INTERNATIONAL COMPARISON EXERCISE ON NEUTRON SPECTRA UNFOLDING IN BONNER SPHERES SPECTROMETRY: PROBLEM DESCRIPTION AND PRELIMINARY ANALYSIS

INTERNATIONAL COMPARISON EXERCISE ON NEUTRON SPECTRA UNFOLDING IN BONNER SPHERES SPECTROMETRY:... Abstract This article describes the purpose, the proposed problems and the reference solutions of an international comparison on neutron spectra unfolding in Bonner spheres spectrometry, organised within the activities of EURADOS working group 6: computational dosimetry. The exercise considered four realistic situations: a medical accelerator, a workplace field, an irradiation room and a skyshine scenario. Although a detailed analysis of the submitted solutions is under preparation, the preliminary discussion of some physical aspects of the problem, e.g. the changes in the unfolding results due to the perturbation of the neutron field by the Bonner spheres, is presented. INTRODUCTION The Bonner sphere spectrometer (BSS)(1) has been well established in neutron spectrometry since the 1960s, mainly because of its nearly isotropic response and wide energy range (from thermal up to some GeV)(2–4). A BSS consists of a thermal neutron detector placed in the centre of moderating spheres of different sizes made of polyethylene. The combination of the thermal sensor and the moderating spheres provides different sensitivity to neutrons over a broad energy range so the neutron spectrum can be inferred from the readings of a set of spheres, by means of an unfolding procedure(5). The working group 6 (WG6: computational dosimetry) of EURADOS recently launched the ‘International comparison on neutron spectra unfolding in Bonner spheres spectrometry’ with the main purpose of surveying the usage of unfolding methods by considering four different scenarios. Overall, 21 participants from 15 countries worldwide submitted their solutions for the comparison. This article summarises the scope of the exercise, the proposed problems and the reference solutions. In addition, a brief description of the changes in the unfolding results due to the perturbation of the neutron field by the Bonner spheres is also presented. This effect may systematically affect the unfolding process by changing the unfolded field relative to the detector absent field. DESCRIPTION OF THE EXERCISE AND PROPOSED SCENARIOS The BSS consists of an idealised 3He thermal neutron detector at the centre of polyethylene spheres (density 0.947 g/cm3) of diameters 2˝, 2˝ + 1 mm Cd, 3˝, 3.5˝, 4˝, 4.5˝, 5˝, 6˝, 7˝, 8˝, 10˝ and 12˝. The detector is a pure 3He sphere with diameter 32 mm surrounded by a 0.5 mm thick steel case. The response matrix was calculated in the range from 10−9 to 20 MeV by Monte Carlo simulation using MCNPX 2.7 code(6) with the ENDF/B-VII neutron cross-section library and the thermal neutron S(α,β) cross-section data for polyethylene(7). Photonuclear (γ,n) reactions and the production of secondary neutrons have been also considered. No variance reduction technique was employed. The response matrix is shown in Figure 1. Figure 1. View largeDownload slide Calculated response matrix. Figure 1. View largeDownload slide Calculated response matrix. The response of the 3He detector was assumed to be the number of 3He(n,p) reactions and calculated using a modified f4 tally in the cell filled with 3He. For each considered scenario, both the neutron spectra at the measurement locations and the corresponding BSS counts were also directly calculated with MCNPX 2.7 following the same method used to obtain the response matrix. Neutron spectra have been obtained in absence of the Bonner spheres and correspond to the unperturbed neutron field. The participants were provided with the description of the scenarios as well as the simulated counts together with the corresponding uncertainties. Then, they were required to unfold the neutron spectra to obtain the energy distribution of fluence in a given energy binning, the ambient dose equivalent calculated using ICRP74/ICRU57 conversion coefficients(8, 9), the fluence averaged energy and the ambient dose equivalent averaged energy. The four scenarios are briefly described below. Medical LINAC A medical accelerator GE Saturne 43, 25 MV, is situated in the centre of a 7 × 7 × 3 m3 treatment room with concrete walls, as it is indicated in Figure 2. The accelerator head is mounted vertically with the top of the target located 2 m above the floor. The field size is 10 × 10 cm2. A 40 × 40 × 40 cm3 water phantom with PMMA walls, 1.5 cm thick, is located in the beam with the upper surface at 90 cm from the top of the target. Distance from target to isocentre is 1 m. Two measurement positions have been considered: one at the entrance of the maze (p1 in the figure) and the other at 1 m from the isocentre (p2 in the figure). The BSS counts for both points have been perturbed using a Poissonian distribution to simulate a counting uncertainty of 2%. Figure 2. View largeDownload slide Schematic view of the medical LINAC scenario, showing the position of the accelerator and the two measurement points within the treatment room: point p1 (at the entrance of the maze, 1 m distance to the walls) and point p2 (1 m from the isocentre). Figure 2. View largeDownload slide Schematic view of the medical LINAC scenario, showing the position of the accelerator and the two measurement points within the treatment room: point p1 (at the entrance of the maze, 1 m distance to the walls) and point p2 (1 m from the isocentre). Workplace field The source consists of an ISO 241Am–Be(10) source suspended in a stainless steel tube, which is clad with a lead shield. The steel tube extends from the floor to the ceiling of a 2.5 × 5.0 × 7.75 m3 metrology laboratory room, which has wooden panels covering all surfaces. Neutron absorbing material is placed behind the panels. The measurement point is fixed at the same height as the source (1.25 m), at 1 m distance from it. Adjacent to the steel tube, and in-line with the measurement point, is a water-filled container of ~50 cm depth that moderates the neutrons. The container is sufficiently wide to cover the solid-angle subtended at the source by the largest Bonner sphere, and is supported by a 2 cm thick wooden table at a height that gives equal coverage in the vertical direction. The simulated uncertainty of the measurements is 4%. Irradiation room with a radionuclide source The source consists of an ISO 241Am–Be source in the centre of an iron sphere (r = 10 cm) located in the centre of an 8 × 8 × 8 m3 irradiation room with 50 cm thick concrete walls, floor and ceiling. The measurement point is fixed at the same height as the source, at 4 m distance from the source along one diagonal of the room’s horizontal plane. The simulated uncertainty of the measurements is 2%. Skyshine scenario The fourth scenario attempts to simulate the situation of an environmental measurement at 100 m from a nuclear plant. The plant is a cylindrical room of height 10 m and radius 10 m. The walls of the plant are made from concrete and the roof is made from thin concrete tiles. A source of (α,n) neutrons is located in the middle of the building; the total activity of the source is unknown. The walls provide strong attenuation of the direct field so the main component of the field at 100 m is from air scattered neutrons. The Bonner spheres were placed with their centres 1.5 m above the ground. The simulated uncertainty of the measurements is 5%. PRELIMINARY RESULTS AND DISCUSSION Focusing on the LINAC scenario, Figure 3 shows the reference spectra (thick solid line) for the two measurement points together with unfolded solutions obtained using MAXED(11), GRAVEL(5) and FRUIT(12) codes (FRUIT has been used in guess spectrum mode). In all the cases, the guess spectrum was a calculated 25 MV spectrum taken from an IAEA compendium(13) to which a thermal Gaussian peak was added. As it can be seen, the unfolded spectra show a systematic discrepancy in the epithermal region (10 eV–10 keV) MeV for point p1 (Figure 3a, entrance of the maze) but not for point p2 (Figure 3b, 1 m from the isocentre).The effect can be appreciated more clearly in Figure 4 by comparing the BSS counts (directly Monte Carlo calculated as it has been explained in previous section) with the count values obtained by folding the unperturbed spectrum with the response matrix (folded counts). In Figure 4a (corresponding to LINAC point p1), the BSS counts (close squares) and the folded counts (open squares) show a systematic difference for spheres 3′–5′. This effect is due to the perturbation in the field caused by the BSS when the measurement point is close to a wall because actual response is affected by the neutron multi-scattering between the moderator of the spheres and the surrounding elements, sometimes called the ping-pong effect(14). As a consequence, the contributions to the neutron spectrum (without BSS perturbation) are different from the contributions to the BSS response. Figure 3. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 3. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 4. View largeDownload slide Medical LINAC scenario: comparison between the simulated BSS counts (solid squares), the folding of the reference spectrum with the response matrix (open squares) and the simulated unperturbed contribution to the BSS counts (plus symbols). Values calculated: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 4. View largeDownload slide Medical LINAC scenario: comparison between the simulated BSS counts (solid squares), the folding of the reference spectrum with the response matrix (open squares) and the simulated unperturbed contribution to the BSS counts (plus symbols). Values calculated: (a) at the entrance of the maze and (b) 1 m from the isocentre. As it can be seen in Figure 4, the perturbation effect depends on sphere size: 2′ sphere is too small to produce a significative change in the neutron field; large spheres (6′ and above) are mainly sensitive to fast neutrons so they are not affected by neutrons that has been multiple scattered by the sphere and the walls. To quantify the ping-pong effect, the unperturbed contribution to the BSS counts (i.e. the counts corresponding to the unperturbed neutron field) has been calculated following two steps: Save the neutron phase space in a spherical surface around the position of the BSS, without the BSS. Use the saved phase space to irradiate the BSS, without the other elements (treatment room, LINAC). As it can be seen also in Figure 4, the BSS counts corresponding to the unperturbed field (plus symbols) give practically the same values than folded counts (open squares), because both the response matrix and the reference spectra do not include any perturbation arising from the spheres. Moreover, the effect of perturbation is only significant for spheres 3′–5′ in point p1 (Figure 4a) but not in point p2 (Figure 4b). In all the other cases, the difference is compatible with the stochastic uncertainty of the calculations, without showing any trend. Unfolding the BSS counts due only to the unperturbed field (Figure 5) permits to obtain neutron spectra that agree pretty well with the reference ones. Figure 5. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained considering only the unperturbed field contribution to the BSS counts and using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 5. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained considering only the unperturbed field contribution to the BSS counts and using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Table 1 shows the neutron fluence, Φ, and ambient dose equivalent H*(10) obtained from the unfolded spectra for point p1 (entrance of the maze), using both the real BSS counts and the unperturbed contribution (discrepancies with the reference value are indicated between brackets). Although the ping-pong effect does not produce a significant change in the unfolded Φ and H*(10) (3 and 1% variations, respectively, are acceptable in most of the practical situation), the fluence distribution in three energy groups give a more accurate view of the phenomena. In particular, discrepancy for the fluence in the energy range between 0.4 eV and 0.1 MeV reduces from around 10–2%, being this the range corresponding to the 3′–5′ spheres where the perturbation contribution is more significant. Table 1. Medical LINAC scenario, point p1 (entrance of the maze): comparison of reference values for fluence and ambient dose equivalent with the corresponding values obtained from the unfolded spectra, using both the real BSS counts and the unperturbed contribution (discrepancy with the reference value is indicated between brackets). Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  View Large Table 1. Medical LINAC scenario, point p1 (entrance of the maze): comparison of reference values for fluence and ambient dose equivalent with the corresponding values obtained from the unfolded spectra, using both the real BSS counts and the unperturbed contribution (discrepancy with the reference value is indicated between brackets). Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  View Large CONCLUSIONS The comparison exercise on neutron spectra unfolding in Bonner spheres spectrometry described above is expected to allow gaining insight into the usage of unfolding methods by analysing the results obtained in realistic simulated scenarios. In particular, the effect on the unfolding results of the perturbation in the neutron field caused by the neutron multi-scattering between the Bonner spheres and the neighbouring elements has been quantified using Monte Carlo simulation. The applied method using the phase space around the BSS shows that, while discrepancy in unfolded fluence can reach up to 13% compared with the reference spectrum for the epithermal region, the discrepancy in ambient dose equivalent, H*(10), is very low (1%) as well in total fluence (3%). Thus and although correction factors could be calculated to take into account the effect of the perturbation, such effort seems not to be required for workplace monitoring where H*(10) is the main quantity to be determined. Acknowledgements The authors also thank Dr Marcel Reginatto and Dr Burkhard Weigel for their very valuable advice. Funding This work is been partially supported by EURADOS (European Radiation Dosimetry Group), within the activities of Working Group 6: Computational Dosimetry. REFERENCES 1 Bramblett, R. L., Ewing, R. I. and Bonner, T. W. A new type of neutron spectrometer. Nucl. Instrum. Methods Phys. Res. A  9, 1– 12 ( 1960). Google Scholar CrossRef Search ADS   2 Thomas, D. J. and Alevra, A. V. Bonner sphere spectrometers—a critical review. Nucl. Instr. Meth. Phys. Res. A  476, 12– 20 ( 2002). Google Scholar CrossRef Search ADS   3 Thomas, D. J. Neutron spectrometry. Radiat. Meas.  45, 1178– 1185 ( 2010). Google Scholar CrossRef Search ADS   4 Howell, R. M., Burgett, E. A., Wiegel, B. and Hertel, N. E. Calibration of a Bonner sphere extension (BSE) for high-energy neutron spectrometry. Radiat. Meas.  45, 1233– 1237 ( 2015). Google Scholar CrossRef Search ADS   5 Matzke, M. Unfolding procedures. Radiat. Prot. Dosim.  107, 155– 174 ( 2003). Google Scholar CrossRef Search ADS   6 Pelowitz, D. B. (editor). MCNPX User’s Manual Version 2.7, Report LA-CP-11–00438 ( 2011). 7 Chadwick, M. B. et al.  . ENDF/B-VII.0: next generation evaluated nuclear data library for nuclear science and technology. Nucl. Data Sheets  107, 2931– 3060 ( 2006). Google Scholar CrossRef Search ADS   8 International Commission on Radiological Protection (ICRP). Conversion coefficients for use in Radiological Protection against external radiation. ICRP Publication 74. Ann. ICRP 26 (3/4) ( 1996). 9 International Commission on Radiation Units and Measurements (ICRU). Conversion coefficients for use in radiological protection against external radiation. ICRU Report 57 (Bethesda, USA) ( 1998). 10 International Standarization Organization (ISO). Reference neutron radiations—Part 1: characteristics and methods of production. ISO 8529-1:2001 ( 2001). 11 Reginatto, M., Goldhagen, P. and Neumann, S. Spectrum unfolding, sensitivity analysis and propagation of uncertainties with the maximum entropy deconvolution code MAXED. Nucl. Instrum. Methods Phys. Res. A  476, 242– 246 ( 2002). Google Scholar CrossRef Search ADS   12 Bedogni, R., Domingo, C., Esposito, A. and Fernández, F. FRUIT: an operational tool for multisphere neutron spectrometry in workplaces. Nucl. Instrum. Methods Phys. Res. A  580, 1301– 1309 ( 2007). Google Scholar CrossRef Search ADS   13 International Atomic Energy Agency (IAEA). Compendium of neutron spectra and detector responses for radiation protection purposes. Tech. Report Series No. 318 (Vienna) ( 1990). 14 Matsumoto, T. et al.  . Measurement of neutron energy spectra behind shields for quasi-monoenergetic neutrons generated by 246 MeV and 389 MeV protons using a Bonner sphere spectrometer. Prog. Nucl. Sci. Technol.  4, 332– 336 ( 2014). Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Protection Dosimetry Oxford University Press

INTERNATIONAL COMPARISON EXERCISE ON NEUTRON SPECTRA UNFOLDING IN BONNER SPHERES SPECTROMETRY: PROBLEM DESCRIPTION AND PRELIMINARY ANALYSIS

Loading next page...
 
/lp/ou_press/international-comparison-exercise-on-neutron-spectra-unfolding-in-VjEPI8ecOT
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0144-8420
eISSN
1742-3406
D.O.I.
10.1093/rpd/ncy002
Publisher site
See Article on Publisher Site

Abstract

Abstract This article describes the purpose, the proposed problems and the reference solutions of an international comparison on neutron spectra unfolding in Bonner spheres spectrometry, organised within the activities of EURADOS working group 6: computational dosimetry. The exercise considered four realistic situations: a medical accelerator, a workplace field, an irradiation room and a skyshine scenario. Although a detailed analysis of the submitted solutions is under preparation, the preliminary discussion of some physical aspects of the problem, e.g. the changes in the unfolding results due to the perturbation of the neutron field by the Bonner spheres, is presented. INTRODUCTION The Bonner sphere spectrometer (BSS)(1) has been well established in neutron spectrometry since the 1960s, mainly because of its nearly isotropic response and wide energy range (from thermal up to some GeV)(2–4). A BSS consists of a thermal neutron detector placed in the centre of moderating spheres of different sizes made of polyethylene. The combination of the thermal sensor and the moderating spheres provides different sensitivity to neutrons over a broad energy range so the neutron spectrum can be inferred from the readings of a set of spheres, by means of an unfolding procedure(5). The working group 6 (WG6: computational dosimetry) of EURADOS recently launched the ‘International comparison on neutron spectra unfolding in Bonner spheres spectrometry’ with the main purpose of surveying the usage of unfolding methods by considering four different scenarios. Overall, 21 participants from 15 countries worldwide submitted their solutions for the comparison. This article summarises the scope of the exercise, the proposed problems and the reference solutions. In addition, a brief description of the changes in the unfolding results due to the perturbation of the neutron field by the Bonner spheres is also presented. This effect may systematically affect the unfolding process by changing the unfolded field relative to the detector absent field. DESCRIPTION OF THE EXERCISE AND PROPOSED SCENARIOS The BSS consists of an idealised 3He thermal neutron detector at the centre of polyethylene spheres (density 0.947 g/cm3) of diameters 2˝, 2˝ + 1 mm Cd, 3˝, 3.5˝, 4˝, 4.5˝, 5˝, 6˝, 7˝, 8˝, 10˝ and 12˝. The detector is a pure 3He sphere with diameter 32 mm surrounded by a 0.5 mm thick steel case. The response matrix was calculated in the range from 10−9 to 20 MeV by Monte Carlo simulation using MCNPX 2.7 code(6) with the ENDF/B-VII neutron cross-section library and the thermal neutron S(α,β) cross-section data for polyethylene(7). Photonuclear (γ,n) reactions and the production of secondary neutrons have been also considered. No variance reduction technique was employed. The response matrix is shown in Figure 1. Figure 1. View largeDownload slide Calculated response matrix. Figure 1. View largeDownload slide Calculated response matrix. The response of the 3He detector was assumed to be the number of 3He(n,p) reactions and calculated using a modified f4 tally in the cell filled with 3He. For each considered scenario, both the neutron spectra at the measurement locations and the corresponding BSS counts were also directly calculated with MCNPX 2.7 following the same method used to obtain the response matrix. Neutron spectra have been obtained in absence of the Bonner spheres and correspond to the unperturbed neutron field. The participants were provided with the description of the scenarios as well as the simulated counts together with the corresponding uncertainties. Then, they were required to unfold the neutron spectra to obtain the energy distribution of fluence in a given energy binning, the ambient dose equivalent calculated using ICRP74/ICRU57 conversion coefficients(8, 9), the fluence averaged energy and the ambient dose equivalent averaged energy. The four scenarios are briefly described below. Medical LINAC A medical accelerator GE Saturne 43, 25 MV, is situated in the centre of a 7 × 7 × 3 m3 treatment room with concrete walls, as it is indicated in Figure 2. The accelerator head is mounted vertically with the top of the target located 2 m above the floor. The field size is 10 × 10 cm2. A 40 × 40 × 40 cm3 water phantom with PMMA walls, 1.5 cm thick, is located in the beam with the upper surface at 90 cm from the top of the target. Distance from target to isocentre is 1 m. Two measurement positions have been considered: one at the entrance of the maze (p1 in the figure) and the other at 1 m from the isocentre (p2 in the figure). The BSS counts for both points have been perturbed using a Poissonian distribution to simulate a counting uncertainty of 2%. Figure 2. View largeDownload slide Schematic view of the medical LINAC scenario, showing the position of the accelerator and the two measurement points within the treatment room: point p1 (at the entrance of the maze, 1 m distance to the walls) and point p2 (1 m from the isocentre). Figure 2. View largeDownload slide Schematic view of the medical LINAC scenario, showing the position of the accelerator and the two measurement points within the treatment room: point p1 (at the entrance of the maze, 1 m distance to the walls) and point p2 (1 m from the isocentre). Workplace field The source consists of an ISO 241Am–Be(10) source suspended in a stainless steel tube, which is clad with a lead shield. The steel tube extends from the floor to the ceiling of a 2.5 × 5.0 × 7.75 m3 metrology laboratory room, which has wooden panels covering all surfaces. Neutron absorbing material is placed behind the panels. The measurement point is fixed at the same height as the source (1.25 m), at 1 m distance from it. Adjacent to the steel tube, and in-line with the measurement point, is a water-filled container of ~50 cm depth that moderates the neutrons. The container is sufficiently wide to cover the solid-angle subtended at the source by the largest Bonner sphere, and is supported by a 2 cm thick wooden table at a height that gives equal coverage in the vertical direction. The simulated uncertainty of the measurements is 4%. Irradiation room with a radionuclide source The source consists of an ISO 241Am–Be source in the centre of an iron sphere (r = 10 cm) located in the centre of an 8 × 8 × 8 m3 irradiation room with 50 cm thick concrete walls, floor and ceiling. The measurement point is fixed at the same height as the source, at 4 m distance from the source along one diagonal of the room’s horizontal plane. The simulated uncertainty of the measurements is 2%. Skyshine scenario The fourth scenario attempts to simulate the situation of an environmental measurement at 100 m from a nuclear plant. The plant is a cylindrical room of height 10 m and radius 10 m. The walls of the plant are made from concrete and the roof is made from thin concrete tiles. A source of (α,n) neutrons is located in the middle of the building; the total activity of the source is unknown. The walls provide strong attenuation of the direct field so the main component of the field at 100 m is from air scattered neutrons. The Bonner spheres were placed with their centres 1.5 m above the ground. The simulated uncertainty of the measurements is 5%. PRELIMINARY RESULTS AND DISCUSSION Focusing on the LINAC scenario, Figure 3 shows the reference spectra (thick solid line) for the two measurement points together with unfolded solutions obtained using MAXED(11), GRAVEL(5) and FRUIT(12) codes (FRUIT has been used in guess spectrum mode). In all the cases, the guess spectrum was a calculated 25 MV spectrum taken from an IAEA compendium(13) to which a thermal Gaussian peak was added. As it can be seen, the unfolded spectra show a systematic discrepancy in the epithermal region (10 eV–10 keV) MeV for point p1 (Figure 3a, entrance of the maze) but not for point p2 (Figure 3b, 1 m from the isocentre).The effect can be appreciated more clearly in Figure 4 by comparing the BSS counts (directly Monte Carlo calculated as it has been explained in previous section) with the count values obtained by folding the unperturbed spectrum with the response matrix (folded counts). In Figure 4a (corresponding to LINAC point p1), the BSS counts (close squares) and the folded counts (open squares) show a systematic difference for spheres 3′–5′. This effect is due to the perturbation in the field caused by the BSS when the measurement point is close to a wall because actual response is affected by the neutron multi-scattering between the moderator of the spheres and the surrounding elements, sometimes called the ping-pong effect(14). As a consequence, the contributions to the neutron spectrum (without BSS perturbation) are different from the contributions to the BSS response. Figure 3. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 3. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 4. View largeDownload slide Medical LINAC scenario: comparison between the simulated BSS counts (solid squares), the folding of the reference spectrum with the response matrix (open squares) and the simulated unperturbed contribution to the BSS counts (plus symbols). Values calculated: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 4. View largeDownload slide Medical LINAC scenario: comparison between the simulated BSS counts (solid squares), the folding of the reference spectrum with the response matrix (open squares) and the simulated unperturbed contribution to the BSS counts (plus symbols). Values calculated: (a) at the entrance of the maze and (b) 1 m from the isocentre. As it can be seen in Figure 4, the perturbation effect depends on sphere size: 2′ sphere is too small to produce a significative change in the neutron field; large spheres (6′ and above) are mainly sensitive to fast neutrons so they are not affected by neutrons that has been multiple scattered by the sphere and the walls. To quantify the ping-pong effect, the unperturbed contribution to the BSS counts (i.e. the counts corresponding to the unperturbed neutron field) has been calculated following two steps: Save the neutron phase space in a spherical surface around the position of the BSS, without the BSS. Use the saved phase space to irradiate the BSS, without the other elements (treatment room, LINAC). As it can be seen also in Figure 4, the BSS counts corresponding to the unperturbed field (plus symbols) give practically the same values than folded counts (open squares), because both the response matrix and the reference spectra do not include any perturbation arising from the spheres. Moreover, the effect of perturbation is only significant for spheres 3′–5′ in point p1 (Figure 4a) but not in point p2 (Figure 4b). In all the other cases, the difference is compatible with the stochastic uncertainty of the calculations, without showing any trend. Unfolding the BSS counts due only to the unperturbed field (Figure 5) permits to obtain neutron spectra that agree pretty well with the reference ones. Figure 5. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained considering only the unperturbed field contribution to the BSS counts and using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Figure 5. View largeDownload slide Reference neutron spectra (thick solid line) and results obtained considering only the unperturbed field contribution to the BSS counts and using different unfolding codes: Maxed (solid line), Gravel (dashed line) and FRUIT (dotted line), for the two measurement points in the medical LINAC scenario: (a) at the entrance of the maze and (b) 1 m from the isocentre. Table 1 shows the neutron fluence, Φ, and ambient dose equivalent H*(10) obtained from the unfolded spectra for point p1 (entrance of the maze), using both the real BSS counts and the unperturbed contribution (discrepancies with the reference value are indicated between brackets). Although the ping-pong effect does not produce a significant change in the unfolded Φ and H*(10) (3 and 1% variations, respectively, are acceptable in most of the practical situation), the fluence distribution in three energy groups give a more accurate view of the phenomena. In particular, discrepancy for the fluence in the energy range between 0.4 eV and 0.1 MeV reduces from around 10–2%, being this the range corresponding to the 3′–5′ spheres where the perturbation contribution is more significant. Table 1. Medical LINAC scenario, point p1 (entrance of the maze): comparison of reference values for fluence and ambient dose equivalent with the corresponding values obtained from the unfolded spectra, using both the real BSS counts and the unperturbed contribution (discrepancy with the reference value is indicated between brackets). Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  View Large Table 1. Medical LINAC scenario, point p1 (entrance of the maze): comparison of reference values for fluence and ambient dose equivalent with the corresponding values obtained from the unfolded spectra, using both the real BSS counts and the unperturbed contribution (discrepancy with the reference value is indicated between brackets). Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  Quantity  Reference values  Unfolded values  Real BSS counts  Unperturbed BSS counts  MAXED  GRAVEL  FRUIT  MAXED  GRAVEL  FRUIT  Φtotal (×106 cm−2 Gy−1)  6.79  7.01 (3.3%)  6.97 (2.6%)  6.94 (2.2%)  6.84 (0.8%)  6.80 (0.1%)  6.81 (0.3%)  ΦE < 0.4 eV  3.08  3.20 (3.8%)  3.17 (2.7%)  3.10 (0.6%)  3.19 (3.6%)  3.13 (1.6%)  3.14 (1.9%)  Φ0.4 eV < E < 0.1 MeV  1.92  2.12 (10.4%)  2.11 (10.1%)  2.18 (13.5%)  1.88 (−1.9%)  1.96 (1.9%)  1.97 (2.6%)  ΦE > 0.1 MeV  1.79  1.69 (−5.2%)  1.69 (−5.5%)  1.65 (−7.8%)  1.76 (−1.3%)  1.71 (−4.2%)  1.74 (−2.8%)  H*(10) (mSv Gy−1)  63.4  62.7 (−1.1%)  62.5 (−1.3%)  62.4 (−1.6%)  63.7 (0.5%)  63.5 (0.2%)  63.7 (0.5%)  View Large CONCLUSIONS The comparison exercise on neutron spectra unfolding in Bonner spheres spectrometry described above is expected to allow gaining insight into the usage of unfolding methods by analysing the results obtained in realistic simulated scenarios. In particular, the effect on the unfolding results of the perturbation in the neutron field caused by the neutron multi-scattering between the Bonner spheres and the neighbouring elements has been quantified using Monte Carlo simulation. The applied method using the phase space around the BSS shows that, while discrepancy in unfolded fluence can reach up to 13% compared with the reference spectrum for the epithermal region, the discrepancy in ambient dose equivalent, H*(10), is very low (1%) as well in total fluence (3%). Thus and although correction factors could be calculated to take into account the effect of the perturbation, such effort seems not to be required for workplace monitoring where H*(10) is the main quantity to be determined. Acknowledgements The authors also thank Dr Marcel Reginatto and Dr Burkhard Weigel for their very valuable advice. Funding This work is been partially supported by EURADOS (European Radiation Dosimetry Group), within the activities of Working Group 6: Computational Dosimetry. REFERENCES 1 Bramblett, R. L., Ewing, R. I. and Bonner, T. W. A new type of neutron spectrometer. Nucl. Instrum. Methods Phys. Res. A  9, 1– 12 ( 1960). Google Scholar CrossRef Search ADS   2 Thomas, D. J. and Alevra, A. V. Bonner sphere spectrometers—a critical review. Nucl. Instr. Meth. Phys. Res. A  476, 12– 20 ( 2002). Google Scholar CrossRef Search ADS   3 Thomas, D. J. Neutron spectrometry. Radiat. Meas.  45, 1178– 1185 ( 2010). Google Scholar CrossRef Search ADS   4 Howell, R. M., Burgett, E. A., Wiegel, B. and Hertel, N. E. Calibration of a Bonner sphere extension (BSE) for high-energy neutron spectrometry. Radiat. Meas.  45, 1233– 1237 ( 2015). Google Scholar CrossRef Search ADS   5 Matzke, M. Unfolding procedures. Radiat. Prot. Dosim.  107, 155– 174 ( 2003). Google Scholar CrossRef Search ADS   6 Pelowitz, D. B. (editor). MCNPX User’s Manual Version 2.7, Report LA-CP-11–00438 ( 2011). 7 Chadwick, M. B. et al.  . ENDF/B-VII.0: next generation evaluated nuclear data library for nuclear science and technology. Nucl. Data Sheets  107, 2931– 3060 ( 2006). Google Scholar CrossRef Search ADS   8 International Commission on Radiological Protection (ICRP). Conversion coefficients for use in Radiological Protection against external radiation. ICRP Publication 74. Ann. ICRP 26 (3/4) ( 1996). 9 International Commission on Radiation Units and Measurements (ICRU). Conversion coefficients for use in radiological protection against external radiation. ICRU Report 57 (Bethesda, USA) ( 1998). 10 International Standarization Organization (ISO). Reference neutron radiations—Part 1: characteristics and methods of production. ISO 8529-1:2001 ( 2001). 11 Reginatto, M., Goldhagen, P. and Neumann, S. Spectrum unfolding, sensitivity analysis and propagation of uncertainties with the maximum entropy deconvolution code MAXED. Nucl. Instrum. Methods Phys. Res. A  476, 242– 246 ( 2002). Google Scholar CrossRef Search ADS   12 Bedogni, R., Domingo, C., Esposito, A. and Fernández, F. FRUIT: an operational tool for multisphere neutron spectrometry in workplaces. Nucl. Instrum. Methods Phys. Res. A  580, 1301– 1309 ( 2007). Google Scholar CrossRef Search ADS   13 International Atomic Energy Agency (IAEA). Compendium of neutron spectra and detector responses for radiation protection purposes. Tech. Report Series No. 318 (Vienna) ( 1990). 14 Matsumoto, T. et al.  . Measurement of neutron energy spectra behind shields for quasi-monoenergetic neutrons generated by 246 MeV and 389 MeV protons using a Bonner sphere spectrometer. Prog. Nucl. Sci. Technol.  4, 332– 336 ( 2014). Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

Radiation Protection DosimetryOxford University Press

Published: Jan 29, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off