This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Radiation Physics and Chemistry 80 (2011) 917–922 Contents lists available at ScienceDirect Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem Thermal neutron fluence in a treatment room with a Varian linear accelerator at a medical university hospital Wen-Shan Liu a, Sheng-Pin Changlai b, Lung-Kwang Pan c, Hsien-Chun Tseng a, Chien-Yi Chen d,e,n a School of Medicine, Chung Shan Medical University and Department of Radiation Oncology, Chung Shan Medical University Hospital, Taiwan, Republic of China Department of Nuclear Medicine, Lin Shin Hospital, Taichung 402, Taiwan, Republic of China c Department of Medical Imaging and Radiological Sciences, Central Taiwan University of Science and Technology, Takun 40609, Taiwan, Republic of China d Department of Medical Image, Chung Shan Medical University Hospital, Taiwan, Republic of China e School of Medical Imaging and Radiological Sciences, Chung Shan Medical University, Taichung 402, Taiwan, Republic of China b a r t i c l e i n f o abstract Article history: Received 14 February 2008 Accepted 31 March 2011 Available online 8 April 2011 The indium foil activation technique has been employed to measure thermal neutron fluences (Fth) among various locations in the treatment room with a 20  20 cm2 field size and a 15 and 10 MV X-ray beam. Spatial Fth are visualized using colored three-dimensional graphical representations; intensities are up to (1.97 7 0.13)  105 and (1.46 7 0.13)  104 n cm  2/Gy-X at isocenter, respectively. The Fth is found to increase with the X-ray energy of the LINAC and decreases as it moves away from the beam center. However, thermal neutron exposure is not assessed in routine dosimetry planning and radiation assessment of patients since neutron dose contributes o 1% of the given therapy dose. However, unlike the accelerated beam limited within the gantry window, photoneutrons are widely spread in the treatment room. Distributions of Fth were measured in water phantom irradiated with 15 MV X-ray beams. The shielding effect of the maze was also evaluated. The experimentally estimated Fth along the maze distance was fitted explicate and the tenth-value layer (TVL) was calculated and discussed. Use of a 10 cm-thick polyethylene door placed at the maze was suitable for radiation shielding. & 2011 Elsevier Ltd. All rights reserved. Keywords: Indium foils activation technique Thermal neutron fluences Water phantom Shielding effect 1. Introduction High-energy electron accelerators, such as the electron linear accelerator (LINAC) are utilized routinely to generate photon beams for cancer therapy (Kamino, 1998). However, when high-energy accelerators are operated roughly at 10 MV or higher, a significant number of neutrons are generated through photonuclear reactions via (e, n) and (x, n), such that high-energy bremsstrahlung react with the target, dynamic wedge and multi-leaves resulting in a mixed radiation field in the radiotherapy area (Konefal et al., 2005; Lennox, 2001; Lin et al., 2001; Kase et al., 1998; AAPM, 1986). With growing interest in the use of high-energy photon beams at Chung Shan Medical University Hospital (CSMUH), detailed measurements of thermal neutron fluences (Fth) are required. Fth have been extensively investigated using calculations and measurement techniques for LINACs with various energies. Table 1 lists the neutron fluences in the 10–25 MV range (Barquero et al., 2005; Konefal et al., 2005; Lin et al., 2001; Paredes et al., 1999; Palta et al., 1984; Gur et al., 1978; Deye and Young, 1977; McGinley et al., 1976; Uwamino et al., 1986). The foil activation technique has been employed for n Corresponding author at: School of Medical Imaging and Radiological Sciences, Chung Shan Medical University, Taiwan, Republic of China. Tel.: þ886 4 24730022 x 17220; fax: þ886 4 247217210. E-mail address: ccy@csmu.edu.tw (C.-Y. Chen). 0969-806X/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.radphyschem.2011.03.022 three-dimensional distribution measurements of Fth. Neutrons are, however, scattered in the human body (Comsan, 1996). The neutron field in the body differs from that in the primary field. Distributions of Fth depth were measured in a water phantom irradiated with a 15 MV X-ray beam. The problem of adequate shielding from maze and neutron scattering into the maze may require additional attention. The tenth-value layer (TVL) of Fth was expressed as an exponential function to the calculated points in the maze. 2. Materials and methods 2.1. Varian CLINAC 21EX A large number of LINACs around the world are currently in use for radiation therapy (Baker and Thomas, 2001; Lennox, 2001; Kamino, 1998). However, only limited studies are focused on the distributions of Fth inside treatment rooms, water phantoms and mazes of medical LINACs. A detailed survey of Fth in treatment rooms, water phantoms and mazes is important for radiotherapy patients. The LINAC (Varian CLINAC 21EX; Palo Alto, CA, USA) at CSMUH is designed for Intensity Modulated Radiation Therapy (IMRT), which is achieved by superposing different MLC segments. Approximately 8500–9000 patients with various pathologies, most of them with nasopharyngeal cancer in the head and neck have been treated at CSMUH since 1999. LINAC generates dual photon energies Author's personal copy 918 W.-S. Liu et al. / Radiation Physics and Chemistry 80 (2011) 917–922 Table 1 Neutron fluence measurements of this study and other investigations. Accelerator Energy (MV) SSD (cm) X-ray filed size (cm2) Method Neutron fluence n cm  2/Gy-X Ref. Varian CLINAC 21EX Varian CLINAC 21EX Varian 2300 Siemens Varian CLINAC 2300 Varian CLINAC 2300 Primus Siemens Siemens Siemens Primus Varian CLINAC 2100 C Mevatron 77 Philips SL/75-20 Varian CLINAC Varian CLINAC-18 Allis-Chalmers Betatron Siemens Microtron, MM22 15 10 22 21 20 18 18 15 15 18 18 18 18 10 25 18 21 100 100 20  20 20  20 In foil In foil 100 10  10 In foil 40  40 P2O5 CR 39, MCNP Au foil In foil 5.1 cm Rh In foil In foil, BF3 BF3 In foil 1.97  105 1.46  104 1.10  104 2.00  104 1.51  106 7.00  103 1.20  104 2.90  105 1.51  106 1.07  105 2.30  105 1.4  106 1.4  104 6.80  104 3.80  105 4.60  105 3.2  105 This study This study Konefal et al., 2005 Konefal et al., 2005 Konefal et al., 2005 Konefal et al., 2005 Konefal et al., 2005 Konefal et al., 2005 Lin et al., 2001 Paredes et al., 1999 Palta et al., 1984 Gur et al., 1978 Deye and Young, 1977 McGinley et al., 1976 McGinley et al., 1976 Barquero et al., 2005 Uwamino et al., 1986 100 100 100 100 100 15  15 10  10 25  25 10  10 10  10 100 10  10 Table 2 Systematic and random errors for the practical evaluation of photo neutron fluence in this study. concrete 2.2 m concrete 2.7 m concrete y 2.2 m x Source One standard deviation Di Others: Daily X-ray constancy Daily electron output constancy LINAC power fluctuation o2% o2% o1% Measurement: Internal normalization of foil Activated foil counting statistics 1.1–1.8% 6.12%  9.4% Dtot 6.90–10.2% Isocenter (0,0,0) slide door A 250 mm Pb Maze B 1.0 m S Control Area Fig. 1. Plane view and measurement locations in the Varian CLINAC 21EX treatment room at CSMUH. The 10 cm-thick PE sliding door was installed to decrease transmitted neutron fluence. (C-69; Wellhofer Co., Germany) calibrated by the National Ionization Standard Laboratory, Institute of Nuclear Energy Research. The errors associated with the derived neutron fluence were calculated as the square root of the sum of the squares of the individual errors Di, as listed in Table 2. The experimental uncertainty includes (1) the dosimetry and power fluctuation of LINAC, which were less than 2% and (2) counting statistics during In foil activation analysis. 2.2. Foil activation technique The foil activation technique is widely utilized for measuring with accelerating voltages up to 6 MV or higher. The electrons are accelerated and hit the gantry head target (Lennox, 2001; Kamino, 1998). Fig. 1 shows a plane view of the radiotherapy area with the Varian CLINAC 21EX at CSMUH. The accelerator isocenter (0, 0 and 0) is located midway between the primary wall barriers at a height of 1.3 m above the floor (Lennox, 2001; Kamino, 1998). The treatment room has a metal slab and 2.2 m thick concrete on the left side of the primary barriers. Right primary wall barriers are fabricated entirely with 2.7 m thick concrete. A 250 mm thick lead slab was employed in the construction of the left side of the primary barriers. The isocenter is 3.55 m from the primary barrier. The treatment room is 7.9 m long by 7.1 m wide by 3.8 m high. A slide door of 10 cm polyethylene (PE) is used to moderate the fast and intermediate energy neutrons (Baker and Thomas, 2001). Lead of 1 cm thick is installed at the maze end. Due to its weight, the sliding door has a motorized opener. The maze below the treatment room is located between the control area of right and approximately rectangular concrete of 6.8 m long  2.7 m wide  3.8 m high (Fig. 1). The X-ray dose rate was confirmed using an ionization chamber Fth (n cm  2/Gy-X). Thermal neutrons can be identified by selecting appropriate foil materials (Table 1; Barquero et al., 2005; Konefal et al., 2005; Chao et al., 2001; Lin et al., 2001; Paredes et al., 1999; Palta et al., 1984; Gur et al., 1978; Deye and Young, 1977; McGinley et al., 1976; Uwamino et al., 1986). The Fth in the treatment room was measured using the following typical conditions with a source-surface distance (SSD) of 100 cm and 10 photon Gy for each experimental run of 2.5 min. All experiments were conducted in triplicate. (1) Experiments were repeated with collimator opened to 20  20 cm2 and the gantry was vertically oriented pointing down at the floor. (2) The accelerators were operated at 15 MV in water phantom. The Fth in the treatment room was measured by activation of indium foils via 115In(n, g)116m1In reaction, which has high activation section (162 barns) for the thermal neutrons and a suitably short half-life (T1/2 ¼54.2 min) (Konefal et al., 2005; Chao et al., 2001; AAPM, 1986; Gur et al., 1978). An indium foil (purity 499.999%; 25 mm long  25 mm wide  1 mm high, Goodfellow Cambridge, Ltd., UK) with an average mass of Author's personal copy W.-S. Liu et al. / Radiation Physics and Chemistry 80 (2011) 917–922 919 15 MV. The water phantom base was 58 cm long  41 cm wide  35.5 cm high; the polymethyl-methacrylate (PMMA) tank had a volume of approximately 7100 cm3. The PMMA tank was filled with water to a height of 30 cm (Fig. 2(b)). The water phantom center was placed directly on the X-ray beam and located at the isocenter (i.e., 15 cm below the water surface). These indium foils were inserted into Plexiglas holders. Each indium foil was placed individually exactly 1 cm apart on the x, y and z axes. In total, 69 detection points were performed subsequently at incremental steps to map the Fth in the water phantom. These phantom measurements can be performed at positions of the central axis ranging in depth from 1 to 30 cm and at two 30 cm off-axis position at 15 cm depth. z isocenter x y 2.4. The maze design of the 15 MV treatment room at CSMUH Fig. 2. Measurements of Fth performed in the water phantom during exposure to 15 MV X-rays. Indium foil locations are indicated by circles (K); the x and y axes correspond to locations in the treatment room for SSD ¼ 100 cm for 15 and 10 MV X-ray exposures, respectively. 4.86 70.07 g was placed at the isocenter as a reference foil, other identical indium foils were suspended anywhere inside the room during experiments (Fig. 2(a); AAPM, 1986; Palta et al., 1984). Consider a thin indium foil of mass m exposed for a period of time ti. After the irradiation if the foil is allowed to cool for time td and then counted for time tc using a high-purity germanium (HPGe) detector (GC3520; Canberra Industries, Meriden, CT, USA), which has an active volume of 145.3 cm3, 35% relative efficiency and resolution of 2.0 keV at 1332 keV, Fth can be derived as Fth ¼ NIn egsK ð1Þ where K ¼ nð1elti Þðeltd Þð1eltc Þ=l, NIn is the number of activated nuclides of 116m1In, l is the decay constant (¼0.693/T1/2), e is the detection efficiency; s is the microscopic cross-section, T1/2 is the half-life of 116m1In, and g is the absolute g-ray intensity (29.2%). n¼ y  m  NA M ð2Þ where n is the number of 115In, y is isotopic abundance of 115In, the indium nuclide (95.7%), NA is Avogadro’s number (6.022  1023 in atoms/g-atom), M is molar mass of the foil material and m is indium foil mass. The measured g-ray spectra were collected using a multichannel analyzer (10 þ ; Canberra Industries, Meriden, CT, USA). These foils were immediately placed on the detector face. The e at the characteristic g-ray energy of 417 keV was emitted from the isomeric transition of 116m1In and calibrated as 4.0%. The efficiency of the GC3520 HPGe detector measured a series of g-energy using calibrated standards for 60Co, 133Ba, 137Cs and 152Eu. The e in counts/ g was extrapolated for a similar geometry. For each irradiation position, the induced radioactivities of 116m1In from 15 and 10 MV LINACs for count times of 3 and 6 min, respectively, were precisely determined. Following exposure, the activated indium foils were allowed to decay for appropriate times. The principal errors are systemic and consist of detector energy responses. Spectra acquired were identified and calculated using Micro SAMPO-90 software and a personal computer. Details of the experimental conditions are given elsewhere (Chen et al., 2007; Uwamino et al., 1986; Gur et al., 1978). Statistical errors were within 10%. Structure shielding design of maze and entrance slide door was built to fulfill the recommendation of NCRP 49 (NCRP, 1976). A maze and shielded entrance slide door is utilized to protect the personnel outside the facility (Baker and Thomas, 2001). Point A (  3.8,  5.9, 0) is on the maze centerline and visible from the isocenter (Fig. 1). All measurements were performed by placing indium foils at locations A and B, (7.8, 5.9 and 0; near the 10 cm PE sliding door) at 0.5–1 m intervals along a straight line. 3. Results and discussion 3.1. The Fth distributions at 15 and 10 MV Fig. 3(a) and (b) shows the foil technique results for Fth at 15 and 10 MV in the 21EX Varian clinic at CSMUH; Fth is a function of distance from the central axis. The difference in Fth between 15 and 10 MV beams is attributed to the difference in X-ray energy of the LINACs. Three-dimensional distributions of the Fth for the treatment room can be mapped using colored profiles that reflect various intensities of the thermal neutron field. The blue color in Fig. 3(a) reflects the great Fth value 41.78  105 n cm  2/Gy-X and 3.6070.16  105 n cm  2/Gy-X was measured at the gantry head, roughly 1.83 times greater than that at the isocenter on the z-axis. This finding is expected as the contribution of scattered components to the total Fth that decreases as the distance to the isocenter increases as shown in (Fig. 3(a) and (b)). Both minor and significant variations of Fth were observed across the treatment room. The representative Fth values on the y-axis from the isocenter edge at 60, 100 and 160 cm are (1.3270.12)  105, (1.2670.11)  105 and (1.2870.12)  105 n cm  2/Gy-X, respectively (Fig. 3(a)). The measured Fth, ranging from 1.78  105 to 1.57  105 (Fig. 3(a)) is shown in red color; this finding is due to the distributions of leaking neutrons through the accelerator head. The measured Fth around the maze entrance at the far end of the isocenter (green color in Fig. 3(a)) is (1.0170.09)  105 n cm  2/Gy-X; the lowest Fth in the treatment room is o(1.0170.09)  105 n cm  2/Gy-X (yellow color in Fig. 3(a)). The Fth at 15 MV was (1.9770.13)  105 n cm  2/Gy-X, approximately 13.5 times greater than that of the 10 MV beam ((1.4670.13)  104 n cm  2/Gy-X) at the isocenter. The Fth in the treatment room irradiated by 10 MV X-ray vary, (1.4670.13)  104  (6.3570.03)  103 n cm  2/Gy-X and are nearly uniform throughout the treatment room (Fig. 3(b)). 3.2. Distributions of Fth in a water phantom 2.3. Field application for water phantom The Fth measurements performed using the foil technique simulated total body irradiation is carried out using a water phantom from Varian CLINAC 21EX accelerators operating at In this study, dose depth equivalent measurements were preformed in a water phantom irradiated with an X-ray beam field of 20  20 cm2 from Varian 21EX accelerators operating at 15 MV. The photoneutrons entered the water moderator and slowed via Author's personal copy 920 2×105 water level 2.00×105 1.78×105 1.57×105 1.35×105 1.13×105 9.10×104 6.90×104 1.0×105 1×105 0.5×104 -2 X ax is 2 )m 0 0 ,y, 0 ( xis 0 (x ,0, 0) 2 -2 m Ya Thermal neutron fluence (n⋅cm-2/Gy-X) Thermal neutron fluence (n⋅cm-2/Gy-X) W.-S. Liu et al. / Radiation Physics and Chemistry 80 (2011) 917–922 isocenter Gantry head z y water level isocenter 105 tank bottom 1.50×104 1.32×104 1.13×104 9.40×103 7.51×103 5.65×103 0.4 0.5×103 0 X -2 ax is ( 0 ,0) 2 )m 0 , 0,y 0 x,0 m -2 2 xis ( Ya Fig. 3. Distributions of Fth in treatment room (Varian CLINAC 21EX) irradiated by a 20  20 cm2 beams of (a) 15 and (b) 10 MV X-rays. Experimental data are displayed using a three-dimensional colored graphical representation (thermal neutron fluence unit is n cm  2/Gy-X). (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.). a 103 0.2 0.1 Z axis (0, 0, z) m 0.0 -0.1 -0.2 isocenter 105 CSMUH isocenter 1m x 104 -0.3 104 0.3 106 Thermal neutron fluence (n⋅cm-2/Gy-X) utron fluence Thermal ne-2 (n⋅cm /Gy-X) 0.5 1.5×104 1.0×104 tank bottom 106 -0.2 -0.1 0.0 0.1 X axis (x,0,0) 0.2 0.3 Fig. 5. Distributions of Fth in a water phantom irradiated by a 20  20 cm2 beam of a 15 MV X-ray. (a) versus depth along the central z-axis; and (b) across the width at a depth of 15 cm on the x-axis. 1293 keV 116m1In 101 2212 keV 23Th 460 keV 40K 104 417 keV 116m1In 100 336 keV 115mIn Counts/Channel 102 102 103 101 100 b 1000 2000 3000 4000 The major and relatively significant variations of Fth, varying from (8.8070.09)  105  (6.8570.07)  104 n cm  2/Gy-X, were observed on the z-axis in the water phantom (Fig. 5(a)). The Fth was a function of distance from the z-axis and penetration depth in the water phantom. The depth of maximum Fth was 1.3 cm on the z-axis that is, (0, 0, and 0.13). Moreover, the Fth was relatively constant inside and outside the 20  20 cm2 field size along the x-axis and roughly uniform throughout the water phantom (Fig. 5(b)). At a depth of 25 cm (0, 0, and 0.10) from water surface, Fth was only 31.1% of the isocenter of the water phantom, the smallest among all water phantoms (Lennox, 2001). 3.3. Thermal neutron dose to patients Channel Fig. 4. Gamma-ray spectra collected by the germanium detection system for radioactivity measurement of an indium foil of 20  20 cm2 exposed to 15 MV X-rays. (a) isocenter and (b) (0.12, 0, 0) outside the beam field. elastic collisions and captured to form 116m1In. The spectra in Fig. 4(a) and (b) were measured at the isocenter and outside the beam field (0.12, 0 and 0) of water phantom, respectively. The photon peak at 336 keV is clearly seen in the spectra (Fig. 4(a)), which is from 115mIn and is associated primarily with high-energy photon interaction through 115In(g, g)115mIn (Chao et al., 2001). Thermal neutron exposure is not considered in routine dosimetry planning; however, patients may require ensuring their safety. McCall (McCall, 1997) stated that thermal neutron dose conversion factor is 260 n/cm2 s¼ 1 mrem/h¼2.78 nSv/s. Therefore, based on experimental data obtained from the isocenter of the water phantom, the thermal neutron dose equivalent is 9.42 mSv/min for a patient receiving a 10 Gy photon dose during a routine 2.5 min treatment. For a 15 MV 20  20 cm2 field size, the thermal neutron dose equivalent at the isocenter of the water phantom is 2.36 mSv/Gy-X, representing a total dose of Author's personal copy 921 W.-S. Liu et al. / Radiation Physics and Chemistry 80 (2011) 917–922 3.4. Fth in the maze The Fth were measured along the line A–B (cf Fig. 1). The Fth at point A was equal to (5.0170.03)  104 ncm  2/Gy-X, 25.4% of that at the isocenter, which was higher than that at other places in the maze due to neutrons leaking through the accelerator head (Comsan, 1996; Kamino, 1998). The Fth of point B, (2.8570.21)  103 n cm  2/Gy-X was approximately 1.45% of that at the isocenter and the smallest in the treatment room. This finding states that photoneutrons are emitted strongly forward after being produced by high-energy X-rays from beam-modified devices. Fig. 6 shows a semi-log plot of the Fth as function of distance (x) along the A–B centerline. The Fth, x (Fig. 6) can be expressed as an exponential function of x Fth,x ¼ Fth  10TVL x This work (Varian Clinac 21Ex) Barquero et al., 2005 (Siemens) Konelfal et al., 2005 (Varian, Siemens) Lin et al., 2001 (Primus Siemens) Uwamino et al., 1986 (Microtron MM22) Plata et al., 1985(Mevatron 77) Gur et al., 1978 (Philips, SL/75-20) Deye and Young 1977 (Varian Clinac) McGinley, 1976 (Varian Clinac-18) 3×106 2.5×106 2×106 1.5×106 1×106 5×105 0 5 10 15 20 25 30 Accelerator energy (MV) Fig. 7. The neutron fluence at the isocenter of medical accelerator operated at different peak X-ray energies. ð3Þ where Fth, x is the Fth of the x position, TVL is tenth-value layer (m) of Fth in the maze. The TVL can be obtained from Fig. 6 by extrapolating the portion of curve with a y value to a point on the A–B straight line. The TVL was estimated at 7.2770.10 m in this study. This experimental result is 4 6.4 m, as obtained by McGinly and Miner, who proposed that a TVL 46.4 m can be classified as hard or the high-energy component of neutrons (McGinley and Miner, 1995). At point ‘‘S’’ outside the sliding door, the Fth were negligible or below the detection limit (DL) of the indium foil activation approach, indicating that 10 cm PE effectively attenuates the penetrating neutrons and an adequate shielding design of the wall and maze (cf. Fig. 1). 3.5. Detection limit (DL) of foil activation technique The DL of Fth for 116m1In in this study was 4.65  DS  (B/t)1/2, where B is the background count rate at the 417 keV of 116m1In, and t is the counting period (Chen, 2003; Currie, 1968). Hence, the detection sensitivity (DS) for 116m1In can be calculated by dividing the counts of 417 keV by the e and absolute g-ray intensity. In the 6 min counting period of the 10 MV LINAC, the DL of Fth was Thermal neutron fluence (n⋅cm-2/G-X) 3.5×106 Neutron fluence (n⋅cm-2/Gy-X) 2.36  10  4%, which is in agreement with the experimental results reported elsewhere (Lin et al., 2001; Paredes et al., 1999). Additionally, the thermal neutron equivalent near the treatment bed is roughly 10  5 of the dose delivered to patients. estimated at 400 n cm  2/Gy-X (Currie, 1968). Using this technique to determine Fth in treatment rooms for LINAC at hospitals is feasible. 3.6. Fth compared with results obtained by other studies Fig. 7 shows the neutron fluence measured at the isocenter within a 20  20 cm2 X-ray field and expressed in n cm  2/Gy-X. The Fth obtained are compared with those obtained by recent studies. The only exceptions relative to Fth are experimental results acquired by Lin et al. (Lin et al., 2001), who used a 40  40 cm2 field size. Generally, these studies suggested that the range of Fth was 1.20  104  1.60  106 n cm  2/Gy-X at 100 cm SSD for LINACs operating at 10–25 MV. Experimental data in this study are in agreement with the average of results obtained by other studies (Barquero et al., 2005; Konefal et al., 2005; Lin et al., 2001; Palta et al., 1984; Gur et al., 1978; Deye and Young, 1977; McGinley et al., 1976). This agreement is demonstrated by the linear scale for Fth (Fig. 7). Conversely, the Fth for 15-MX Varian CLINAC 21EX is roughly 26% lower than that of Primus Siemens (Lin et al., 2001). This difference is due to the different materials used to construct the accelerator head of the target collimators and scattering shields. 105 4. Conclusion A B 104 HPGe GC3520 10 10+MCA PC / SAMPO-90 3 0 2 4 6 8 10 Maze (m) Fig. 6. Exponential fitting curves of Fth in the maze. Locations of the Fth estimation (A–B) points are shown with the corresponding distances from the maze entrance. The Fth produced from a Varian CLINAC 21EX operated at 15 and 10 MV were successfully measured and the accompanied neutron dose for personnel was also elucidated herein. The foil activation technique is an effective method of measuring Fth in treatment rooms. The Fth for the 15 MV, (1.9770.13)  105 n cm  2/Gy-X was roughly 13.5 times that of the 10 MV beam, (1.4670.13)  104 n cm  2/Gy-X at the isocenter. The Fth measurements for Varian CLINAC 21EX accelerators indicate that Fth is a function of distance from the central axes of x and y and penetrating depth in the water phantom, representing only 2.36  10  4% of the total dose. The Fth of the Varian 21EX are in agreement with those reported in literature (McGinley and Miner, 1995). Thus, a 10 cm thick PE is capable of suppressing the neutron dose for personnel and is also a reliable way to reduce the Fth at the outer maze entrance. A maze with 10 cm thick PE effectively attenuates the penetrating neutrons and provides adequate shielding for the wall and maze. Author's personal copy 922 W.-S. Liu et al. / Radiation Physics and Chemistry 80 (2011) 917–922 Acknowledgements The authors wish to thank radiotherapists of CSMUH for their assistances in the operation of the accelerator. This work was financially supported by the National Science Council of Republic of China under Contract no. NSC 97-2221-E-040-008. References American Association of Physicists in Medicine, 1986. Neutron Measurements Around High Energy X-ray Radiotherapy Machine. AAPM No. 19, New York. Baker, C.R., Thomas, S.J., 2001. Neutron transport in a clinical linear accelerator bunker: comparison of materials for reducing the photo-neutron dose at the maze entrance. Radiat. Phys. Chem. 61, 633–634. Barquero, R., Mendez, R., Vega-Carrillo, H.R., Iniguez, M.P., Edwards, T.M., 2005. Neutron spectra and dosimetric features around an 18 MV LINAC accelerator. Health Phys. 88 (1), 48–58. Chao, J.H., Hsu, P.C., Liu, H.M., 2001. Measurement of high dose rates by photon activation of indium foils. Appl. Radiat. Isot. 55, 549–556. Chen, C.Y., 2003. Optimal conditions for identifying 80Br and 128I in health food Angelica keiskei using rapid epithermal neutron activation analysis. Appl. Radiat. Isot. 58 (4), 423–429. Chen, C.Y., Lin, D.B., Chen, W.K., 2007. Elements in the sera of preschool children living in central Taiwan. Chin. J. Chem. 25, 515–520. Comsan, M.N.H., 1996. Shielding calculations for Inshas cyclotron. Radiat. Phys. Chem. 47 (5), 735–738. Currie, L.A., 1968. Limits for qualitative detection and quantitative determination. Anal. Chem. 40, 586–593. Deye, J.A., Young, F.C., 1977. Neutron production from 10 MV medical LINAC. Phys. Med. Biol. 22 (1), 90–94. Gur, D., Rosen, J.C., Bukovitz, A.G., Gill, A.W., 1978. Fast and slow neutrons in an 18 MV photo beam from a Philips SL/75-20 linear accelerator. Med. Phys. 5 (3), 221–222. Kamino, Y., 1998. 10 MeV 25 kW industrial electron LINAC. Radiat. Phys. Chem. 52 (1–6), 573–576. Kase, K.R., Mao, X.S., Nelson, W.R., Liu, J.C., Kleck, J.H., Elsalim, M., 1998. Neutron fluence and energy spectra around the varian CLINAC 2100C/2300C medical accelerator. Health Phys. 74, 38–47. Konefal, A., Dybek, M., Zipper, W., Łobodziec, W., Szczucka, K., 2005. Thermal and epithermal neutrons in the vicinity of the Primus Siemens biomedical accelerator. Nukleonika 50 (2), 73–81. Lennox, A.J., 2001. Accelerators for cancer therapy. Radiat. Phys. Chem. 61, 223–226. Lin, J.P., Chua, T.C., Lin, S.Y., Liu, M.T., 2001. The measurement of photoneutrons in the vicinity of a Siemens Primus linear accelerator. Appl. Radiat. Isot. 55, 315–321. McCall, R.C., 1997. Shielding for thermal neutrons. Med. Phys. 24 (1), 135–136. McGinley, P.H., Miner, M.A., 1995. Method of eliminating the maze door of medical accelerator rooms. Radiat. Prot. Manage. 12 (5), 29–37. McGinley, P.H., Wood, M., Millis, M., Rodriguez, R., 1976. Dose levels due to neutrons in the vicinity of high-energy medical accelerators. Med. Phys. 3 (6), 397–402. National Council on Radiation Protection and Measurements, 1976. National Council on Radiation Protection and Measurements Report 49. Structural Shielding Design and Evaluation for Medical Use of X-rays and Gamma-rays of Energies up to 10 MeV, Washington, DC. Palta, J.R., Hogstrom, K.R., Tannanonta, C., 1984. Neutron leakage measurements from a medical accelerator. Med. Phys. 11 (4), 498–501. Paredes, L., Genis, R., Balcazar, M., Tavera, L., Camacho, E., 1999. Fast neutron leakage in 18 MeV medical electron accelerator. Radiat. Meas. 31, 475–478. Uwamino, Y., Nakamura, T., Ohkubo, T., Hara, A., 1986. Measurement and calculation of neutron leakage from a medical electron accelerator. Med. Phys. 13 (3), 374–384.