Proc. Int. Symp. on Radiation Detectors and Their Uses (ISRD2016) https://doi.org/10.7566/jpscp.11.050002 High Sensitive Neutron-detection by using a Self-activation of Iodine-containing Scintillators for the Photo-neutron Monitoring around X-ray Radiotherapy Machines Akihiro NOHTOMI*, Genichiro WAKABAYASHI 1, Hiroyuki KINOSHITA, Soichiro HONDA, Ryosuke KURIHARA, Junichi FUKUNAGA 2, Yoshiyuki UMEZU 2, Yasuhiko NAKAMURA 2, Saiji OHGA 2, and Katsumasa NAKAMURA 3 Department of Health Sciences, Kyushu University, Fukuoka 812-8582, Japan 1 Kinki University Atomic Energy Research Institute, Higashi-Osaka 577-8502, Japan 2 Department of Radiology, Kyushu University Hospital, Fukuoka 812-8582, Japan 3 Department of Radiation Oncology, Hamamatsu University School of Medicine, Hamamatsu 431-3192, Japan E-mail: nohtomi@hs.med.kyushu-u.ac.jp (Received February 24, 2016) A novel method for evaluating the neutron dose-equivalent as well as neutron fluence around high-energy X-ray radiotherapy machines has been proposed and examined by using the self-activation of a CsI scintillator. Several filtering conditions were used to extract energy information of the neutron field. The shapes of neutron energy spectra were assumed to be practically unchanged at each three energy s (thermal, epi-thermal and fast s) for different irradiations around an X-ray linac whose acceleration potential was fixed to be a certain value. In order to know the actual neutron energy spectrum, an unfolding process was carried out for saturated activities of 128 I generated inside the CsI scintillator under different filtering conditions; the response function matrix for each filtering condition was calculated by a Monte Carlo simulation. As the result, neutron dose-equivalent was estimated to be 0.14 [msv/gy] at 30 cm from the isocenter of linac. It has been revealed that fast neutron component dominated the total dose-equivalent. KEYWORDS: scintillator, 128 I neutron dose equivalent, X-ray linac, the self-activation method, CsI 1. Introduction It has been known that neutrons are generated in the radio-therapeutic treatment with high-energy accelerators. Effective energies of photo-neutrons leaked from X-ray therapy machines range from a few hundred kev to a few MeV. Thus, it is of concern that such photo-neutrons can increase the risk of secondary cancer due to high biological effectiveness. For measuring them, so-called the activation method has been conventionally used with activation materials such as 197 Au. However, the activation method is generally time-consuming and not practical for routine monitoring in a hospital. 050002-1 2016 The Author(s) This article is available under the terms of the Creative Commons Attribution 4.0 License. Any further distribution of this work must maintain attribution to the author(s) and the title of the article, journal citation, and DOI.
050002-2 In our recent paper [1], applicability of the self-activation of an NaI scintillator has been studied for the photo-neutron monitoring around an X-ray linac. In the method, neutron fluence rates are derived from the measurement of 128 I activity generated in an NaI scintillator; ß-rays of 128 I are detected efficiently by the scintillator itself. So, high-sensitive neutron detection is realized with on-line read out. As an extension of this, the use of CsI, in place of NaI, may be advantageous in terms of manufacturing, handling and price. 2. Materials and Methods 2.1 Experimental The response of a CsI detector [2.5 x 2.5 x 2.5 cm 3 ] to photo-neutrons was investigated around a 10 MV clinical linac (Varian Clinac 21EX). The detector was arranged at 30 cm from the isocenter position outside of the primary irradiation fileld. To extract the neutron energy information, two cylindrical polyethylene moderators (P.E.) having different dimensions ( 13 cm x H 13 cm, 20 cm x H 20 cm) were used for energy filtering. A 1 mm-thick Cd foil was also used to eliminate thermal components. After the termination of each irradiation (90 Gy at the isocenter), the pulse height spectra were recorded by a MCA every 1 min [1]. From the decay curves of count rate, the initial count rates were derived by fitting them with an exponential function having half life of 25 min [2]. The initial count rates were converted to the saturated activities A sat. The evaluated saturated activities are summarized in Table I. Table I. Evaluated saturated activities A sat for different filtering conditions. Filtering condition A sat [Bq] Bare 4750 Cd foil (1 mm-t) 3560 13 cm P.E. 17200 20 cm P.E. 11800 2.2 Evaluation of neutron energy spectrum and dose equivalent The saturated activity A sat is expressed by the following equation of neutron activation analysis, A sat = N V S (1) where is neutron fluence rate [n/(s cm 2 )], N number density of iodine atom in CsI
050002-3 crystal [1/cm 3 ], differential capture cross section of 127 I (n, ) 128 I [cm 2 ], V volume of CsI crystal [cm 3 ], S self-shielding correction factor. In eq. (1), the values of and S both depend on neutron energy spectrum, and also on detector dimensions to some extent. So, if actual neutron energy spectrum is known, the term of N VS can be calculated by a proper Monte Carlo simulation. As a general observation that we have made, the shapes of partial neutron energy spectra are practically unchanged at each three energy s (thermal, epi-thermal and fast s) and only the ratio among them varies with the situation for different irradiations; such tendency is actually common for other measurements and calculations reported in many research papers, for example [3-5]. Therefore, under this practical assumption, neutron energy spectra (E) are approximated by the following three formulae at each energy in the present study, (E) = E 1.5 exp( - E / 2.53 x 10-8 ) for E (MeV) < 5 x 10-7 : thermal (2) (E) = 1 / E for 5 x 10-7 E (MeV) < 1 x 10-2 : epi-thermal (3) (E) = E 0.3 exp( - E / 0.32 ) for 1 x 10-2 E (MeV) : fast (4) where eq.(2) is Maxwell distribution at room temperature, 0.0253 ev. Equation (3) is referred from the assumption used in ref. [6]. Equation (4) is estimated by fitting a simulated spectrum based on the Monte Carlo calculation for a 10 MV linac [7]. By using those partial energy spectra of eqs. (2) - (4), the detector response R = N VS in eq. (1) was calculated by the Monte Carlo simulation code PHITS [8] for each filtering condition as indicated in Table II. In order to know actual neutron energy spectrum, those evaluations were used as response function matrix R ij for unfolding the measurements of A sat with different filtering conditions as follows, ( A sat ) i = R ij j (5) j where i indicates the filtering condition and j the neutron energy. We successfully used "OpenSolver for Excel, version 2.7.1" to carry out this unfolding process [9].
Neutron Fluence Rate (n/s/cm 2 / Lethargy) Proceedings of International Symposium on Radiation Detectors and Their Uses (ISRD2016) 050002-4 Table II. Evaluated response function matrix R ij. Energy j Filtering condition i Thermal Epi-thermal Fast Bare R = 0.54 0.41 0.033 Cd foil (1 mm-t) 0.00 0.39 0.032 13 cm P.E. 0.23 0.40 0.42 20 cm P.E. 0.13 0.20 0.42 As the result of this unfolding, fluence rates were determined to be 2.0 x 10 3 [n/(s cm 2 )] for thermal, 6.8 x 10 3 [n/(s cm 2 )] for epi-thermal and 2.9 x 10 4 [n/(s cm 2 )] for fast. Finally-obtained neutron energy spectrum is shown in Fig. 1. By convoluting the energy spectrum with dose equivalent conversion factors given in ICPR Pub. 74 [10], ambient dose equivalent H*(10) was calculated at each energy. As indicated in Table III, total neutron dose equivalent was evaluated to be 0.14 [msv/gy], which was almost dominated by the fast neutron component. The result is consistent with other evaluations found in research papers [7, 11]. 10MV Linac Table III. Evaluated neutron ambient dose equivalent H*(10) per unit Gy to isocenter. Energy Ambient dose equivalent [msv/gy] Thermal 4.12 x 10 4 Epi-thermal 1.40 x10 3 Fast 1.38 x 10 1 Total 1.40 x 10 1 Fig. 1 Evaluated neutron energy spectrum.
050002-5 3. Conclusion A novel method has been proposed for evaluating the dose-equivalent as well as neutron fluence by assuming the partial neutron energy-spectra at each energy of thermal, epi-thermal and fast neutron. The self-activation of iodine in a CsI scintillation detector was successfully applied for the evaluation of neutron dose equivalent around a 10 MV X-ray radiotherapy machine. The use of CsI, in place of NaI, may be advantageous for the application of self-activation method in terms of manufacturing, handling and price. Acknowledgment This study was performed in part under the Cooperative Research at Kinki University Reactor supported by the Graduate School of Engineering, Osaka University. The first author thanks Mr. R. Kakino of the Department of Health Sciences, Kyushu University for his discussions on data treatment. References [1] G. Wakabayashi, A. Nohtomi, E. Yahiro, T. Fujibuchi, J. Fukunaga, Y. Umezu, Y. Nakamura, K. Nakamura, M. Hosono and T. Itoh: Radiol. Phys. Techno. 8 (2015) 125. [2] A. Nohtomi and G. Wakabayashi: Nucl. Instrum. Meth. A800 (2015) 6. [3] A. Esposito, R. Bedogni, L. Lembo and M. Morelli: Radiation Measurements 43 (2008) 2. [4] J. Pena, L. Franco, F. Gomez, A. Iglesias, J. Pardo and M. Pombar: Phys. Med. Biol. 50 (2005) 5921. [5] H. M. Garnica-Garza: Phys. Med. Biol. 50 (2005) 531. [6] S. Yamaguchi, H. Hanada, K. Igarashi and G. Irie: J. Jpn. Radiol. Society 42 (1982) 783. [7] K. Yabuta, H. Monzen, M. Tamura, T. Tsuruta, T, Itoh, A. Nohtomi and K. Nishimura: Jpn. J. Med. Phys. 34 (2014) 139. [8] T. Sato, K. Niita, N. Matsuda, S. Hashimoto, Y. Iwamoto, S. Noda, T. Ogawa, H. Iwase, H. Nakashima, T. Fukahori, K. Okumura, T. Kai, S. Chiba, T. Furuta and L. Sihver: J. Nucl. Sci. Technol. 50 (2013) 913. [9] Web site of Open Solver for Excel http://opensolver.org (2016. 2. 1) [10] ICRP, 1996. Conversion Coefficients for use in Radiological Protection against External Radiation. ICRP Publication 74. Ann. ICRP 26 (3-4). [11] K.R. Kase, X.S. Mao, W.R. Nelson, J.C. Liu, J.H. Kleck and M. Elsalim: Health Physics 74 (1998) 38.