Radiation Protection Dosimetry (2014), Vol. 161, No. 1–4, pp. 216 –220 Advance Access publication 6 December 2013
doi:10.1093/rpd/nct324
CORRECTION AND VERIFICATION OF AECL BONNER SPHERE RESPONSE MATRIX BASED ON MONO-ENERGETIC NEUTRON CALIBRATION PERFORMED AT NPL J. Atanackovic1,*, D. J. Thomas2, N. J. Roberts2, S. Witharana3, J. Dubeau3 and A. Yonkeu1 1 Atomic Energy of Canada Limited, Chalk River Laboratories, Chalk River, ONT, Canada K0J 1J0 2 National Physical Laboratory, Hampton Road, Teddington, Middlesex TW11 0LW, UK 3 DETEC, Gatineau, QC, Canada J8 T 4J1 *Corresponding author: [email protected]
INTRODUCTION In this paper a method for correction and verification of a matrix of Bonner Sphere response functions was presented. This correction is based on a correction for the evaluated 3He number density inside the central SP9 counter, precise polyethylene mass density of each sphere and ISO-8529(1) prescribed calibration with mono-energetic neutrons. Using these results and methods outlined by Thomas(2) and Thomas and Soochak(3), the authors were able to generate a matrix of a new AECL BSS response functions. The correction is verified using measurements of several neutron sources in three standard laboratories. Atomic Energy of Canada Limited (AECL) BSS consists of seven polyethylene spheres with diameters of 3, 3.5, 4, 5, 6, 8 and 10 inches(4). The central detector is a spherical 3He proportional counter, SP9, manufactured by Centronic, UK. A set of measurements results in eight count rates, corresponding to seven spheres and a bare 3He proportional counter. The measured neutron spectrum is unfolded using two unfolding codes. Those are STAY’SL, a FORTRAN-based least-squares neutron spectral unfolding code(5) and MAXED, a FORTRAN unfolding code based on maximum entropy(6, 7).
MATERIALS AND METHODS The AECL BSS response matrix consists of 52 energy bins ( per sphere) covering the neutron energy range from thermal to 20 MeV. The bins are structured in such a way that there are five identical logarithmic
energy bins per decade, over 10 energy decades, with logarithmic mid-points located at: 2.512, 3.981, 6.310, 1.000 and 1.585 eV, keV, MeV, etc. The logarithmic mid-point of each bin is given by the expression: pffiffiffiffiffiffiffiffiffiffi ð1Þ Emid ¼ Eu El where Eu and El are the upper and lower boundaries of the particular energy bin, respectively. Before calibrations were performed in the National Physical Laboratory (NPL) mono-energetic fields, the AECL BSS response matrix was based on Monte Carlo (MCNP)(8) calculated responses. The calculated values for 7 spheres and bare SP9 detector for 52 energy binned structure are given in the PhysikalischTechnische Bundesanstalt (PTB) report (9). These MCNP calculations were based on a realistic model of SP9 detector as described in this report. The PTB model assumed the mean polyethylene sphere density to be 0.946 g cm23 and a nominal pressure of 3He gas at room temperature to be 200 kPa. The latter corresponds to the 3He number density of 4.9418` 1019 cm – 3. However, the AECL BSS polyethylene density was evaluated to be 0.951 g cm23. The 3He number density inside the AECL SP9 counter was measured in the NPL thermal field. The absolute response of the counter to NPL thermal fields was measured to be 2.73+0.08 cm2. According to the calculations performed by Thomas and Soochak(3) this SP9 response corresponds to the 3He number density of (4.9+0.2)`1019 cm – 3. Using the methods outlined in refs (2) and (3) the original
# The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
The AECL Bonner Sphere Spectrometer (BSS) was taken to National Physical Laboratory (NPL) for calibration in monoenergetic neutron fields and bare 252Cf neutron fields. The mono-energetic radiations were performed using ISO-8529 prescribed neutron energies: 0.071, 0.144, 0.565, 1.2, 5 and 17 MeV. A central SP9 proportional counter was also evaluated at the NPL thermal neutron calibration facility in order to assess an effective pressure of 3He inside the counter, i.e. number density of 3He atoms. Based on these measurements and methods outlined by Thomas and Soochak, a new BSS response matrix was generated. The response matrix is then verified by unfolding spectra corresponding to various neutron fields. Those are NPL bare 252Cf source, National Institute of Standards and Technology bare and heavy water moderated 252Cf source and 241AmBe calibration source located at National Research Council. A good agreement was observed with expected neutron fluence rates, as well as derived dosimetric quantities, such as International Commission on Radiological Protection-74 ambient dose equivalent.
BONNER SPHERE RESPONSE MATRIX CORRECTION
AECL BSS response matrix was formed by correcting the PTB calculated response matrix for the difference in polyethylene mass density and 3He number density, according to the following equation: nnew nold Rnew ¼ Rold 1 þ S nold r rold 1 þ S 0 new rold
RESULTS AND DISCUSSION
ð2Þ
In the above equation, R, n and r represent the detector response, 3He the number density inside the central counter and the mass density of the Bonner sphere, respectively. S and S 0 are calculated by Thomas(2) using an ANISN code and represent a rate of change of a response with varying 3He number density and polyethylene mass density, respectively. Lastly, the final AECL BSS response matrix contains the corrected values based on mono-energetic neutron measurements performed at NPL.
Table 1. Ratio of AECL calculated and NPL measured response function values at ISO-8529 energies for 3-, 3.5- and 4-inch spheres. 3 inch
71 keV 144 keV 565 keV 1.2 MeV 5 MeV 17 MeV Correction factor
3.5 inch
4 inch
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
0.99 0.98 1.01 0.94 0.96 n/a 1.02
1.01 1.00 1.03 0.96 0.98 n/a 1.00
0.97 0.97 1.03 0.95 0.99 n/a 1.02
0.99 0.99 1.05 0.97 1.01 n/a 1.00
0.94 0.93 0.98 0.94 0.97 n/a 1.05
0.99 0.97 1.03 0.99 1.02 n/a 1.00
Table 2. Ratio of AECL calculated and NPL measured response function values at ISO-8529 energies for 5-, 6-, 8- and 10-inch spheres. 5 inch
71 keV 144 keV 565 keV 1.2 MeV 5 MeV 17 MeV Correction factor
6 inch
8 inch
10 inch
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
0.95 0.95 0.98 0.93 0.96 n/a 1.05
1.00 1.00 1.03 0.97 1.00 n/a 1.00
n/a 0.95 0.96 0.93 0.92 n/a 1.06
n/a 1.01 1.02 0.99 0.98 n/a 1.00
n/a 0.98 0.97 0.89 0.91 n/a 1.07
n/a 1.05 1.04 0.96 0.98 0.97 1.00
n/a 1.00 0.93 0.88 0.87 n/a 1.08
n/a 1.08 1.01 0.95 0.94 0.87 1.03
217
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
Tables 1 and 2 contain correction progress of the BSS response matrix for all spheres. Columns designated as Corr. 1 correspond to the ratio of the BSS responses before mono-energetic neutron corrections were applied. The row designated as correction factor corresponds to the reciprocal of the average Corr. 1 values. It is of interest that correction factors for all spheres were .1, suggesting the overall over-response of the pre-calibrated BSS. This was in accordance with the results obtained before and reported elsewhere(4). Finally, each sphere’s response function values were multiplied by the corresponding correction factor and a new column (designated as Fin. Ratio) was formed. The Fin. (Final) Ratio represents the average deviation of the final response function values from corresponding measured values. Evidently, the Fin. Ratio for each sphere is 1, except for the 10-inch sphere, where this value is 1.03. However, it improved significantly, since it dropped from 1.08 to 1.03. Once the final BSS response matrix was constructed, four neutron spectra at three institutions
J. ATANACKOVIC ET AL.
were measured and unfolded. The 252Cf measurement at NPL was performed using a shadow cone subtraction method. On the other hand, the measurements at National Research Council (NRC) and National Institute of Standards and Technology (NIST) include the scattering component, i.e. the shadow cone subtraction method was not used in these institutions. The unfolded neutron spectra are shown in Figures 1 and 2. The FORTRAN-based codes STAY’SL and MAXED were used for unfolding. All four spectra are presented using BSS binning based on the 52 energy bin structure, while the 241AmBe spectrum is also presented using ISO-8529 fine binning in order to emphasise the fine structures in
the spectrum. A priori (guess) spectra were taken from ISO-8529, with the addition of a prominent thermal peak for the NIST 252Cf spectra. The 252Cf heavy water-moderated spectrum, presented in Figure 1b was unfolded using STAY’SL by two operators (authors) designated as ‘J’ and ‘S’ and using MAXED only by operator ‘J’. Table 3 represents the quantitative summary of all results. Where applicable, the integral neutron fluence and derived dosimetric quantities, obtained by two authors (designated as ‘J’ and ‘S’) are averaged and reported in the table. The ratio of unfolded/expected is also given in the table. In the case of NPL, the direct (non-scattered) neutron fluence was calculated
Figure 2. NRC 241AmBe unfolded neutron spectrum using MAXED and STAY’SL. (a) STAY’SL unfolding and (b) MAXED unfolding including fine binning and ISO-8529 default spectrum. Note that fluences ,10 – 4 MeV are negligible.
218
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
Figure 1. Bare and D2O moderated 252Cf neutron spectra measured at NPL and NIST. (a) STAY’SL unfolded bare 252Cf spectra acquired at NPL and NIST. (b) NIST 252Cf D2O moderated unfolded neutron spectra. MAXED and STAY’SL independent unfolding by two authors designated as J and S.
BONNER SPHERE RESPONSE MATRIX CORRECTION Table 3. Tabular representation of the results. Expected values
252
Cf (NPL) Cf (NIST) AmBe (NRC) 252 Cf mod. (NIST) 252 241
Calculated STAY’SL
Calculated MAXED
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
162+2 n/a 13 n/a
22.4+0.4 5.4 1.77 15.9
165.4 46.7 13 498
22.8 5.7 1.8 17.6
165.1 n/a 13.3 504
22.8 n/a 1.9 17.4
Ratio
f(E) H*(10)
1.02 n/a 1.01 n/a
1.02 1.06 1.04 1.1
Expected and unfolded neutron fluence rates, together with the corresponding expected and derived values for International Commission on Radiological Protection-74 ambient dose equivalent. Ratio values (unfolded/expected) are also included.
CONCLUSION The AECL BSS response matrix was created based on methods proposed by Wiegel et al.(9), Thomas(2) and Thomas and Soochak(3). The response matrix was further corrected for the mono-energetic neutron
measurements taken and NPL. In order to experimentally verify the response matrix, four neutron measurements were taken at three laboratories: NPL, NIST and NRC. Good agreement with expected values both for integrated neutron fluence and derived dosimetric quantities was observed in all four cases. ACKNOWLEDGEMENTS The authors would like to thank Atomic Energy of Canada Nuclear Platform program for funding this work. Also the authors would like to thank Dr Burkhard Wiegel of PTB for kindly providing Report PTB-N-21. Finally, the authors would like to thank Dr Alan Keith Thompson for helping to organise and perform measurements at NIST. FUNDING This work is funded from AECL Nuclear Platform Research Program on neutron metrology, dosimetry and neutron activation analysis for in vivo applications for fiscal years 12/13 and 13/14. REFERENCES 1. Reference neutron radiations. Tech. Rep. 8529. ISO (2001). 2. Thomas, D. Use of the program ANISN to calculate response functions for a Bonner Sphere Set with a 3He detector. Tech. Rep. NPL Report RSA(EXT) 31. NPL (1992). 3. Thomas, D. and Soochak, N. Determination of the 3He number density for the proportional counter used in the NPL Bonner Sphere System. Tech. Rep. NPL Report RS(EXT) 104. NPL (1988). 4. Atanackovic, J. et al. Neutron spectrometry and dosimetry study at two research nuclear reactors using Bonner Sphere Spectrometer (BSS), ROtational SPECtrometer (ROSPEC) and Nested Neutron Spectrometer (NNS). Radiat. Prot. Dosim. 150(3), 364–374 (2013).
219
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
from the emission rate, which was previously measured by means of the manganese sulphate bath. Furthermore, the anisotropy of the neutron emission was taken into account. On the other hand, in the case of NRC, the total neutron fluence that includes the scattering component was calculated using inhouse correction methods according to ISO-8529(1). The same is applicable to the NRC ambient dose rate. However, NIST does not report expected (official) value for the neutron fluence rate; however, the ambient dose rate is reported. This value was also derived using in-house methods based on ISO-8529 document. Where applicable, the final unfolded value was calculated as yet another average of MAXED and STAY’SL results. With the final corrected response matrix, the authors were able to achieve a very good agreement with the NPL and NRC sources. Note that the overall conservative uncertainties of the measurements are 5 and 15 % for the integrated fluence and derived dosimetric quantities, respectively(4). Also, a reasonable agreement, within the above-stated uncertainties is observed with the NIST sources. The measurement at NPL was performed using a shadow cone set, so that the scattering component of the spectrum was subtracted before unfolding. However, shadow cones were not used for NIST and NRC measurements. It would then be of interest to correct the BSS measurements for the bare 252Cf at NIST and the AmBe at NRC using shadow cones. This would allow the BSS results to be compared directly with the fluence rate and ambient dose equivalent rates due to uncollided neutrons at those facilities.
J. ATANACKOVIC ET AL. 5. Perey, F. Least-squares dosimetry unfolding: the program STAY’SL. Tech. Rep. ORNL/TM-6062. Oak Ridge National Laboratory (1977). 6. Reginatto, M. and Goldhagen, P. MAXED, a computer code for maximum entropy deconvolution of multisphere neutron spectrometer data. Health Phys. J. 77(5), 579–583 (1999). 7. Reginatto, M. et al. Spectrum unfolding, sensitivity analysis and propagation of uncertainties with the maximum
entropy deconvolution code MAXED. Nucl. Instrum. Methods A 476, 242 –246 (2002). 8. MCNP—a general Monte Carlo N-Particle transport code. Tech. Rep. LA-UR-03-1987. Los Alamos National Laboratory (2008). 9. Wiegel, B. et al. Calculations of the response funtions of Bonner Spheres with a spherical 3He proportional counter using realistic detector model. Tech. Rep. Report PTB-N21. PTB (1988).
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
220
doi:10.1093/rpd/nct324
CORRECTION AND VERIFICATION OF AECL BONNER SPHERE RESPONSE MATRIX BASED ON MONO-ENERGETIC NEUTRON CALIBRATION PERFORMED AT NPL J. Atanackovic1,*, D. J. Thomas2, N. J. Roberts2, S. Witharana3, J. Dubeau3 and A. Yonkeu1 1 Atomic Energy of Canada Limited, Chalk River Laboratories, Chalk River, ONT, Canada K0J 1J0 2 National Physical Laboratory, Hampton Road, Teddington, Middlesex TW11 0LW, UK 3 DETEC, Gatineau, QC, Canada J8 T 4J1 *Corresponding author: [email protected]
INTRODUCTION In this paper a method for correction and verification of a matrix of Bonner Sphere response functions was presented. This correction is based on a correction for the evaluated 3He number density inside the central SP9 counter, precise polyethylene mass density of each sphere and ISO-8529(1) prescribed calibration with mono-energetic neutrons. Using these results and methods outlined by Thomas(2) and Thomas and Soochak(3), the authors were able to generate a matrix of a new AECL BSS response functions. The correction is verified using measurements of several neutron sources in three standard laboratories. Atomic Energy of Canada Limited (AECL) BSS consists of seven polyethylene spheres with diameters of 3, 3.5, 4, 5, 6, 8 and 10 inches(4). The central detector is a spherical 3He proportional counter, SP9, manufactured by Centronic, UK. A set of measurements results in eight count rates, corresponding to seven spheres and a bare 3He proportional counter. The measured neutron spectrum is unfolded using two unfolding codes. Those are STAY’SL, a FORTRAN-based least-squares neutron spectral unfolding code(5) and MAXED, a FORTRAN unfolding code based on maximum entropy(6, 7).
MATERIALS AND METHODS The AECL BSS response matrix consists of 52 energy bins ( per sphere) covering the neutron energy range from thermal to 20 MeV. The bins are structured in such a way that there are five identical logarithmic
energy bins per decade, over 10 energy decades, with logarithmic mid-points located at: 2.512, 3.981, 6.310, 1.000 and 1.585 eV, keV, MeV, etc. The logarithmic mid-point of each bin is given by the expression: pffiffiffiffiffiffiffiffiffiffi ð1Þ Emid ¼ Eu El where Eu and El are the upper and lower boundaries of the particular energy bin, respectively. Before calibrations were performed in the National Physical Laboratory (NPL) mono-energetic fields, the AECL BSS response matrix was based on Monte Carlo (MCNP)(8) calculated responses. The calculated values for 7 spheres and bare SP9 detector for 52 energy binned structure are given in the PhysikalischTechnische Bundesanstalt (PTB) report (9). These MCNP calculations were based on a realistic model of SP9 detector as described in this report. The PTB model assumed the mean polyethylene sphere density to be 0.946 g cm23 and a nominal pressure of 3He gas at room temperature to be 200 kPa. The latter corresponds to the 3He number density of 4.9418` 1019 cm – 3. However, the AECL BSS polyethylene density was evaluated to be 0.951 g cm23. The 3He number density inside the AECL SP9 counter was measured in the NPL thermal field. The absolute response of the counter to NPL thermal fields was measured to be 2.73+0.08 cm2. According to the calculations performed by Thomas and Soochak(3) this SP9 response corresponds to the 3He number density of (4.9+0.2)`1019 cm – 3. Using the methods outlined in refs (2) and (3) the original
# The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
The AECL Bonner Sphere Spectrometer (BSS) was taken to National Physical Laboratory (NPL) for calibration in monoenergetic neutron fields and bare 252Cf neutron fields. The mono-energetic radiations were performed using ISO-8529 prescribed neutron energies: 0.071, 0.144, 0.565, 1.2, 5 and 17 MeV. A central SP9 proportional counter was also evaluated at the NPL thermal neutron calibration facility in order to assess an effective pressure of 3He inside the counter, i.e. number density of 3He atoms. Based on these measurements and methods outlined by Thomas and Soochak, a new BSS response matrix was generated. The response matrix is then verified by unfolding spectra corresponding to various neutron fields. Those are NPL bare 252Cf source, National Institute of Standards and Technology bare and heavy water moderated 252Cf source and 241AmBe calibration source located at National Research Council. A good agreement was observed with expected neutron fluence rates, as well as derived dosimetric quantities, such as International Commission on Radiological Protection-74 ambient dose equivalent.
BONNER SPHERE RESPONSE MATRIX CORRECTION
AECL BSS response matrix was formed by correcting the PTB calculated response matrix for the difference in polyethylene mass density and 3He number density, according to the following equation: nnew nold Rnew ¼ Rold 1 þ S nold r rold 1 þ S 0 new rold
RESULTS AND DISCUSSION
ð2Þ
In the above equation, R, n and r represent the detector response, 3He the number density inside the central counter and the mass density of the Bonner sphere, respectively. S and S 0 are calculated by Thomas(2) using an ANISN code and represent a rate of change of a response with varying 3He number density and polyethylene mass density, respectively. Lastly, the final AECL BSS response matrix contains the corrected values based on mono-energetic neutron measurements performed at NPL.
Table 1. Ratio of AECL calculated and NPL measured response function values at ISO-8529 energies for 3-, 3.5- and 4-inch spheres. 3 inch
71 keV 144 keV 565 keV 1.2 MeV 5 MeV 17 MeV Correction factor
3.5 inch
4 inch
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
Corr. 1, AECL/NPL
Fin. Ratio, AECL/NPL
0.99 0.98 1.01 0.94 0.96 n/a 1.02
1.01 1.00 1.03 0.96 0.98 n/a 1.00
0.97 0.97 1.03 0.95 0.99 n/a 1.02
0.99 0.99 1.05 0.97 1.01 n/a 1.00
0.94 0.93 0.98 0.94 0.97 n/a 1.05
0.99 0.97 1.03 0.99 1.02 n/a 1.00
Table 2. Ratio of AECL calculated and NPL measured response function values at ISO-8529 energies for 5-, 6-, 8- and 10-inch spheres. 5 inch
71 keV 144 keV 565 keV 1.2 MeV 5 MeV 17 MeV Correction factor
6 inch
8 inch
10 inch
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
Corr. 1, AECL/ NPL
Fin. Ratio, AECL/ NPL
0.95 0.95 0.98 0.93 0.96 n/a 1.05
1.00 1.00 1.03 0.97 1.00 n/a 1.00
n/a 0.95 0.96 0.93 0.92 n/a 1.06
n/a 1.01 1.02 0.99 0.98 n/a 1.00
n/a 0.98 0.97 0.89 0.91 n/a 1.07
n/a 1.05 1.04 0.96 0.98 0.97 1.00
n/a 1.00 0.93 0.88 0.87 n/a 1.08
n/a 1.08 1.01 0.95 0.94 0.87 1.03
217
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
Tables 1 and 2 contain correction progress of the BSS response matrix for all spheres. Columns designated as Corr. 1 correspond to the ratio of the BSS responses before mono-energetic neutron corrections were applied. The row designated as correction factor corresponds to the reciprocal of the average Corr. 1 values. It is of interest that correction factors for all spheres were .1, suggesting the overall over-response of the pre-calibrated BSS. This was in accordance with the results obtained before and reported elsewhere(4). Finally, each sphere’s response function values were multiplied by the corresponding correction factor and a new column (designated as Fin. Ratio) was formed. The Fin. (Final) Ratio represents the average deviation of the final response function values from corresponding measured values. Evidently, the Fin. Ratio for each sphere is 1, except for the 10-inch sphere, where this value is 1.03. However, it improved significantly, since it dropped from 1.08 to 1.03. Once the final BSS response matrix was constructed, four neutron spectra at three institutions
J. ATANACKOVIC ET AL.
were measured and unfolded. The 252Cf measurement at NPL was performed using a shadow cone subtraction method. On the other hand, the measurements at National Research Council (NRC) and National Institute of Standards and Technology (NIST) include the scattering component, i.e. the shadow cone subtraction method was not used in these institutions. The unfolded neutron spectra are shown in Figures 1 and 2. The FORTRAN-based codes STAY’SL and MAXED were used for unfolding. All four spectra are presented using BSS binning based on the 52 energy bin structure, while the 241AmBe spectrum is also presented using ISO-8529 fine binning in order to emphasise the fine structures in
the spectrum. A priori (guess) spectra were taken from ISO-8529, with the addition of a prominent thermal peak for the NIST 252Cf spectra. The 252Cf heavy water-moderated spectrum, presented in Figure 1b was unfolded using STAY’SL by two operators (authors) designated as ‘J’ and ‘S’ and using MAXED only by operator ‘J’. Table 3 represents the quantitative summary of all results. Where applicable, the integral neutron fluence and derived dosimetric quantities, obtained by two authors (designated as ‘J’ and ‘S’) are averaged and reported in the table. The ratio of unfolded/expected is also given in the table. In the case of NPL, the direct (non-scattered) neutron fluence was calculated
Figure 2. NRC 241AmBe unfolded neutron spectrum using MAXED and STAY’SL. (a) STAY’SL unfolding and (b) MAXED unfolding including fine binning and ISO-8529 default spectrum. Note that fluences ,10 – 4 MeV are negligible.
218
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
Figure 1. Bare and D2O moderated 252Cf neutron spectra measured at NPL and NIST. (a) STAY’SL unfolded bare 252Cf spectra acquired at NPL and NIST. (b) NIST 252Cf D2O moderated unfolded neutron spectra. MAXED and STAY’SL independent unfolding by two authors designated as J and S.
BONNER SPHERE RESPONSE MATRIX CORRECTION Table 3. Tabular representation of the results. Expected values
252
Cf (NPL) Cf (NIST) AmBe (NRC) 252 Cf mod. (NIST) 252 241
Calculated STAY’SL
Calculated MAXED
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
f(E) (cm22 s21)
H*(10)`1022 (mSv h21)
162+2 n/a 13 n/a
22.4+0.4 5.4 1.77 15.9
165.4 46.7 13 498
22.8 5.7 1.8 17.6
165.1 n/a 13.3 504
22.8 n/a 1.9 17.4
Ratio
f(E) H*(10)
1.02 n/a 1.01 n/a
1.02 1.06 1.04 1.1
Expected and unfolded neutron fluence rates, together with the corresponding expected and derived values for International Commission on Radiological Protection-74 ambient dose equivalent. Ratio values (unfolded/expected) are also included.
CONCLUSION The AECL BSS response matrix was created based on methods proposed by Wiegel et al.(9), Thomas(2) and Thomas and Soochak(3). The response matrix was further corrected for the mono-energetic neutron
measurements taken and NPL. In order to experimentally verify the response matrix, four neutron measurements were taken at three laboratories: NPL, NIST and NRC. Good agreement with expected values both for integrated neutron fluence and derived dosimetric quantities was observed in all four cases. ACKNOWLEDGEMENTS The authors would like to thank Atomic Energy of Canada Nuclear Platform program for funding this work. Also the authors would like to thank Dr Burkhard Wiegel of PTB for kindly providing Report PTB-N-21. Finally, the authors would like to thank Dr Alan Keith Thompson for helping to organise and perform measurements at NIST. FUNDING This work is funded from AECL Nuclear Platform Research Program on neutron metrology, dosimetry and neutron activation analysis for in vivo applications for fiscal years 12/13 and 13/14. REFERENCES 1. Reference neutron radiations. Tech. Rep. 8529. ISO (2001). 2. Thomas, D. Use of the program ANISN to calculate response functions for a Bonner Sphere Set with a 3He detector. Tech. Rep. NPL Report RSA(EXT) 31. NPL (1992). 3. Thomas, D. and Soochak, N. Determination of the 3He number density for the proportional counter used in the NPL Bonner Sphere System. Tech. Rep. NPL Report RS(EXT) 104. NPL (1988). 4. Atanackovic, J. et al. Neutron spectrometry and dosimetry study at two research nuclear reactors using Bonner Sphere Spectrometer (BSS), ROtational SPECtrometer (ROSPEC) and Nested Neutron Spectrometer (NNS). Radiat. Prot. Dosim. 150(3), 364–374 (2013).
219
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
from the emission rate, which was previously measured by means of the manganese sulphate bath. Furthermore, the anisotropy of the neutron emission was taken into account. On the other hand, in the case of NRC, the total neutron fluence that includes the scattering component was calculated using inhouse correction methods according to ISO-8529(1). The same is applicable to the NRC ambient dose rate. However, NIST does not report expected (official) value for the neutron fluence rate; however, the ambient dose rate is reported. This value was also derived using in-house methods based on ISO-8529 document. Where applicable, the final unfolded value was calculated as yet another average of MAXED and STAY’SL results. With the final corrected response matrix, the authors were able to achieve a very good agreement with the NPL and NRC sources. Note that the overall conservative uncertainties of the measurements are 5 and 15 % for the integrated fluence and derived dosimetric quantities, respectively(4). Also, a reasonable agreement, within the above-stated uncertainties is observed with the NIST sources. The measurement at NPL was performed using a shadow cone set, so that the scattering component of the spectrum was subtracted before unfolding. However, shadow cones were not used for NIST and NRC measurements. It would then be of interest to correct the BSS measurements for the bare 252Cf at NIST and the AmBe at NRC using shadow cones. This would allow the BSS results to be compared directly with the fluence rate and ambient dose equivalent rates due to uncollided neutrons at those facilities.
J. ATANACKOVIC ET AL. 5. Perey, F. Least-squares dosimetry unfolding: the program STAY’SL. Tech. Rep. ORNL/TM-6062. Oak Ridge National Laboratory (1977). 6. Reginatto, M. and Goldhagen, P. MAXED, a computer code for maximum entropy deconvolution of multisphere neutron spectrometer data. Health Phys. J. 77(5), 579–583 (1999). 7. Reginatto, M. et al. Spectrum unfolding, sensitivity analysis and propagation of uncertainties with the maximum
entropy deconvolution code MAXED. Nucl. Instrum. Methods A 476, 242 –246 (2002). 8. MCNP—a general Monte Carlo N-Particle transport code. Tech. Rep. LA-UR-03-1987. Los Alamos National Laboratory (2008). 9. Wiegel, B. et al. Calculations of the response funtions of Bonner Spheres with a spherical 3He proportional counter using realistic detector model. Tech. Rep. Report PTB-N21. PTB (1988).
Downloaded from http://rpd.oxfordjournals.org/ at Harvard Library on April 22, 2015
220
