Science, Ami, Ami-machi, Inashiki-gun, Ibaraki , Japan. Ibaraki , Japan. Japan. *Corresponding author:

Transcription

1 Radiation Protection Dosimetry (2012), Vol. 148, No. 2, pp Advance Access publication 10 February 2011 doi: /rpd/ncr005 INDUCED RADIOACTIVE NUCLIDES OF 10-MeV RADIOTHERAPY ACCELERATORS DETECTED BY USING A PORTABLE HP-Ge SURVEY METER Toshioh Fujibuchi 1,2, *, Satoshi Obara 1, Ichiro Yamaguchi 3, Masaya Oyama 4, Hiroshi Watanabe 5, Takeji Sakae 2 and Kazuaki Katoh 6 1 Department of Radiological Sciences, School of Health Sciences, Ibaraki Prefectural University of Health Science, Ami, Ami-machi, Inashiki-gun, Ibaraki , Japan 2 Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tennodai, Tsukuba, Ibaraki , Japan 3 National Institute of Public Health, Minami, Wako-shi, Saitama , Japan 4 National Hospital Organization Tokyo Medical Center, Higashigaoka, Meguro, Tokyo , Japan 5 Yokohama Rousai Hospital, 3211 Kodzukue-cho, Kohuku-ku, Yokohama-shi, Kanagawa , Japan 6 Research Institute of Sophisticated Subjects, D , Kenkyu-gakuen, Tsukuba, Ibaraki , Japan *Corresponding author: fujibuchi@ipu.ac.jp Received May , revised December , accepted January The radioactivation of linear accelerator components for radiation therapy is interest for radiation protection in general, and particularly, when decommissioning these structures. The energy spectra of gamma rays emitted from the heads of two accelerator models, EXL-15SP and Clinac ix, after 10-MeV X-ray irradiation, were measured using a high-purity germanium semiconductor survey meter. After spectrum analyses, activities of 24 Na, 28 Al, 54 Mn, 56 Mn, 57 Ni, 58 Co, 60 Co, 64 Cu, 65 Zn, 122 Sb, 124 Sb, 181 W, 187 W, 196 Au, and 198 Au were detected. One centimetre deep dose-equivalent rate of the heads of the linear accelerator was measured using the survey meter. The dose rate decreased to 10 % of its initial rate after 1 week. Longterm activations were few, the radioactivity level was low, and a cooling time of several days was effective for reducing dose rate to an acceptable level for decommissioning. INTRODUCTION Currently in radiotherapy practice, both high-energy X-rays and electron beams are commonly used in the treatment of malignant tumors. These beams are generated by linear accelerators. The high-energy beams enable radiation to penetrate deeply into tumors and protect against skin damage. At radiotherapy facilities that use linear accelerators, neutrons are produced primarily through the interaction of highenergy photons with high-z materials such as tungsten or lead. In turn, these neutrons activate the structure of the linear accelerator through neutron capture reactions adding previous photonuclear reactions. Because linear accelerators are widely employed in radiotherapy, more information concerning the radioactivation of accelerator components would be of interest for radiation protection in general, and particularly when decommissioning these structures (1 5). The degree of radioactivation by neutrons varies according to the materials used in the construction (1 5) of the linear accelerator. Some reports have described the type and quantity of radiation produced by obtaining measurements during the dismantling of structures; however, only one report has evaluated the relatively long half-lived induced radioactivity found in 10-MeV machines (6). For the radiotherapy machine, it is important to confirm what nuclides were generated. The present study aims to confirm this by measurements. The energy spectrum of gamma rays emitted from the heads of two linear accelerator models after 10-MeV X-ray irradiation was measured, using a high-purity germanium (HP-Ge) semiconductor survey meter. Moreover, the 1-cm dose equivalent was measured to estimate the exposure of workers who do maintenance and demolishing work. MATERIALS AND METHODS Accelerators involved in the study EXL-15SP (Mitsubishi Electric, Tokyo, Japan) and Clinac ix (Varian Medical Systems, Palo Alto, USA) linear accelerators were evaluated. The EXL- 15SP was installed at the Ibaraki prefectural university of health sciences in 1995, and the operating period of the Clinac ix which installed at the Tsukuba University hospital is 1 month. The maximum acceleration energy of each machine is # The Author Published by Oxford University Press. All rights reserved. For Permissions, please journals.permissions@oup.com

2 INDUCED RADIOACTIVE NUCLIDES OF 10-MEV LINAC 10-MeV. The geometry of all machines provides a focus-isocentre distance of 100 cm. Gamma spectroscopy from the heads of linear accelerators after irradiation Most of the activation products will emit gamma radiation during their decay, and are thus accessible to gamma spectroscopy. Unfortunately, scintillation detectors do not provide the spectral resolution required to analyse complicated spectra. On the other hand, high-resolution semiconductor detectors are traditionally stationary due to the mechanical sensitivity and the need of permanent cooling. The gamma-ray spectrum at the surface of the accelerator heads was measured with the HP-Ge detector (Detective-EX; ORTEC, Oak Ridge, USA), which is an electric cooling type of survey meter (shown in Figure 1a). The detector was calibrated in terms of energy scale and absolute efficiency with standard gamma sources composed of 137 Cs. The spectrometer was installed on a trolley and put in operational mode outside of the treatment room to avoid radiation damage during beam-on. The accelerator was programmed to deliver a 5000 monitor units (MU) 10 MeV photon beam at the maximum output rate and a gantry angle of 908. During the measurement, the field size during and after irradiation was set to 1010 cm, assumed to be an average treatment field size. The detector was covered with a 5-cm-thick lead block to eliminate natural radiations and gamma rays emitted from the surrounding activated material (Figure 1b). The duration of data acquisition for each measurement was set at 60 with 10 h intervals. Radioactivity concentration of 60 Co in an accelerator head Radionuclides were identified by photon energies and confirmed by comparing the rate of decrease of the peak height with the half-life. The count efficiency was calculated under two different geometrical settings as mentioned below, viz. wide distribution and localised distribution, by using EGS5 (7). The iron head was assumed to have a cylindrical shape, 20 cm in radius and 90 cm in height. The distance from the surface of the head to the detector was set to 10 cm. At the wide distribution setting, 60 Co was set uniformly to a depth of 4 5 cm from the surface in the column, while at the localised distribution setting, 60 Co was set to a depth of 0 1 cm from the surface in the centre of the column, with a 0.5 cm radius. The Monte Carlo technique was used to track the energy deposition of photons emitted from 60 Co in the head. The cut-off energy for photons and electrons was 10 kev. One centimetre deep dose-equivalent rate measurement in the heads of linear accelerator for a week For 1 week after 5000 MU irradiation, the 1-cmdeep dose-equivalent rate of the heads of the EXL-15SP was measured using an NaI scintillator type survey meter (TCS-161, Aloka, Tokyo, Japan) (Figure 2). The distance from the target to the top of the scintillator detector was 20 and 50 cm. This survey meter is calibrated in terms of 1-cm-deep dose-equivalent rate by the substitution method with standard detector by using a 137 Cs gamma source (8). Figure 1. HP-Ge semiconductor survey meter. (a) Overview and (b) measuring head of linear accelerator. Figure 2. Measurement in the heads of linear accelerator with NaI survey meter for a week. 169

3 RESULTS Measurement of gamma-ray energy spectra from the heads of linear accelerators after irradiation The resultant gamma-ray energy spectra from the EXL-15SP and Clinac ix heads are shown in Figures 3 and 4. The main peaks were 58 ( 181 W), 134, 479, 551, 618, 686, 744, 772 ( 187 W), 510 (annihilation), 332, 355 ( 196 Au), 412 ( 198 Au), 846 ( 56 Mn), 1377 ( 57 Ni), 1692 ( 124 Sb) kev, detected EXL-15SP only were 835 ( 54 Mn), 811 ( 58 Co), 1115 ( 65 Zn), 1173, 1332 ( 60 Co) kev and Clinac ix only were 1097, 1293, 1508 ( 116m In), 1346 ( 64 Cu), 1370 ( 24 Na), 1778 ( 28 Al) kev (see Table 1). There were differences in the radionuclides produced by each model. Annihilation may be from 57 Ni, 64 Cu, 65 Zn, 122 Sb and 196 Au. The half-life of annihilation was h, i.e. contributions of 64 Cu are large. Estimate radioactivity concentration of 60 Co The calculation results showed that the radioactivity concentrations of 60 Co were 10 mbq g 21 in the activated region for the wide distribution setting and 3 mbq g 21 in the activated region for the localised T. FUJIBUCHI ET AL. distribution setting. In each setting, the total amount of induced 60 Co was,0.1 Bq. Dose equivalent measurements in the heads of linear accelerator Results of 1-cm-deep dose equivalent measurements for a week are shown in Figure 5. The dose rate decreased to 10 % in 1 week. The dose 20 cm from the target was 10 times as high as that at 50 cm. From the spectral data of EXL-15SP, the counts of 54 Mn, 124 Sb, 60 Co, 181 W, 187 W, 198 Au and annihilation were high. The dose equivalent from these spectral data for comparison with the scintillator measurements was calculated. ICRP publication 74 (9) for fluence-to-dose equivalent conversion data was used. The half-life of annihilation calculated from the spectral data was h. Because the measurement points were different, the dose equivalent values were normalised at 25 h from the time of irradiation. The relative strengths of the various emissions at some post-irradiation reference time were given as part of this comparison. The results are shown in Figure 6. Figure 3. The gamma energy spectra measured by EXL-15SP linear accelerators. 170

4 INDUCED RADIOACTIVE NUCLIDES OF 10-MEV LINAC Figure 4. The gamma energy spectra measured by Clinac ix linear accelerators. Table 1 Gamma-ray energy and nuclides from the EXL-15SP and Clinac ix heads. Radionuclide Production Decay Main, gamma-ray energy (kev) Half-life 181 W 187 W 196 Au 198 Au 56 Mn 82 Br 122 Sb 124 Sb (EXL-15SP only) 54 Mn 57 Ni 58 Co 60 Co 65 Zn (Clinac ix only) 24 Na 28 Al 64 Cu 116m In 182 W(g, n) 181 W EC d 186 W(n, g) 187 W b 2 134, 479, 551, 618, 686, 744, h 197 Au(g, n) 196 Au b þ 332, d 197 Au(n, g) 198 Au b d 55 Mn(n, g) 56 Mn b 2 846, 1810, h 81 Br(n, g) 82 Br b 2 554, 776, 828, 1044, 1317, h 121 Sb(n, g) 122 Sb b þ /b d 123 Sb(n, g) 124 Sb b 2 603, 646, 723, d 55 Mn(g, n) 54 Mn, 54 Fe(n, p) 54 Mn EC d 58 Ni(g, n) 57 Ni b þ h 59 Co(g, n) 58 Co, 58 Ni(n, p) 58 Co EC d 61 Ni(g, p) 60 Co, 60 Ni(n, p) 60 Co b , y 66 Zn(g, n) 65 Zn EC, b þ d 27 Al(n, a) 24 Na b h 27 Al(n, g) 28 Al b m 63 Cu(n, g) 64 Cu b þ /b h 115 In(n, g) 116m In b , 1293, min DISCUSION Various kinds of materials are used as component parts of linear accelerators, such as those around the beam line with the target, the collimator and the flattening filter. Radioactivation caused by the highenergy beams used for radiotherapy results from the photonuclear reaction with the primary beam and 171

5 T. FUJIBUCHI ET AL. Table 2. Threshold energy of photonuclear reactions (7). Nucleus (g, n) (g, p) Nucleus (g, n) (g, p) 27 Al Ca Mn Fe Fe Co Ni Ni Cu Cu Zn Zn Sb Ta W W W W W Au Pb Pb Pb Bi Figure 5. One centimetre deep dose-equivalent rate in the head of linear accelerator. Target to top of scintillator detector was 20 and 50 cm. Figure 6. Dose equivalent values calculated from spectral data, and dose from the scintillator survey meter measurement. The values at 25 h from irradiation were normalised to 1. secondary radioactivation from neutrons produced by the photonuclear reaction. A photonuclear reaction has a threshold energy (Table 2) (10), and the amount of neutrons produced by the 10-MeV X-rays with increasing Z on the material. When a treatment beam and multi-leaf collimators collide directly, as during intensity-modulated radiation therapy, the effect of neutrons on the patient must also be taken into consideration. It is assumed that all important isotopes have been identified with the possible exception of very short-lived or pure beta-emitting nuclides. The use of a portable HP-Ge semiconductor survey meter enables high-energy resolution spectral measurements in the treatment room without dismantling the equipment. This type of survey meter can effectively analyse radioactivated materials that cannot be removed from the treatment room because of the high-radiation levels or cannot be detached from the equipment. By analysing the amount of induced radionuclides, useful information for determining the radioactive cooling period for the decommissioning equipment can be acquired. The amount of radioactivation is relatively low with 10-MeV X-rays. However, when materials are activated with long half-life radionuclides, proper management of the operating conditions is required to control the amount of induced radioactivity. The proposed method is not a quantitative evaluation, and additional measurements are needed to estimate the amount of radioactivity. The nuclides detected in the EXL-15SP and Clinac ix accelerators were different. The reason why long-term activations were detected in the EXL- 15SP could be because the operating period was long. Long-term activations are few, the radioactivity level is low, and a cooling time of several days is effective for reducing dose rate to an acceptable level for decommissioning. As seen in Figure 6, the dose decay results determined through calculations involving the spectral data and through measurements with the survey meter were highly comparable. The key nuclides of exposure were confirmed. The most commonly used energy of linear accelerators for radiotherapy is 10 MeV in Japan (11).No induced radioactivity was detected on the walls in this study. Thus, for a 10-MeV accelerator, care should be taken only for the target and its assembly. By the end of a 1-week cooling period, the radiation dose rate around the accelerator was greatly reduced. 172

6 INDUCED RADIOACTIVE NUCLIDES OF 10-MEV LINAC Therefore, 1 week is regarded as an adequate cooling time for decommissioning (12, 13). For an accelerator employed.10 MeV, the residual radioactivity around the accelerator would be near the clearance level (14). These results are important for establishing appropriate regulations based on scientific data. CONCLUSION The energy spectra of gamma rays emitted from the heads of two linear accelerator models were measured using a portable HP-Ge semiconductor detector. 24 Na, 28 Al, 57 Ni, 58 Co, 54 Mn, 56 Mn, 60 Co, 64 Cu, 65 Zn, 124 Sb, 181 W, 187 W, 196 Au and 198 Au were detected. The dose rate decreased to 10 % in 1 week. The estimated amount of 60 Co in the head was,0.1 Bq. Long-term activations were few, the radioactivity level was low and a cooling time of several days was effective at the time of decommissioning. ACKNOWLEDGEMENTS The authors wish to thank the reviewer who spent a great deal of time and gave informative advice for improving the manuscript. The authors thank Seiko EG&G Co., Ltd, Tokyo Japan, for providing HP-Ge detector. FUNDING This survey was conducted as a part of the research by the JSRT research group on the control and disposal of radioactive waste and was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Health, Labour and Welfare for research on medical safety and health technology assessment entitled Research on securing safety of medical radiation (H19-medical treatment-common-003) (chief researcher: Makoto Hosono). REFERENCES 1. Ahlgren, L. and Olsson, L. E. Induced activity in a high-energy linear accelerator. Phys. Med. Biol. 33, (1988). 2. Brusa, A., Cesana, A., Stucchi, C., Terrani, M. and Zenellati, F. Long term activation in a 15 MeV radiotherapy accelerator. Med. Phys. 35, (2008). 3. Konefal, A., Polaczek-Grelik, K. and Zipper, W. Undesirable nuclear reactions and induced radioactivity as a result of the use of the high-energy therapeutic beams generated by medical linacs. Radiat. Prot. Dosim. 128, (2008). 4. Konefal, A., Orlef, A., Dybek, M., Maniakowski, Z., Polaczek-Grelik, K. and Zipper, W. Correlation between radioactivity induced inside the treatment room and the undesirable thermal/resonance neutron radiation produced by linac. Phys. Med. 24, (2008). 5. Fischer, W. H., Tabot, B. and Poppe, B. Comparison of activation products and induced dose rate in different high-energy medical linear accelerators. Health Phys. 94, (2008). 6. Yamaguchi, I., Terada, H., Takahashi, M. and Sugiyama, H. Quantitative activation analysis of medical linear accelerator target assemblies. J. Radioanal. Nucl. Chem. 278, (2008). 7. Hirayama, H., Namito, Y., Bielajew, A. F., Wilderman, S. J. and Nelson, W. R. The EGS5 code system, SLAC- R-730 (2005) and KEK Report (2005). 8. JIS (Japanese Industrial Standard). Methods of calibration for exposure meters, air kerma meters, air absorbed dose meters and dose-equivalent meters. JIS Z4511 (2005). 9. ICRP (International Commission on Radiological Protection). Conversion coefficients for use in radiological protection against external radiation. ICRP Publication 74. Pergamon Press (1996). 10. IAEA, and Handbook on Photonuclear Data for Applications: Cross-Sections and Spectra. IAEATECDOC IAEA (2000). 11. Yamaguchi, I., Tanaka, S., Fujibuchi, T., Kida, T., Nagaoka, H. and Watanabe, H. Nationwide survey on the operational status of electron accelerators for radiation therapy in Japan. Radiol. Phys. Tech. 3, (2010). 12. Almen, A., Ahlgren, L. and Mattsson, S. Absorbed dose to technicians due to induced activity in linear accelerators of radiation therapy, Phys. Med. Biol. 36, (1991). 13. Wan, Y., Evans, M. G. and Podgorsak, E. Characteristics of induced activity from medical linear accelerators. Med. Phys. 32, (2005). 14. International Atomic Energy Agency. Application of the Concepts of Exclusion, Exemption and Clearance Safety Guide No.RS-G-1.7. IAEA (2004). 173