensity, an additional normalization is not required. Hence, the observable RSI,1 defined by:
Z
B ionization chamber
Gd 0.03 mm
can be used to assess the areal density of the Cd transmission sample that is placed in the beam. This is illustrated in Fig. 9, which plots this observable as a function of the areal density of the Cd transmission sample. A signature that is more related to the quantity derived from a SINRD measurement can be derived from the difference of the total counts resulting from the measurements with either a Gd or Cd SINRD filter in the beam. This signature RSI,2 is defined by:
RSI;2 ¼ 1=ðCGd CCd Þ
ð6Þ
0:4 eV
RSI;1 ¼ 1=
ðCðEÞ BðEÞÞdE 0:08 eV
ð5Þ
where CGd and CCd are the total counts of the spectra taken with the Gd and Cd SINRD filters, respectively.
374
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 12. Experimental signature RSI,2 (Eq. (6)) as a function of the areal density of the Cd transmission sample placed in the beam. The data are results from measurements with different neutron detectors: a self-indication detector consisting of a Cd capture sample combined with C6D6 detectors and ionization chambers loaded with 235U and 10B, respectively. They have been normalized to have RSI, 2 = 1 for the measurements without sample in the beam.
samples were not performed and are not present in the table. As for the self-indication measurements, the total neutron counts obtained with the Gd SINRD filter decrease with the thickness of the Cd transmission sample. The uncertainty of the values due to counting statistics is given and it is below 1% for the 235U fission chamber and below 0.2% for the 10B ionization chamber. From the spectra obtained with the 235U and 10B chambers a signature based on Eq. (6) was derived to estimate the areal density of the Cd transmission sample. The results are compared in Fig. 12 with the signature derived from the measurements with the self-indication detector. This figure clearly demonstrates that a higher sensitivity to the amount of Cd is obtained with the self-indication detector. The signature derived from the measurements with the 235U chamber is more sensitive to the amount of Cd compared to the one derived from the 10B measurements due to the 0.274 eV resonance of 235U. The results in Fig. 12 reveal that the best results are obtained with a neutron detector which has an enhanced efficiency close to a resonance of the material of interest. Hence, to determine the amount of 239Pu in a spent fuel assembly, a neutron detector based on a 239Pu convertor, e.g. 239Pu based fission chamber, is recommended. 3. Summary and conclusions
Table 3 contains the neutron counts obtained in the ideal case by integrating the time-of-flight spectra between 0.08 and 0.4 eV, and by using the SINRD filters. The background was subtracted to obtain the values in Table 3. With both approaches the neutron counts decrease with the thickness of the Cd transmission sample, due to the neutron absorption of this material. The uncertainty of the values due to counting statistics is given and it is always lower than 2%. The SINRD signatures obtained with the two approaches, i.e. RSI,1 and RSI,2, are compared in Fig. 9. This figure reveals that the sensitivity of RSI,2 to the Cd areal density is very close to the one of RSI,1. The latter results from an almost ideal measurement, using an optimised filter and a dedicated procedure to account for the background contribution. This demonstrates the effectiveness of the use of the Gd and Cd SINRD filters to focus on the region close to the 113Cd resonance. The results obtained with a self-indication detector, i.e. using a Cd capture sample combined with the C6D6 detector, can be compared with results of measurements using a 235U fission chamber and 10B ionization chamber as neutron detectors. Figs. 10 and 11 show the spectra normalized to the total neutron intensity taken with the 235U and 10B chambers, respectively. The spectra measured with the Gd and Cd SINRD filters are given separately. The results obtained with the bare 235U fission chamber are also shown. Measurements with a bare 10B ionization chamber were not performed. To optimise the available beam time, the measurements with the Cd SINRD filter were limited to a measurement with only the 1.0 mm thick filter in the beam without any additional Cd transmission sample. However, considering the results in Fig. 8, negligible effects are expected from measurements with additional Cd transmission samples. The spectra in Figs. 10 and 11 illustrate that the 235U and 10B detectors do not show an enhanced efficiency close to the 0.178 eV resonance. The peak below 0.1 eV is due to the shape of the neutron spectrum which has a maximum at about 40 meV. The impact of 235U resonances is clearly visible. The 0.274 eV resonance of 235U slightly enhances the efficiency for neutrons close to the 0.178 eV resonance of 113Cd. The neutron counts obtained with the SINRD filters and differ235 ent Cd transmission samples are reported in Table 4 for the U fis10 sion chamber and for the B ionization chamber. The values with the 1.0 mm-Cd SINRD filter and additional Cd transmission
Experiments were carried out at the time-of-flight facility GELINA to validate the effectiveness of SINRD, a non-destructive technique that can be used to determine the amount of plutonium in a PWR spent fuel assembly. The technique is based on the basic principles of self-indication cross section measurements. It makes use of filters to absorb neutrons, i.e. Gd and Cd filters, and a neutron detector which has an enhanced efficiency in the energy region of a resonance of the material of interest. Transmission measurements were performed to verify the quality of the nuclear data that is needed to optimise the characteristics of the neutron absorption filters. The results show some differences between the experimental and theoretical transmissions. However, the quality of the nuclear data is good enough to define the optimum thickness for both the Gd and Cd SINRD filters. When the amount of Pu is determined from the neutron flux reduction due to the 0.296 eV resonance of 239Pu, the optimum combination is a 0.13 mm thick Gd filter and a 1.0 mm thick Cd filter. In addition, self-indication time-of-flight experiments were carried out. The results of these measurements reveal the effectiveness of the Gd and Cd SINRD filters. It was also confirmed that the highest sensitivity is obtained using a neutron detector with an enhanced efficiency for a resonance of the material of interest. Therefore, a neutron detector based on a 239Pu convertor is recommended for the characterisation of spent fuel by SINRD. Future work will focus on the development and testing of a prototype of the SINRD detector for the measurement of spent fuel assemblies. The thickness of the SINRD filters will be based on the results of the transmission experiments presented in this article. In addition, the availability of 239Pu fission chambers will be further explored since the self-indication detector showed the highest sensitivity to the target material. Acknowledgement The research is sponsored by Engie in the framework of the cooperation agreement CO-90-07-2124 between SCKCEN and Engie. The authors acknowledge the support of the EURATOM Fission 7th Framework Programme’s project GENTLE (Grant No. 323304) for the experiments at the GELINA facility of JRC Geel (Belgium). We are grateful to the GELINA operators for their support during the measurements at the JRC Geel (Belgium).
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
References Becker, B. et al., 2012. Data reduction and uncertainty propagation of time-of-flight spectra with AGS. In: J. Instrum. 7, P11002–P11021. Becker, B., et al., 2014. AGS - Analysis of Geel Spectra Users’ Manual. NEA Data Bank NEA/DB/DOC (2014) 4. Chadwick, M.B. et al., 2006. ENDF/B-VII.0 next generation evaluated nuclear data library for science and technology. Nucl. Data Sheets 107 (12). Chadwick, M.B. et al., 2011. ENDF/B-VII.1 nuclear data for science and technology: Cross sections, covariances, fission product yields and decay data. Nucl. Data Sheets 112 (12). Fröhner, F.H., et al., 1966. Accuracy of neutron resonance parameters from combined area and self-indication ratio measurements. In: P. Hemmig, P. (E. d.), Proceedings of the Conference on Neutron Cross Section Technology, CONF660 303, Washington D.C. (1966) Book 1, pp. 55–66. Hu, J. et al., 2012. The performance of self-interrogation neutron resonance densitometry in measuring spent fuel. J. Nucl. Mater. Manage., Volume XL 3 (2012), 36–44. Hu, J. et al., 2013. The performance of self-interrogation neutron resonance densitometry in measuring spent fuel. Nucl. Instrum. Methods Sect. A 6729 (2013), 247–253. Koning, A.J. et al., 2010. Status of the JEFF nuclear data library. Proc. Int. Conf. Nucl. Data Sci. Technol. 2010, 1057–1062. Kopecky, S. et al., 2009. The total cross section and resonance parameters for the 0.178 eV resonance of 113Cd. Nucl. Instrum. Methods. http://dx.doi.org/10.1016/ j.nimb.2009.04.010. LaFleur, A.M., 2011. Development of Self-Interrogation Neutron Resonance Densitometry (SINRD) to Measure the Fissile Content in Nuclear Fuel. Ph.D. dissertation at Texas A&M University. LaFleur, A.M. et al., 2012a. Development of self-interrogation neutron resonance densitometry to quantify the fissile content in PWR spent LEU and MOX assemblies. Nucl. Sci. Eng. 171 (2012), 175–191. LaFleur, A.M. et al., 2012b. Comparison of fresh fuel experimental measurements to MCNPX calculations using self-interrogation neutron resonance densitometry. Nucl. Instrum. Methods Sect. A 680 (2012), 168–178. LaFleur, A.M. et al., 2013. Development of self-interrogation neutron resonance densitometry to improve detection of partial defects in PWR spent fuel assemblies. Nucl. Technol. 181 (2013), 354–370.
375
LaFleur, A.M. et al., 2015. Analysis of experimental measurements of PWR fresh and spent fuel assemblies using Self-Interrogation Neutron Resonance Densitometry. Nucl. Instrum. Methods Sect. A 781 (2015), 86–95. Lampoudis, C., 2013. Neutron transmission and capture cross section measurements for 241Am at the GELINA facility. Euro. Phys. J. Plus 128 (8), 86. Massimi, C. et al., 2011. Neutron resonance parameters of 197Au from transmission, capture and self-indication measurements. J. Korean Phys. Soc. 59, 1689–1692. Massimi, C. et al., 2014. Neutron capture cross section measurements for 197Au from 3.5 to 84 keV at GELINA. Eur. Phys. J. A 50 (2014), 124. Menlove, H.O. et al., 1969. A resonance self-indication technique for isotopic assay of fissile materials. Nucl. Appl. 6, 401–408. Mondelaers, W. et al., 2006. GELINA, a neutron time-of-flight facility for highresolution neutron data measurements. Notiziario Neutroni e Luce di Sincrotrone I (2), 19–25. Paradela, C., et al., 2015. Characterization of special nuclear material by neutron resonance spectroscopy. In: Proceedings of the 37th ESARDA Symposium, pp. 859–864. Rossa, R., et al., 2014. Optimization of the filters thickness for the SINRD technique applied to spent fuel verification. In: Proceedings of the 55th INMM Annual Meeting. Rossa, R. et al., 2015a. Investigation of the Self-Interrogation Neutron Resonance Densitometry applied to spent fuel using Monte Carlo simulations. Ann. Nucl. Energy 75 (2015), 176–183. Rossa, R. et al., 2015b. Neutron absorbers and detector types for spent fuel verification using the Self-Interrogation Neutron Resonance Densitometry. Nucl. Instrum. Methods Sect. A 791 (2015), 93–100. Schillebeeckx, P. et al., 2012. Determination of resonance parameters and their covariances from neutron induced reaction cross section data. Nucl. Data Sheets 113, 3054–3100. Schillebeeckx, P., et al., 2014. Neutron resonance spectroscopy for the characterization of materials and objects. JRC Science and Policy Reports – Report EUR 26848 EN. Shibata, K. et al., 2011. JENDL-4.0: A New Library for Nuclear Science and Engineering. Journal of Nuclear Science and Technology 48 (1), 1–30. Soppera, N. et al., 2014. JANIS 4: an improved version of the NEA Java-based nuclear data information system. Nucl. Data Sheets 120, 294–296. Volev, K. et al., 2013. Evaluation of resonance parameters for neutron induced reactions in cadmium. Nucl. Instrum. Methods Sect. B 300 (2013), 11–29.
Contents lists available at ScienceDirect
Annals of Nuclear Energy journal homepage: www.elsevier.com/locate/anucene
SINRD validation experiments at the time-of-flight facility GELINA Riccardo Rossa a,b, Gery Alaerts c, Alessandro Borella a,⇑, Jan Heyse c, Stefan Kopecky c, Pierre-Etienne Labeau b, Carlos Paradela c, Nicolas Pauly b, Peter Schillebeeckx c, Klaas van der Meer a, Ruud Wynants c a b c
SCKCEN, Belgian Nuclear Research Centre, Boeretang, 200, B2400 Mol, Belgium Université libre de Bruxelles, Ecole polytechnique de Bruxelles, Service de Métrologie Nucléaire (CP165/84), Avenue F.D. Roosevelt, 50, B1050 Brussels, Belgium European Commission Joint Research Centre, Directorate G, Retieseweg 111, B-2440 Geel, Belgium
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 11 October 2016 Accepted 24 November 2016 Available online 13 January 2017 Keywords: Self-Interrogation Neutron Resonance Densitometry SINRD Spent fuel measurements Safeguards Non-destructive assay 239 Pu Time-of-flight measurements GELINA
a b s t r a c t Self-interrogation neutron resonance densitometry (SINRD) is a non-destructive analysis technique that can be used to quantify the amount of 239Pu in spent nuclear fuel. It is a passive method that relies on the detection of neutrons, which are emitted by the fuel. The amount of 239Pu is estimated from the ratio of the neutron intensity in the fast energy region and in a region close to the 0.296 eV resonance of 239Pu. The neutron intensity in the resonance region is obtained from a detection system with a high sensitivity to 0.296 eV neutrons. This can be realized by using two neutron detectors with 239Pu as convertor material. One of the detectors is covered by a thin Gd foil and the other by a thin Cd foil. The Gd and Cd foils are referred to as SINRD filters. An approach based on the measurement of a fuel assembly in air and surrounded by a slab of polyethylene was developed at SCKCEN. This approach foresees the insertion of small neutron detectors in the guide tubes of the assembly, and optimisation studies of SINRD were based on Monte Carlo simulations. Experiments to support the results of such simulations were carried out at the time-of-flight facility GELINA of the Joint Research Centre (JRC) in Geel (Belgium). Transmission measurements were performed to verify the quality of the nuclear data that are used to define the optimum thickness of the SINRD filters. Results of self-indication measurements were used to confirm the basic principle of SINRD, that is, that the best results are obtained with a detector that has a high sensitivity to neutrons with energy close to the energy of a strong resonance of the material under investigation. The results of these experiments are presented in this work. Ó 2017 Elsevier Ltd. All rights reserved.
1. Basic principles of SINRD Self-Interrogation Neutron Resonance Densitometry (SINRD) (LaFleur, 2011; LaFleur et al., 2012a,b; Hu et al., 2012; Hu et al., 2013; LaFleur et al., 2013; LaFleur et al., 2015) is a passive neutron technique to determine the amount of fissile material in spent fuel. It originates from a technique proposed in 1968 (Menlove et al., 1969) which was based on active interrogation with an external neutron source. This technique, referred to as Self-Indication Neutron Resonance Absorption Densitometry, was inspired by the basic principles of self-indication measurements. The latter is a well-known technique to study cross section data in the resonance region (Fröhner et al., 1966). Applying SINRD to spent fuel, the interrogation relies on prompt fission neutrons from spontaneous ⇑ Corresponding author. E-mail address: [email protected] (A. Borella). http://dx.doi.org/10.1016/j.anucene.2016.11.037 0306-4549/Ó 2017 Elsevier Ltd. All rights reserved.
fission of 244Cm present in the fuel without the use of an external neutron source. Therefore, the term self-interrogation was introduced by LaFleur (2011). The total microscopic cross-section for 239Pu is reported in Fig. 1 for neutron energy close to the resonance at 0.296 eV. In addition, the transmitted neutron fluxes calculated in air through two homogeneous samples containing 238U and different percentages of 239Pu are shown. The samples had a density of 10.4 g/cm3 and thickness equal to the radius of a fuel pin in a PWR 17 17 fuel assembly, i.e. 0.4025 mm. The figure shows the corresponding attenuation of the neutron flux at the energy of the 239Pu resonance, and indicates that the attenuation is more evident for the sample with higher quantity of 239Pu. The attenuation of the neutron flux observed in Fig. 1 can be related to the 239Pu content in the spent fuel. Such attenuation is measured with SINRD by calculating the SINRD signature (RSI) according to Formula (1).
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
material both in the detector and in the sample attenuating the flux.
4
1.0 3
0.6
2
10
0.4 1
10
239
239 0
10
2.1. Setup of the transmission experiments
0.8
1% Pu 239 4% Pu
Pu
Transmission
Total cross-section (barn)
10
10
0.2 0.0
0.1
1
Neutron energy (eV) Fig. 1. Total cross-section of 239Pu (left axis) and transmitted neutron fluxes through 239Pu samples (right axis) as a function of the neutron energy.
RSI ¼
CF C Gd C Cd
369
ð1Þ
The SINRD signature was defined in (LaFleur, 2011) as the ratio between the neutron intensity in the fast energy region and in a region close to the 239Pu resonance. The use of SINRD for spent fuel measurement was first proposed by (LaFleur, 2011) by measuring spent fuel under water, and by placing a set of 235U fission chambers on one side of the fuel assembly. This differs significantly from the SINRD approach proposed by SCKCEN (Rossa et al., 2014; Rossa et al., 2015a,b) that focuses on measurements of PWR spent fuel assemblies in dry conditions. Neutrons emitted from the fuel are measured with small detectors which are inserted in the guide tubes of a nuclear fuel assembly. According to our approach (Rossa et al., 2015a), the neutron intensity in the fast region is derived from the response of a 238U fission chamber (CF), while the neutron intensity in the resonance region is taken as the difference between the neutron counts of two 239 Pu fission chambers covered with Gd and Cd foils, denoted by CGd and CCd, respectively. The choice a 239Pu fission chambers allows the implementation of the self-indication principle (Fröhner et al., 1966), with an increased sensitivity to the neutron flux around the energy region close to the 239Pu resonance. According to the self-indication principle, one utilizes the same sample material, 239Pu in this case, both in the detector and in the sample attenuating the flux being measured. 2. Validation experiments at GELINA Most of the optimization studies of SINRD, e.g. (LaFleur et al., 2012a; Hu et al., 2013; Rossa et al., 2015a,b) are based on results of Monte Carlo simulations. The optimum characteristics of the SINRD filters were defined by Rossa et al. (2015b) based on theoretical calculations. These characteristics differ from those proposed by LaFleur et al. (2012a). To validate the results of Rossa et al. (2015b), transmission experiments were carried out at the time-of-flight (TOF) facility GELINA. A detailed description of this facility, which is operated by the Joint Research Centre, Geel, Belgium, can be found in Mondelaers et al. (2006). Transmission factors through well-characterized Cd and Gd samples of different thickness were obtained and compared with the obtained ones when using the data libraries used for the optimization (Rossa et al., 2015b). In addition, self-indication experiments were carried out to demonstrate the basic principles of SINRD, and an increased sensitivity to the neutron flux around a given energy region can be achieved by utilizing the same sample
Transmission measurements can be used for the determination of total neutron cross-section data (Fröhner et al., 1966; Massimi et al., 2011; Schillebeeckx et al., 2014), and the schematic view of the measurement setup is shown in Fig. 2. The experiments were carried out to measure the transmitted neutron flux (utr) through several samples of Gd and Cd as a function of the time-of-flight. Transmission measurements were carried out at a 10 m transmission station of GELINA (Paradela et al., 2015), with the accelerator operating at 50 Hz. The moderated neutron spectra originated from GELINA can be approximated by a Maxwellian distribution in the thermal region, a 1/E0.85 dependence in the epithermal region, and a Watt function in the fast region (Schillebeeckx et al., 2014). A detailed view of the experimental setup is shown in Fig. 3. An automatic sample changer is positioned at 7.7 m from the neutron producing target, allowing an automated alternation of sample-in and sample-out measurements. A second sample changer, which is placed close to the sample position, is used for anti-overlap and background filters. Neutrons passing through the sample and the filters are detected by a 6.35 mm 76 mm 76 mm Li-glass scintillator, placed at a 11 m distance from the neutron producing target. The detector, which is enriched to 95% in 6Li, is directly viewed by one photomultiplier tube. A good transmission geometry is realized by proper collimation of the neutron beam. A set of Li, B, Cu, Ni and Pb collimators with decreasing diameter are placed between the neutron producing target and the sample; a similar sequence of collimators with increasing diameter are placed between the sample and the detector. Experiments in good transmission geometry are required to assure that all detected neutrons traverse the sample and scattered neutrons do not reach the detector (Schillebeeckx et al., 2012). Transmission measurements using Gd and Cd metal foils or discs of different thicknesses were carried out. The characteristics of the samples are summarized in Table 1. The beam at the sample position had a 10 mm diameter. To reduce the influence of the cray flash in the detector a permanent Pb filter was installed in the beam line. The experimental transmission, Texp, was obtained from the ratio of a sample-in measurement Ctr and a sample-out measurement C0, both corrected for their background contributions Btr and B0, respectively:
T exp ¼
C tr B0 C tr B0
ð2Þ
The TOF spectra (Ctr, Btr, C0, B0) in Eq. (2) were corrected for losses due to the dead time in the detector and electronics chain, and all spectra were normalized to the same neutron intensity and TOF-bin width. The background, which is a sum of TOF
Fig. 2. Schematic view of a transmission experiment.
370
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 3. Experimental setup of the transmission measurements.
Table 1 Characteristics of the samples used for the transmission measurements. All samples were in the form of a metal foil or disc. Sample-ID
Element
Nominal thickness (mm)
Weight (g)
Area (cm2)
Areal density (104 atom/barn)
TP-NP 07-32 SN3S-2015-01-04 Gd-disc NS06001A NS06001C NS06001D
Gd Gd Gd Cd Cd Cd
0.030 0.100 0.200 0.500 1.000 1.000
1.0982 ± 0.0001 2.093 ± 0.001 8.242 ± 0.001 8.6474 ± 0.0001 17.1120 ± 0.0001 17.1446 ± 0.0001
50.3859 ± 0.0007 25.0455 ± 0.0022 50.2084 ± 0.0003 19.6309 ± 0.0005 19.6433 ± 0.0017 19.6397 ± 0.0030
0.8347 ± 0.001 3.2003 ± 0.0016 6.2865 ± 0.0008 23.598 ± 0.001 46.668 ± 0.004 46.765 ± 0.007
independent and dependent components, was determined by applying the black resonance technique (Schillebeeckx et al., 2012). The resonance dips resulting from Co and Na filters placed permanently in the beam were used to account for the impact of the presence of the sample on the background. The average background (Bin, Bout) below 1 eV is about 4% of the sample-in and sample-out measurements (Cin, Cout). The AGL and AGS codes were then used to process the data obtained during the transmission experiments (Becker et al., 2012, 2014). The AGL code was used in the first step of data reduction to sort the list mode data files and to calculate raw TOF spectra. The output files generated with the AGL code were then introduced in the AGS package to calculate the transmission for each SINRD filter. The AGS code carries out the most important spectra manipulations, such as: dead time correction, background fitting and subtraction. The package performs a full propagation of uncertainties, starting from the uncorrelated uncertainties due to counting statistics. The final transmission includes a complete covariance matrix accounting for both uncorrelated and correlated uncertainty components.
Fig. 4. Transmission through different Gd and Cd foils. The experimental transmission is compared with a theoretical transmission based on Eq. (3). The theoretical transmission does not account for the response of the TOF-spectrometer. The total cross-section of 239Pu and 113Cd are also reported.
2.2. Results from the transmission measurements The results of the transmission measurements obtained with the different Gd and Cd foils are shown in Fig. 4. The experimental transmission obtained from Eq. (3) is compared with its theoretical estimate Tan. The latter is derived from the expression:
T an ¼ e
P
nk rtot;k
ð3Þ
where rtot,k is the Doppler broadened total cross section and nk the areal number density of nuclide k present in the sample. This equation is valid for a parallel neutron beam which is perpendicular to a homogeneous sample. For these calculations the Gd cross sections were taken from the ENDF/B-VII.0 nuclear data library (Chadwick et al., 2006), while those for Cd were taken from JEFF-3.2 (Koning et al., 2010). The latter are based on an evaluation that was reported in (Volev et al., 2013). The cross sections were obtained from the JANIS software (Soppera et al., 2014) by using a logarithmic interpolation from 109 to 20 MeV and taking 100 points per decade. In the calculations of the theoretical estimate, the response of the TOFspectrometer was not included. In the low energy region, the influence of this response is small compared to the broadening of the resonance profiles due to the Doppler effect and the total width, i.e. sum of neutron and radiation width, of the resonance. The total cross-section values for 239Pu and 113Cd are also shown in Fig. 4, and the values were taken from the JEFF-3.2 and ENDF/B-VII.0 data library, respectively. The spectra in Fig. 4 illustrate that all Gd foils have a cut-off energy below the Cd resonance at 0.178 eV (Kopecky et al., 2009), whereas the cut-off energy for Cd is slightly above 0.3 eV. Evidently the cut-off energy increases with increasing filter thickness. For a clear absorption of neutrons with energy below about 0.5 eV, a Cd filter with a minimum thickness of 1.0 mm is required. The results in Fig. 4 show a good agreement between the experimental and theoretical transmissions. In addition, the theoretical transmissions were calculated for the Gd and Cd filters with the ENDF/B-VII.0 (Chadwick et al., 2006) and ENDF/B-VII.1 (Chadwick et al., 2011), JEFF-3.2 (Koning et al., 2010) and JENDL-4.0 (Shibata et al., 2011) data libraries to check whether the choice of the data library influences the selection of the optimal combination of SINRD filters. However, similar values of the transmission were obtained for all data libraries. The small differences that are observed might be important for nuclear safety calculations, but
371
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
they do not have an impact on the calculation of the optimum filter thickness. Hence, the total cross sections for Gd and Cd in these libraries can be used for SINRD optimisation studies. To derive the amount of 239Pu from the absorption properties of the 0.296 eV resonance the optimum thickness for a Gd filter is 0.13 mm and 1.0 mm for a Cd filter. This confirms the results reported in (Rossa et al., 2015b). This combination was selected because Gd and Cd foils with these thicknesses have a cut-off energy slightly below and above the resonance of 239Pu at 0.296 eV. 2.3. Setup of the self-indication experiments Self-indication experiments were carried out to demonstrate the basic principles of SINRD. The measurements were performed at a station of GELINA that is mostly used for capture cross section measurements (Lampoudis et al., 2013) (Massimi et al., 2014). This station was already used for self-indication measurements to
Fig. 5. Schematic representation of the self-indication experiments carried out at GELINA.
Fig. 6. Experimental setup for the self-indication experiments. The 0.027 mm Cd capture sample is placed in the neutron beam and is surrounded by 4 C6D6 scintillator detectors.
determine the spin of low energy resonance of 197Au (Massimi et al., 2011). A schematic representation of the set-up is shown in Fig. 5. The incoming neutron flux (u0) interacts with a transmission sample under investigation and with the so-called SINRD filter to obtain the transmitted flux (utr). The sensitivity to the sample is enhanced by combining measurements with different SINRD filters and a detector sensitive to a specific resonance of a material of interest in the sample. Self-indication measurements without the SINRD filter in the neutron beam were also carried out. As indicated in Section 1, 239Pu should be present both in the transmission sample and detector, when applying SINRD to spent fuel. During our measurement campaign, however, due to safety regulations and limitations of available material, measurements on samples with different amounts of 239Pu could not be carried out. Therefore, natural Cd-samples with different thicknesses were used to mimic the presence of 239Pu. The neutron cross section of 113 Cd has a strong resonance at 0.178 eV, which is relatively close in energy to the 0.296 eV resonance of 239Pu. The areal density of the Cd transmission sample that is placed in the beam can be derived from the attenuation of the neutron beam due to the 0.178 eV resonance. The disadvantage of choosing Cd to mimic 239 Pu is that Cd is present both in the transmission sample and the SINRD filter; therefore it is expected that the measurements with the Cd SINRD filter show a response (CCd) that is not sensitive to the Cd transmission sample. Nevertheless, the sensitivity to the Cd transmission sample is present in the measurement with the Gd SINRD filter and therefore, due to the self-indication principle, a correlation with Cd amount will be observed. To construct a detector with a high sensitivity to the 0.178 eV resonance, a thin Cd capture sample (0.027 mm thick) was surrounded by 4 C6D6 liquid scintillators detecting the prompt crays emitted after (n,c) reaction. This detection system was placed at a 13.2 m distance from the neutron producing target, and Fig. 6 shows the experimental setup of the detection system during the self-indication measurements. To enhance the sensitivity to the 0.178 eV resonance, measurements were performed with a Gd and Cd SINRD filter in the beam. The thickness of these filters was optimized for the detection of a 0.178 eV neutron following the criterion used in the previous section for the definition of the SINRD filters thickness for spent fuel measurements. The filter thicknesses were therefore 0.03 mm for the Gd filter and 1.0 mm for the Cd filter for self-indication measurements with Cd as material of interest. The change in thickness of the Gd filters compared to the choice for spent fuel measurement is due to the lower energy of the Cd resonance compared to the one of 239Pu as visible in Fig. 4. The SINRD filters were placed far from the detectors to avoid background contribution of neutron capture reactions in the filters. Measurements were performed using cadmium transmission samples with different thicknesses to mimic spent fuel with different 239Pu contents. The SINRD filters and the transmission samples were placed at about 6.5 m from the neutron producing target. The characteristics of these samples are summarized in Table 2. The results obtained with an ideal self-indication geometry were compared with results derived from measurements with a detector based on the 10B(n,a)7Li and 235U(n,f) conversion reactions. These reactions do not show an enhanced efficiency at the
Table 2 Characteristics of the transmission samples used for the self-indication measurements at GELINA. All samples were in the form of a metal disc with a 80 mm diameter. Sample-ID
Element
Nominal thickness (mm)
Weight (g)
Area (cm2)
Areal density (104 atom/barn)
TP-NP 07-14 TP-NP 07-13 TP-NP 07-12
Cd Cd Cd
0.030 0.050 0.075
1.2814 ± 0.0001 2.0993 ± 0.0001 3.1837 ± 0.0001
50.4334 ± 0.0011 50.4699 ± 0.0018 50.3927 ± 0.0012
1.3611 ± 0.0001 2.2283 ± 0.0001 3.3845 ± 0.0001
372
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
6 RSI,1 RSI,2
5
RSI
4 3 2 1 0 0.0000
0.0001
0.0002
0.0003
0.0004
0.0005
Areal density (at/b) Fig. 7. Spectra of the self-indication experiments with Cd transmission samples in the beam. The detector consists of a 0.027 mm thick Cd capture sample surrounded by 4 C6D6 liquid scintillators. The spectrum obtained with the detector only, i.e. no transmission sample in the beam, is reported for comparison together with the background contribution estimated by Eq. (4). All spectra were normalized to the same beam intensity.
energy of the resonance of interest, i.e. the 0.178 eV resonance. The detectors, which consist of a thin layer of 10B or 235U mounted in a Frisch-gridded ionization chamber, were positioned at a 7.65 m distance from the neutron production target.
Fig. 9. Experimental SINRD signature RSI,1 (Eq. (5)) and RSI,2 (Eq. (6)) as a function of the areal density of the Cd transmission sample placed in the beam. The results are normalized to have RSI = 1 for the measurements without Cd sample in the beam.
Table 3 Neutron counts obtained with the self-indication measurements for different Cd transmission samples. The ideal case refers to the integration of the Time-of-Flight spectra in the region 0.08–0.4 eV whereas the cases with the SINRD filters are the integral value over the full energy range. The values are in 104 counts and are normalized to the same beam intensity. Ideal case
2.4. Results from the self-indication measurements Self-indication experiments using a neutron detector consisting of a 0.027 mm thin Cd capture sample combined with 4 C6D6 detectors were carried out with and without Gd and Cd SINRD filters in the beam. Measurements were performed with 0.03 mm, 0.05 mm, 0.075 mm and 0.105 mm Cd transmission samples placed in the neutron beam. The results with and without SINRD filters are shown in Figs. 7 and 8, respectively. All spectra were normalized to the same neutron intensity using the total counts of a BF3 proportional counter which was installed in the concrete ceiling of the GELINA target hall. The spectrum obtained without a Cd transmission sample is also shown. It is evident that the detection system enhances the detection efficiency in the energy region around 0.178 eV. The figure also reveals the reduction in detector response due to the presence of the Cd transmission samples for the measurements with the Gd SINRD filter. Fig. 8 shows that the
No Cd sample Cd 0.03 mm Cd 0.05 mm Cd 0.075 mm Cd 0.105 mm
18.3 ± < 0.1 9.9 ± 0.1 7.1 ± <0.1 4.7 ± <0.1 3.4 ± <0.1
SINRD filters Gd 0.03 mm
Cd 1.0 mm
22.1 ± <0.1 15.1 ± 0.1 12.0 ± 0.1 10.8 ± <0.1 9.6 ± 0.1
7.0 ± 0.1 7.5 ± 0.1 6.3 ± 0.1 6.8 ± 0.1 6.7 ± 0.1
spectra resulting from the measurements with the 1.0 mm thick Cd SINRD filter are practically insensitive to the presence of the Cd transmission samples. Since almost all neutrons with energy below 1 eV have been absorbed, this part of the spectrum reflects the background contribution. This background is a sum of timeindependent and time-dependent contributions, as discussed in detail in (Schillebeeckx et al., 2012). The total background below 5 eV was approximated by the function:
BðEÞ ¼ a0 þ a1 ðE=1 eVÞa2
ð4Þ
Fig. 8. Spectra of the self-indication experiments with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filter in the beam. The detector consists of a 0.027 mm thick Cd capture sample surrounded by 4 C6D6 liquid scintillators. The spectrum obtained with the detector only, i.e. no transmission sample and SINRD filter in the beam, is reported for comparison together with the background contribution estimated by Eq. (4).
373
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 10. Spectra obtained for the 235U fission chamber with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filter in the beam. Moreover, several Cd transmission samples were used with the Gd SINRD filter to simulate the neutron absorption by fuel pins containing 239Pu.
Fig. 11. Spectra obtained for the 10B ionization chamber with a 0.03 mm Gd (left) and 1.0 mm Cd (right) SINRD filters in the beam. Moreover, several Cd transmission samples were used with the Gd SINRD filter to simulate the neutron absorption by fuel pins containing 239Pu.
Table 4 Neutron counts obtained with the 235U fission chamber and 10B ionization chamber for different Cd transmission samples. The values refer to the integral value of the neutron counts with the SINRD filter mentioned in each column. The values are in 104 counts and are normalized to the same beam intensity. 235
10
U fission chamber
No Cd sample Cd 0.03 mm Cd 0.05 mm Cd 0.075 mm Cd 0.105 mm
Cd 1.0 mm
Gd 0.03 mm
Cd 1.0 mm
37.1 ± 0.1 30.5 ± 0.1 27.7 ± 0.1 25.6 ± 0.1 23.9 ± 0.1
19.8 ± <0.1 – – – –
38.4 ± <0.1 30.7 ± <0.1 27.6 ± <0.1 24.8 ± <0.1 22.6 ± <0.1
16.2 ± <0.1 – – – –
The free parameters in Eq. (4) were adjusted by a fit to the data in the energy regions between 0.0022 eV and 0.004 eV, 0.008 eV and 0.009 eV, 1 eV and 1.3 eV, and between 3 eV and 4 eV. The result of such an adjustment is shown in Figs. 7 and 8. From the data in Fig. 7, an ideal observable reflecting a SINRD measurement can be derived by integrating the spectrum C(E) corrected for its background contribution B(E) between 0.08 eV and 0.4 eV. Since all spectra are normalized to the total neutron beam intensity, an additional normalization is not required. Hence, the observable RSI,1 defined by:
Z
B ionization chamber
Gd 0.03 mm
can be used to assess the areal density of the Cd transmission sample that is placed in the beam. This is illustrated in Fig. 9, which plots this observable as a function of the areal density of the Cd transmission sample. A signature that is more related to the quantity derived from a SINRD measurement can be derived from the difference of the total counts resulting from the measurements with either a Gd or Cd SINRD filter in the beam. This signature RSI,2 is defined by:
RSI;2 ¼ 1=ðCGd CCd Þ
ð6Þ
0:4 eV
RSI;1 ¼ 1=
ðCðEÞ BðEÞÞdE 0:08 eV
ð5Þ
where CGd and CCd are the total counts of the spectra taken with the Gd and Cd SINRD filters, respectively.
374
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
Fig. 12. Experimental signature RSI,2 (Eq. (6)) as a function of the areal density of the Cd transmission sample placed in the beam. The data are results from measurements with different neutron detectors: a self-indication detector consisting of a Cd capture sample combined with C6D6 detectors and ionization chambers loaded with 235U and 10B, respectively. They have been normalized to have RSI, 2 = 1 for the measurements without sample in the beam.
samples were not performed and are not present in the table. As for the self-indication measurements, the total neutron counts obtained with the Gd SINRD filter decrease with the thickness of the Cd transmission sample. The uncertainty of the values due to counting statistics is given and it is below 1% for the 235U fission chamber and below 0.2% for the 10B ionization chamber. From the spectra obtained with the 235U and 10B chambers a signature based on Eq. (6) was derived to estimate the areal density of the Cd transmission sample. The results are compared in Fig. 12 with the signature derived from the measurements with the self-indication detector. This figure clearly demonstrates that a higher sensitivity to the amount of Cd is obtained with the self-indication detector. The signature derived from the measurements with the 235U chamber is more sensitive to the amount of Cd compared to the one derived from the 10B measurements due to the 0.274 eV resonance of 235U. The results in Fig. 12 reveal that the best results are obtained with a neutron detector which has an enhanced efficiency close to a resonance of the material of interest. Hence, to determine the amount of 239Pu in a spent fuel assembly, a neutron detector based on a 239Pu convertor, e.g. 239Pu based fission chamber, is recommended. 3. Summary and conclusions
Table 3 contains the neutron counts obtained in the ideal case by integrating the time-of-flight spectra between 0.08 and 0.4 eV, and by using the SINRD filters. The background was subtracted to obtain the values in Table 3. With both approaches the neutron counts decrease with the thickness of the Cd transmission sample, due to the neutron absorption of this material. The uncertainty of the values due to counting statistics is given and it is always lower than 2%. The SINRD signatures obtained with the two approaches, i.e. RSI,1 and RSI,2, are compared in Fig. 9. This figure reveals that the sensitivity of RSI,2 to the Cd areal density is very close to the one of RSI,1. The latter results from an almost ideal measurement, using an optimised filter and a dedicated procedure to account for the background contribution. This demonstrates the effectiveness of the use of the Gd and Cd SINRD filters to focus on the region close to the 113Cd resonance. The results obtained with a self-indication detector, i.e. using a Cd capture sample combined with the C6D6 detector, can be compared with results of measurements using a 235U fission chamber and 10B ionization chamber as neutron detectors. Figs. 10 and 11 show the spectra normalized to the total neutron intensity taken with the 235U and 10B chambers, respectively. The spectra measured with the Gd and Cd SINRD filters are given separately. The results obtained with the bare 235U fission chamber are also shown. Measurements with a bare 10B ionization chamber were not performed. To optimise the available beam time, the measurements with the Cd SINRD filter were limited to a measurement with only the 1.0 mm thick filter in the beam without any additional Cd transmission sample. However, considering the results in Fig. 8, negligible effects are expected from measurements with additional Cd transmission samples. The spectra in Figs. 10 and 11 illustrate that the 235U and 10B detectors do not show an enhanced efficiency close to the 0.178 eV resonance. The peak below 0.1 eV is due to the shape of the neutron spectrum which has a maximum at about 40 meV. The impact of 235U resonances is clearly visible. The 0.274 eV resonance of 235U slightly enhances the efficiency for neutrons close to the 0.178 eV resonance of 113Cd. The neutron counts obtained with the SINRD filters and differ235 ent Cd transmission samples are reported in Table 4 for the U fis10 sion chamber and for the B ionization chamber. The values with the 1.0 mm-Cd SINRD filter and additional Cd transmission
Experiments were carried out at the time-of-flight facility GELINA to validate the effectiveness of SINRD, a non-destructive technique that can be used to determine the amount of plutonium in a PWR spent fuel assembly. The technique is based on the basic principles of self-indication cross section measurements. It makes use of filters to absorb neutrons, i.e. Gd and Cd filters, and a neutron detector which has an enhanced efficiency in the energy region of a resonance of the material of interest. Transmission measurements were performed to verify the quality of the nuclear data that is needed to optimise the characteristics of the neutron absorption filters. The results show some differences between the experimental and theoretical transmissions. However, the quality of the nuclear data is good enough to define the optimum thickness for both the Gd and Cd SINRD filters. When the amount of Pu is determined from the neutron flux reduction due to the 0.296 eV resonance of 239Pu, the optimum combination is a 0.13 mm thick Gd filter and a 1.0 mm thick Cd filter. In addition, self-indication time-of-flight experiments were carried out. The results of these measurements reveal the effectiveness of the Gd and Cd SINRD filters. It was also confirmed that the highest sensitivity is obtained using a neutron detector with an enhanced efficiency for a resonance of the material of interest. Therefore, a neutron detector based on a 239Pu convertor is recommended for the characterisation of spent fuel by SINRD. Future work will focus on the development and testing of a prototype of the SINRD detector for the measurement of spent fuel assemblies. The thickness of the SINRD filters will be based on the results of the transmission experiments presented in this article. In addition, the availability of 239Pu fission chambers will be further explored since the self-indication detector showed the highest sensitivity to the target material. Acknowledgement The research is sponsored by Engie in the framework of the cooperation agreement CO-90-07-2124 between SCKCEN and Engie. The authors acknowledge the support of the EURATOM Fission 7th Framework Programme’s project GENTLE (Grant No. 323304) for the experiments at the GELINA facility of JRC Geel (Belgium). We are grateful to the GELINA operators for their support during the measurements at the JRC Geel (Belgium).
R. Rossa et al. / Annals of Nuclear Energy 102 (2017) 368–375
References Becker, B. et al., 2012. Data reduction and uncertainty propagation of time-of-flight spectra with AGS. In: J. Instrum. 7, P11002–P11021. Becker, B., et al., 2014. AGS - Analysis of Geel Spectra Users’ Manual. NEA Data Bank NEA/DB/DOC (2014) 4. Chadwick, M.B. et al., 2006. ENDF/B-VII.0 next generation evaluated nuclear data library for science and technology. Nucl. Data Sheets 107 (12). Chadwick, M.B. et al., 2011. ENDF/B-VII.1 nuclear data for science and technology: Cross sections, covariances, fission product yields and decay data. Nucl. Data Sheets 112 (12). Fröhner, F.H., et al., 1966. Accuracy of neutron resonance parameters from combined area and self-indication ratio measurements. In: P. Hemmig, P. (E. d.), Proceedings of the Conference on Neutron Cross Section Technology, CONF660 303, Washington D.C. (1966) Book 1, pp. 55–66. Hu, J. et al., 2012. The performance of self-interrogation neutron resonance densitometry in measuring spent fuel. J. Nucl. Mater. Manage., Volume XL 3 (2012), 36–44. Hu, J. et al., 2013. The performance of self-interrogation neutron resonance densitometry in measuring spent fuel. Nucl. Instrum. Methods Sect. A 6729 (2013), 247–253. Koning, A.J. et al., 2010. Status of the JEFF nuclear data library. Proc. Int. Conf. Nucl. Data Sci. Technol. 2010, 1057–1062. Kopecky, S. et al., 2009. The total cross section and resonance parameters for the 0.178 eV resonance of 113Cd. Nucl. Instrum. Methods. http://dx.doi.org/10.1016/ j.nimb.2009.04.010. LaFleur, A.M., 2011. Development of Self-Interrogation Neutron Resonance Densitometry (SINRD) to Measure the Fissile Content in Nuclear Fuel. Ph.D. dissertation at Texas A&M University. LaFleur, A.M. et al., 2012a. Development of self-interrogation neutron resonance densitometry to quantify the fissile content in PWR spent LEU and MOX assemblies. Nucl. Sci. Eng. 171 (2012), 175–191. LaFleur, A.M. et al., 2012b. Comparison of fresh fuel experimental measurements to MCNPX calculations using self-interrogation neutron resonance densitometry. Nucl. Instrum. Methods Sect. A 680 (2012), 168–178. LaFleur, A.M. et al., 2013. Development of self-interrogation neutron resonance densitometry to improve detection of partial defects in PWR spent fuel assemblies. Nucl. Technol. 181 (2013), 354–370.
375
LaFleur, A.M. et al., 2015. Analysis of experimental measurements of PWR fresh and spent fuel assemblies using Self-Interrogation Neutron Resonance Densitometry. Nucl. Instrum. Methods Sect. A 781 (2015), 86–95. Lampoudis, C., 2013. Neutron transmission and capture cross section measurements for 241Am at the GELINA facility. Euro. Phys. J. Plus 128 (8), 86. Massimi, C. et al., 2011. Neutron resonance parameters of 197Au from transmission, capture and self-indication measurements. J. Korean Phys. Soc. 59, 1689–1692. Massimi, C. et al., 2014. Neutron capture cross section measurements for 197Au from 3.5 to 84 keV at GELINA. Eur. Phys. J. A 50 (2014), 124. Menlove, H.O. et al., 1969. A resonance self-indication technique for isotopic assay of fissile materials. Nucl. Appl. 6, 401–408. Mondelaers, W. et al., 2006. GELINA, a neutron time-of-flight facility for highresolution neutron data measurements. Notiziario Neutroni e Luce di Sincrotrone I (2), 19–25. Paradela, C., et al., 2015. Characterization of special nuclear material by neutron resonance spectroscopy. In: Proceedings of the 37th ESARDA Symposium, pp. 859–864. Rossa, R., et al., 2014. Optimization of the filters thickness for the SINRD technique applied to spent fuel verification. In: Proceedings of the 55th INMM Annual Meeting. Rossa, R. et al., 2015a. Investigation of the Self-Interrogation Neutron Resonance Densitometry applied to spent fuel using Monte Carlo simulations. Ann. Nucl. Energy 75 (2015), 176–183. Rossa, R. et al., 2015b. Neutron absorbers and detector types for spent fuel verification using the Self-Interrogation Neutron Resonance Densitometry. Nucl. Instrum. Methods Sect. A 791 (2015), 93–100. Schillebeeckx, P. et al., 2012. Determination of resonance parameters and their covariances from neutron induced reaction cross section data. Nucl. Data Sheets 113, 3054–3100. Schillebeeckx, P., et al., 2014. Neutron resonance spectroscopy for the characterization of materials and objects. JRC Science and Policy Reports – Report EUR 26848 EN. Shibata, K. et al., 2011. JENDL-4.0: A New Library for Nuclear Science and Engineering. Journal of Nuclear Science and Technology 48 (1), 1–30. Soppera, N. et al., 2014. JANIS 4: an improved version of the NEA Java-based nuclear data information system. Nucl. Data Sheets 120, 294–296. Volev, K. et al., 2013. Evaluation of resonance parameters for neutron induced reactions in cadmium. Nucl. Instrum. Methods Sect. B 300 (2013), 11–29.