- Email: [email protected]
hole traps under pre-irradiation and bleaching conditions
Physica Medica 38 (2017) 81–87
Contents lists available at ScienceDirect
Physica Medica journal homepage: http://www.physicamedica.com
Original paper
Sensitivity and stability of optically stimulated luminescence dosimeters with filled deep electron/hole traps under pre-irradiation and bleaching conditions So-Yeon Park a,b,c, Chang Heon Choi a,b,c, Jong Min Park a,b,c,d, Minsoo Chun a,b,c, Ji Hye Han a,c, Jung-in Kim a,b,c,⇑ a
Department of Radiation Oncology, Seoul National University Hospital, Seoul, Republic of Korea Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Republic of Korea c Biomedical Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea d Center for Convergence Research on Robotics, Advance Institutes of Convergence Technology, Suwon, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 1 September 2016 Received in Revised form 14 March 2017 Accepted 10 May 2017 Available online 17 May 2017 Keywords: Dosimetry Optically stimulated luminescence Optimal bleaching condition nanoDot
a b s t r a c t Purpose: We aimed to evaluate the characteristics of optically stimulated luminescence dosimeters (OSLDs) with fully filled deep electron/hole traps, and determine the optimal bleaching conditions for these OSLDs to minimize the changes in dose sensitivity or linearity according to the accumulated dose. Methods: InLight nanoDots were used as OSLDs. The OSLDs were first pre-irradiated at a dose greater than 5 kGy to fill the deep electron and hole traps, and then bleached (OSLDfull). OSLDfull characteristics were investigated in terms of the full bleaching, fading, regeneration of luminescence, dose linearity, and dose sensitivity with various bleaching conditions. For comparison, OSLDs with un-filled deep electron/hole traps (OSLDempty) were investigated in the same manner. Results: The fading for OSLDfull exhibited stable signals after 10 min, for 1 and 10 Gy. The mean supralinear index values for OSLDfull were 1.001 ± 0.001 for doses from 2 to 10 Gy. Small variations in dose sensitivity were obtained for OSLDfull within standard deviations of 0.85% and 0.71%, whereas those of OSLDempty decreased by 2.3% and 4.2% per 10 Gy for unfiltered and filtered bleaching devices, respectively. Conclusions: Under the bleaching conditions determined in this study, clinical dosimetry with OSLDfull is highly stable, minimizing the changes in dose sensitivity or linearity for the clinical dose. Ó 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
1. Introduction The use of anion-deficient sapphire doped with carbon (Al2O3: C) to obtain optically stimulated luminescence (OSL) is known to have advantages for dosimetry, such as good reproducibility, low energy dependency, and low angular dependency compared to LiF thermoluminescence (TL) dosimetry for energies above MeV [1–5]. Therefore, OSL dosimeters (OSLDs) are widely used for not only in vivo dosimetry for cancer patients in the field of radiation therapy, but also personal radiation monitoring in the field of radiation protection [6–8]. Beyond in vivo dosimetry and monitoring, OSLDs are also an important tool for auditing and research in the field of radiation therapy [9–11]. ⇑ Corresponding author at: Department of Radiation Oncology, Seoul National University Hospital, Seoul 110-744, Republic of Korea. E-mail address: [email protected] (J.-i. Kim).
Several studies have explained the OSL mechanism using the energy-band model [3,12–14]. When OSLDs are irradiated, free electrons and holes are first generated, and they subsequently separate. The free electrons can potentially be captured by large numbers of shallow, dosimetric, or deep electron traps, depending on the energy band levels [3]. Further, the holes can be captured in deep hole traps and at luminescence centers. Electrons captured by shallow electron traps are released at room temperature within a few minutes, whereas electrons captured by dosimetric electron traps can be held at room temperature for more than 100 days. In addition, electrons are released from dosimetric electron traps when OSL is generated via visible-light stimulation (at wavelengths of 390–780 nm) or at 190 °C temperature [12]. For deep electron traps, delocalization conditions have been reported to involve a temperature of 900 °C and ultraviolet irradiation [12,14,15]. Furthermore, Umisedo et al. have previously demonstrated that wavelengths less than 495 nm force electrons in deep
http://dx.doi.org/10.1016/j.ejmp.2017.05.057 1120-1797/Ó 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
82
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
electron traps to move to dosimetric electron traps, which may affect the overall luminescence [16]. In OSLDs, electrons released from dosimetric electron traps recombine with holes at the luminescence center, emitting light with peaks centered at 410– 420 nm; this light is associated with the absolute dose on the OSLD and is used as a signal for dose determination [8,17]. On the other hand, the recombination of released electrons and holes in deep hole traps leads to heat emission, and hence cannot be used as a signal for dose determination. Deep hole traps compete with luminescent centers for released electrons and, therefore, reduce the overall luminescence. Omotayo et al. have reported that the signals in OSLDs are primarily affected by the bleaching time, bleaching source wavelength spectrum, stimulation wavelength, and accumulated dose [15]. By using the OSL energy-band model, it has been shown that competition between luminescence centers and deep hole traps, as well as between dosimetric and deep electron traps, causes variation in the OSLD dose linearity and dose sensitivity with accumulated dose [3,12]. For doses greater than 3 Gy, the OSL signal-todose (S/D) ratio trend exhibits a supra-linear curvature because a high dose leads to the filling of the deep electron traps; consequently, dosimetric electron traps are more likely to be filled, causing the gradient of the OSL S/D curve to increase [18,19]. In contrast to these results, other study has shown that supra-linearity was observable at a dose less than 1 Gy [12]. In various studies, the dose sensitivity has exhibited different trends in response to the accumulated dose under different conditions of bleaching time and bleaching-device wavelength spectrum [3,12,15,20]. It has been demonstrated that the change in dose sensitivity results from the complexity of the competing mechanisms involving the various free electron and hole traps. Thus, Jursinic has suggested the filling of the deep electron and hole traps to remove the effect of the OSLD supra-linearity [12]. By using the Jursinic’s approach, the dose linearity could be maintained through the preirradiation of OSLDs with dose values greater than 1 kGy. Despite the previous research on OSLDs (OSLDfull, which are OSLDs with deep electron/hole traps fully filled by pre-irradiation with dose values greater than 1 kGy), uncertainties associated with dose sensitivity still hinder the clinical use of OSLDs [7,12,21]. Two factors affecting the stable use of OSLDs are the bleaching conditions and the competition discussed in terms of the energy-band model above. In general, the dose sensitivities of OSLDfull can be changed without consideration of the bleaching condition because of the competition between the optically released electrons and deep hole traps. Therefore, it is of interest to determine whether the OSLDfull sensitivity stabilizes with respect to the accumulated dose as a result of the full filling of the free electrons and holes in the deep electron/hole traps; similarly, investigations of the optimal bleaching sources and times for such devices are also important. The present study aimed to evaluate the characteristics of OSLDfull and to determine the optimal bleaching conditions for reducing their dose measurement uncertainty. In addition, the clinical use of OSLDfull with low variation in dose sensitivity or linearity was investigated.
mode with an illumination read period of 1 s using weakstimulation light-emitting diode (LED) modes. InLight MicroStar reader uses the green stimulation LED having a median wavelength of 530 nm for weak-stimulation of OSLDs. This wavelength forces electrons to be moved from dosimetric electron traps and doesn’t have a great influence on the sensitivity change of OSLDs. Before readout, measurements to check the stability of the reader were performed 3–5 times using three of the reader modes: ‘‘DRK,” ‘‘CAL,” and ‘‘LED.” These measurements were conducted by assessing the photomultiplier tube (PMT) signal in response to stimuli. Note that ‘‘DRK” represents the PMT response without stimulus, which is an indicator of the electric noise or dark current. ‘‘CAL” is the PMT response to C-14 (T1/2 = 5730 years) encapsulated in a powdered phosphor. Finally, ‘‘LED” is the PMT response to the LED source in the reader used to stimulate OSLDs during readout; this is a reading of the counts recorded when the LED is activated [15]. After these standard measurements, each OSLD signal was recorded an average value of five readings, yielding a coefficient of variation (CV) < 0.9%. After the OSLDs had been read, a bleaching device (Hanil Nuclear, Inc., Gyeonggi-do, Korea) was used for bleaching. 36 LED chips (ATI-5730PWHB-L, ATI LED Inc., Jiangsu, China) with 15-W 6100-K color temperature (white light) measuring 38.0 3.4 cm2 were used as a bleaching source in the OSL dosimetry system. The bleaching source had a wide wavelength in the 400–750-nm range, with two peaks at approximately 450 (blue) and 550 nm (yellow). A 520 nm long-pass filter (Edmund Optics, Inc., Barrington, NJ) was applied to block LED light components with wavelengths less than 520 nm. This filter was used selectively for the experiment including the dose sensitivity, linearity and finding optimal bleaching conditions. The wavelength spectra of the bleaching source for unfiltered and filtered light were measured 2–3 times at several points (anterior, posterior, middle, left, and right sections in bleaching source) using a USB4000 spectrometer (Ocean Optics, Inc., Dunedin, FL) and there has been no difference between those spectra. Then, we have chosen the wavelengths of bleaching source measured as Fig. 1.
2.2. Pre-irradiation and experimental setup The experimental sample consisted of 60 OSLDs that were preirradiated with an accumulated dose of 5 kGy by using a Co-60 gamma-ray source. The Co-60 source (MDS Nordion, Ottawa, ON,
2. Materials and methods 2.1. OSL dosimetry system The OSL dosimeter used in the present study was an InLight nanoDot device (Landauer, Inc., Glenwood, IL), which was composed of Al2O3:C in the form of a disk of 5 mm diameter and 0.2 mm thickness within a 10 10 2 mm3 lightproof plastic case. The OSLDs were read with an InLight MicroStar reader (Landauer, Inc., Glenwood, IL) operated in the continuous-wave (CW)
Fig. 1. Wavelength spectra of unfiltered and filtered bleaching devices. A 520 nm long-pass filter was used to block the LED light components with wavelengths less than 520 nm.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Canada) had 3% uncertainty in dose delivery. This pre-irradiation process was conducted to fill the deep electron/hole traps fully (i.e., to obtain OSLDfull, following the approach of Jursinic et al. [12]). Although Jursinic recommended a pre-irradiation dose of 1 kGy, 5 kGy dose irradiation was chosen in this study to ensure the delivery of a high dose to the 60 OSLDfull devices. We also designated 30 OSLDs as the control group; these devices were not preirradiated (OSLDempty). A 6 MV photon beam delivered with a Varian ClinacÒ iX linear accelerator (Varian Medical Systems, Palo Alto, CA) has been used for all OSLD irradiations in this study past the pre-irradiation. To account for the daily output variation, absolute dose measurements were performed using a Farmer-type ionization chamber (model N30013, PTW, Freiburg, Germany) and a water phantom before each OSLD irradiation, in accordance with the American Association of Physicists in Medicine TG-51 protocol [22]. For the experiments, the OSLDs were positioned above a 10 cm solid water phantom (Standard Grade Solid Water, Gammex, Middleton, WI) in order to deliver a sufficient backscattered dose, and a 1.5 cm bolus was placed on the OSLDs to prevent air-gap effects. A source-to-surface distance (SSD) was 100 cm. All sets of OSLDs were exposed to dose values ranging from 1 to 10 Gy for each experiment. 2.3. Characteristics of OSLDfull
2.3.1. First bleaching after pre-irradiation Before further experimentation, the OSLDfull were first bleached in order to remove the large number of electrons captured in the dosimetric traps as a result of the supplied dose values of up to 5 kGy. Ten 5-kGy-irradiated OSLDfull from the experimental sample were used to determine the bleaching time necessary to remove the trapped electrons from dosimetric traps. The bleaching was performed under filtered bleaching conditions attaching 520 nm long-pass filter to LED source in bleaching device, and for each examined device, the OSLDfull signal was recorded as a function of bleaching time until the signal was stable below the low level of residual signals. For further experimentation, residual signals were defined as the signal values less than 50 which could be converted to a dose value of about 0.01 cGy corresponding to a background value for dose. After investigating the required bleaching time in the initial experiment, the remaining 40 OSLDfull were bleached in preparation for the subsequent experiments. 2.3.2. Fading The fading effect is one of the primary OSLD characteristics, and it occurs as a result of free-electron release in shallow electron traps, which occurs even at room temperature. For the experiments following the initial bleaching, two sets of five OSLDfull each were subjected to dose values of 1 Gy for one set and 10 Gy for the other. Immediately after the OSLD irradiation, the signals were recorded as functions of elapsed time until constant signals were obtained. For comparison, two sets of OSLDempty were investigated in the same manner. 2.3.3. Regeneration of the luminescence The regeneration of the luminescence should be considered as an important correction factor for OSL dosimetry because OSLD signals could increase over time even in the absence of radiation in a dark room and then background dose values could be altered. Jursinic et al. demonstrated regeneration of luminescence when in the dark room was due to filling deep electron traps and was more prominent with high accumulated dose [12]. In this study, we located one set of five OSLDfull in dark room. The readouts from the one set of five OSLDfull were recorded as a function of time.
83
For comparison, one set of five OSLDempty were investigated in the same manner. 2.4. Effects of various bleaching conditions on OSLDfull and OSLDempty In OSL dosimetry, competition between deep electron traps and deep hole traps for electrons and holes, respectively, is the primary cause of changes in the OSLD characteristics. In addition, certain bleaching conditions may affect the electrons/holes in a certain trap, causing them to move to another trap. The wavelength of blue light (shorter than 495 nm) with a long bleaching time increases the probability of partial emptying of deep electron traps by electrons [16]. Based on the theory discussed above, it was presumed that the electrons in the deep traps could affect the dose sensitivity and linearity of the OSLDs examined in this study. 2.4.1. Experimental strategy for determination of optimal bleaching conditions The optimal bleaching conditions investigated here correspond to the bleaching time and bleaching-device wavelength for which the OSLD signal is stable below an OLSD signal baseline, with the dose sensitivity and linearity remaining unchanged. In the OSLD energy-band model, it was assumed that the maximum possible number of electrons and holes was captured in a stable manner by the deep electron/hole traps (i.e., the deep traps were fully filled) under the optimal bleaching conditions. In the case of the unfiltered and filtered bleaching device, the appropriate bleaching times of OSLDfull and OSLDempty were obtained when the OSLD signals were stable for a constant signal (below a low level of residual signal). Five OSLDfull and five OSLDempty for each set were used for both filtered and unfiltered bleaching sources. The bleaching device was filtered using a 520 nm long-pass filter. Further, in order to evaluate the effect of the bleaching time and wavelength combination on the variation in dose sensitivity with the accumulated dose and dose linearity, a longer bleaching time for the unfiltered and filtered bleaching device was used for comparison. 2.4.2. Variation in dose sensitivity with accumulated dose To assess the variation in dose sensitivity with the accumulated dose, two sets which have ten OSLDfull and ten OSLDempty in each set were exposed to a dose of 5 Gy, which corresponds to the general delivered maximum dose per fraction (excluding stereotactic ablative radiotherapy doses) typically administered in clinical applications. In the case of the unfiltered bleaching device, one set underwent repeated irradiation-readout-bleaching-readout cycles with 5 Gy irradiation intervals, until the sensitivity tendencies of OSLDfull and OSLDempty were observed under the appropriate bleaching time and a longer bleaching time. The results were recorded as S/D values as functions of the accumulated doses. The other sets were bleached for the two different bleaching times with the filtered bleaching device after the OSLDs were irradiated to 5 Gy under the same irradiation setup and read in the reader. This irradiation-readout-bleaching-readout cycle was repeated several times to quantify the sensitivity behavior. The results were recorded as S/D values as functions of the accumulated dose. 2.4.3. Dose linearity To examine the dose linearity, ten OSLDfull and ten OSLDempty were employed. The doses delivered to the OSLDfull and OSLDempty were 0.1, 0.2, 0.4, 0.7, 1, 2, 4, and 10 Gy. The bleaching process was conducted using the two bleaching conditions discussed in Section 2.4.2. That is, the appropriate bleaching times for an unfiltered bleaching device determined in Section 2.4.2. and a longer bleaching time of 240 h for a filtered bleaching device. The supra-linearity index (f(D)) was used to quantify the deviation from linearity; [17]
84
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
this index is defined as the S/D value at any dose D divided by the reference signal per dose, assuming that the dose response is linear. Since the reference dose was defined as 1 Gy, f(D) is expressed as
f ðDÞ ¼ ðS=DÞ=ðS1Gy =D1Gy Þ
ð1Þ
3. Results 3.1. Characteristics of OSLDfull
3.1.1. First bleaching after pre-irradiation Fig. 2 shows the OSLDfull signal as a function of bleaching time obtained under filtered bleaching conditions, for the initial bleaching of the OSLDfull devices after the 5 kGy pre-irradiation. The initial mean OSLDfull signal (before bleaching) obtained for the ten examined devices was 7.5 105 counts. Within 600 min, the mean OSLDfull signal had decreased exponentially to less than approximately 3300 counts. Following the initial 600 min bleaching, the slope of the mean OSLD signal decreased slowly. Finally, a mean signal of 43.2 was obtained at 28,800 min (480 h). The 480 h bleaching time was sufficient to lower the OSLDfull signal below the low level of residual level after these devices had been irradiated with a dose of 5 kGy. 3.1.2. Fading The fading effects for four sets of 5 OSLDfull and 5 OSLDempty exposed to 1 or 10 Gy are shown in Fig. 3. Within 10 min, the mean signal obtained from the results for the 10 OSLDfull and 10 OSLDempty decreased exponentially with time, independent of the delivered dose. The mean signal did not vary in the period from 10 to 60 min. The 10 min fading time was sufficient for free electrons to be released from the shallow electron traps for both OSLDfull and OSLDempty. Therefore, it was possible to read out the results of OSLDfull and OSLDempty after a minimum of 10 min following irradiation. 3.1.3. Regeneration of the luminescence Fig. 4 shows the measurements of signals for five OSLDfull and five OSLDempty after several times in the dark room following bleaching to the background values. Four hours was used as a
Fig. 2. Mean OSLDfull signal as a function of bleaching time with a filtered bleaching source. Each OSLDfull was read, assuming that the loss of the OSLDfull signal with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
Fig. 3. Mean signals for OSLDfull and OSLDempty exposed to doses of 1 and 10 Gy as functions of fading time. The average of three readouts was obtained for each data point, assuming that the loss of the signal of OSLDfull and OSLDempty with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
bleaching time for these experiments. Right after the bleaching time, the mean values of five OSLDfull and five OSLDempty were 35.0 and 26.0, respectively. The curves of mean signal had linear relationship to the time after bleaching. At the time of 13 days, the mean signals of 5 OSLDfull and 5 OSLDempty were 3359.3 and 2719.0 counts in dark room, respectively which corresponded to doses of 7.3 and 6.7 cGy. 3.2. Effects of various bleaching conditions on OSLDfull and OSLDempty As shown in Fig. 5(a), the optimal bleaching time was determined to be 4 h for an unfiltered bleaching source for both OSLDfull and OSLDempty. The reduction in stored signal has decreased to a level below the residual signal and additional bleaching time has no further effect. For a filtered bleaching source, we confirmed that 240 h is the appropriate bleaching time. Fig. 6 shows the effect of bleaching time on the dose sensitivity of OSLDempty and OSLDfull for an unfiltered bleaching device, for accumulated dose values ranging from 5 to 115 Gy. With a bleaching time of 4 h, the mean sensitivity of OSLDfull was stable within a
Fig. 4. The measurements of mean signals for OSLDfull and OSLDempty in the dark room after bleaching are shown. Error bars are the standard deviation of the mean.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
85
standard deviation of 0.85%, whereas the mean sensitivity of OSLDempty decreased by approximately 2.3% per 10 Gy. The mean sensitivities of OSLDfull for all accumulated dose values were greater than those of OSLDempty by a factor of 1.42. For the first 48 h (long bleaching time) of bleaching, the sensitivity rapidly decreased by 6.0% and 10.5% for the OSLDempty and OSLDfull sets, respectively. Subsequently, the sensitivity decreased evenly by 3.1% and 5.4% per 10 Gy for the OSLDempty and OSLDfull sets, respectively. For the filtered bleaching device, the sensitivity of OSLDfull did not change; however, the sensitivity of OSLDempty varied, as shown in Fig. 7. The mean sensitivity of OSLDfull was stable with a maximum standard deviation of 0.71% relative to the first sensitivity value of OSLDfull (which was used as a reference). In contrast, the mean sensitivity of OSLDempty decreased by approximately 4.2% per 10 Gy with the same filtered bleaching source. Fig. 8 shows the f(D) of OSLDfull and OSLDempty devices for dose responses ranging from 0.1 to 10 Gy. The mean values of f(D) for OSLDfull were 1.001 ± 0.001 for all doses under both 4 h bleaching with the unfiltered bleaching device and 240 h with the filtered bleaching device. The mean values of f(D) for OSLDempty were 1.001 ± 0.000 for doses less than 1 Gy showing a linear response to doses while those for OSLDempty were 1.000, 1.006, 1.012, and 1.021 for 2, 4, 7, and 10 Gy, respectively, under 4 h bleaching with the unfiltered bleaching device. In the case of bleaching for 240 h with the filtered bleaching device, the mean values of f(D) for OSLDempty were 1.000 ± 0.001 for doses less than 1 Gy while the values of f(D) for OSLDempty were 1.001, 1.005, 1.011, and 1.019 for 2, 4, 7, and 10 Gy, respectively.
4. Discussion
Fig. 5. Mean signals observed for OSLDempty and OSLDfull exposed to 5 Gy as functions of bleaching time, for (a) an unfiltered bleaching device and (b) a filtered bleaching device. Each OSLD was read five times, assuming that the loss of the OSLD signal with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
Fig. 6. Change in mean sensitivity of OSLDempty and OSLDfull with respect to the accumulated dose and bleaching time for an unfiltered bleaching device. The OSLDempty which was already exposed to 10 Gy was used. In this study, the sensitivity was defined as the signal per dose, given that the delivered dose was 5 Gy at every cycle. For accumulated doses ranging from 5 to 65 Gy, a bleaching time of 4 h was used, whereas for doses beyond 65 Gy, a bleaching time of 48 h was used. The error bars indicate the standard deviations.
The aim of this work was to assess the characteristics of OSLDfull and to find the optimal bleaching conditions for achieving a highly accurate and stable OSL dosimetric system. Jursinic has reported that the dose sensitivity of OSLDempty decreased by 4% per 10 Gy for accumulated doses greater than 20 Gy and the probability of electron transference from the deep electron traps to dosimetric electron traps increased at wavelengths less than 495 nm [3]. In this study, the dose sensitivity of OSLDempty shown in Fig. 5
Fig. 7. Change in mean sensitivity of OSLDempty and OSLDfull with respect to accumulated dose for a filtered bleaching source. In this study, the sensitivity was defined as the signal per dose, given that the delivered dose was 5 Gy at every cycle. For accumulated doses ranging from 5 to 30 Gy, a bleaching time of 240 h was used, whereas for doses beyond 30 Gy, a bleaching time of 480 h was used. The error bars indicate the standard deviations.
86
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Fig. 8. Dose linearity of 10 OSLDfull and 10 OSLDempty to doses ranging from 0.1 to 10 Gy for two types of bleaching conditions: 4 h with an unfiltered bleaching device (4 h, unfiltered) and 10 days with a filtered bleaching device (240 h, filtered). The supra-linearity index is defined as the ratio between the signal per dose at a certain dose and the signal per dose at 1 Gy, assuming that the dose response for a 1 Gy dose is linear.
decreased beyond 10 Gy because the OSLDempty already had an accumulated dose of approximately 10 Gy. A long bleaching time with a filtered bleaching device also changed the dose sensitivity of OSLDempty; this result is shown in Fig. 7. The changes in dose sensitivity with respect to accumulated dose originated from the failure to fill the deep electron/hole traps fully, regardless of the bleaching conditions employed. Scarboro et al. have demonstrated the characteristics of OSLDs in patient-specific computed tomography (CT) dosimetry by investigating important dosimetry behaviors such as signal depletion, signal fading, dose linearity, angular dependence, and energy dependence [23]. However, Scarboro et al. did not discuss changes in dose sensitivity, because the characteristics of OSLDempty were investigated under a low accumulated dose. Note that OSLDempty could be applied for the measurement of low dose at an accumulated dose of less than 20 Gy in applications such as cone beam computed tomography (CBCT) and CT dosimetry [23]. In the present study, the sensitivity of OSLDfull did not change with the accumulated dose when the OSLDfull were bleached for the optimal bleaching time of 4 h with an unfiltered bleaching source, as shown in Fig. 6. Further, the OSLDfull sensitivity remained constant when the optimal bleaching time of 240 h with a filtered source was applied for bleaching, despite employing a very long bleaching time (greater than the 240 h bleaching time for which a linear response to dose was obtained). The optimal bleaching time of 240 h is not suitable in the clinic. When the spectra of filtered bleaching source were measured, the absolute intensity under filtered condition was five times lower than those under unfiltered condition which was the reason why efficiency of filtered bleaching was low. Further study using a long-pass filter having a small thickness will be performed in the future. The dose sensitivity of OSLDfull decreased for 48 h bleaching time with the unfiltered bleaching device. Overall, it was empirically demonstrated that a filtered bleaching device with a longer wavelength does not affect the electrons in the deep electron traps, irrespective of the bleaching time, whereas an unfiltered bleaching device could affect the electrons. Therefore, the use of an unfiltered bleaching device is inappropriate. However, it was experimentally determined that the dose sensitivity of OSLDfull was unchanged, as shown in Fig. 6. We are unable to explain this phenomenon theo-
retically, which is a limitation of this study and analysis of the energy level of OSLDfull will be performed in the future. The results of the investigation of the bleaching-wavelength effect are similar to those of a previous study, which reported a quite stable sensitivity for OSLDs exposed to 1 and 10 Gy irradiations with 26 W fluorescent lamps, a long-pass filter, and a bleaching time of 2000 min [15]. The authors of that study concluded that OSLDs may be re-used with an accuracy of 3% and 5% for 7 and 70 Gy accumulated dose, respectively. The OSLDs used in that study did not have fully filled deep electron/hole traps, and the changes in dose sensitivity with respect to the accumulated dose were due to competition between the OSLD luminescence centers and deep hole traps, as well as between the dosimetric electron traps and deep electron traps. These changes may constitute large uncertainties in the corresponding dosimetric results. Dose linearity has also been reported in several studies. For OSLDempty, the response to D is supra-linear above 3 Gy [3], 3.5 Gy [28], and 1 Gy [29], which is consistent with our results (Fig. 8). Further, Jursinic observed that the dose-response behavior of OSLDfull is linear up to a dose of 10 Gy [3], which is consistent with our results (Fig. 8). Therefore, it has been demonstrated that OSLDs with fully filled deep electron/hole traps have a linear dose response up to a dose of 10 Gy. Pre-irradiation of OSLD with doses larger than 5 kGy using a linear accelerator may be impractical and the use of OSLDfull could be limited in clinic. In a limited situation, the uncertainty of OSLDempty should be checked for in vivo dosimetry and auditing. The uncertainty of OSLDempty could be increased by two characteristics; dose sensitivity and dose linearity. In this study, it has been demonstrated that maximum variations of dose sensitivity and dose linearity of OSLDempty were 6.0 and 1.9% and periodic calibration of OSLDempty should be required. Fading behavior has been investigated in several previous studies [3,7,15,24–26]. For the current experimental conditions, the fading time of OSLDfull is in close agreement with that for OSLDempty. Specifically, it was previously reported that the transient signal due to free-electron release from shallow electron traps began to disappear within a period of 8 min [3]. Furthermore, Omotayo et al. observed a similar fading time of 10 min when OSLDempty were exposed to 1 or 10 Gy, although the initial slopes of the decay curves depended on the delivered dose [15]. In addition, Dunn et al. performed a fading experiment with OSLDs exposed to 2 Gy and demonstrated that the time to readout should be greater than 16 min [24]. In our study, the OSLDfull signal was shown to be stable within a standard deviation of 0.4% for a fading time ranging from 10 to 60 min (Fig. 3). As shown in Fig. 3, the signal is stable within a standard deviation of 2% for 2.5 days [12,27], and a 4–5% reduction over 9 months from the initial signal has been reported [24]. Our study suggests that a highly stable OSLD dosimetry system can be constructed with fully filled deep traps and appropriate bleaching conditions. Fig. 9 shows the proposed procedure for the setup of a highly stable OSLDfull system for use in radiation therapy. The following guidelines should be followed: 1. OSLDfull should be bleached with filtered bleaching source to remove electrons from only dosimetric electron traps. 2. Characteristics of OSLDfull in terms of the fading and regeneration of luminescence conditions has similar tendency with those of OSLDempty. Fading time suggested in our study is 10 min and maximum doses increased by regeneration of luminescence in dark room is 7.3 cGy after 13 days. 3. OSLDfull has dose linearity for exposed doses over a range of 10 Gy. 4. The dose sensitivity of OSLDfull can be stable under the optimal bleaching conditions.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Fig. 9. Flowchart for the setup of a highly stable OSLDfull system for use in radiation therapy.
5. Conclusions We investigated the characteristics of OSLDs with fully filled deep electron/hole traps (OSLDfull) and found the optimal bleaching conditions for OSL dosimetry by considering the effects of the bleaching time and the bleaching wavelength on the dose sensitivity and linearity of OSLDfull. Clinical dosimetry with OSLDfull having fully filled deep traps under appropriate bleaching conditions was shown to be highly stable and accurate, with no change in either the dose sensitivity or linearity. Acknowledgements This work was supported by a National Research Foundation of Korea (NRF) grant, funded by the Korea government (MISP) (No. 2014M2B2A4031164), and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI14C3459). The authors would like to thank the following groups and individuals for their assistance: Korea Multi-Purpose Accelerator Complex (KOMAC) for initiating this work, Yoonseok Ka in the Nano Science and Technology program at Seoul National University for measuring the wavelength spectrum of the bleaching source, Hanil Nuclear for providing the bleaching device, Nascimento Luana Freitas in SCKCEN Belgian Nuclear Research Centre for reviewing this manuscript carefully and Hyunseok Lee in Korea Institute of Nuclear Safety and Hyunseok Kim in Convergence Science and Technology at Seoul National University for giving assistance in this study. References [1] McKeever SW, Moscovitch M. On the advantages and disadvantages of optically stimulated luminescence dosimetry and thermoluminescence dosimetry. Radiat Prot Dosimetry 2003;104:263–70. [2] Kerns JR, Kry SF, Sahoo N, Followill DS, Ibbott GS. Angular dependence of the nanoDot OSL dosimeter. Med Phys. 2011;38:3955–62.
87
[3] Jursinic PA. Characterization of optically stimulated luminescent dosimeters, OSLDs, for clinical dosimetric measurements. Med Phys. 2007;34:4594–604. [4] Jursinic PA. Angular dependence of dose sensitivity of nanoDot optically stimulated luminescent dosimeters in different radiation geometries. Med Phys. 2015;42:5633–41. [5] Lehmann J, Dunn L, Lye JE, Kenny JW, Alves AD, Cole A, et al. Angular dependence of the response of the nanoDot OSLD system for measurements at depth in clinical megavoltage beams. Med Phys. 2014;41:061712. [6] Lee SY, Lee KJ. Development of a personal dosimetry system based on optically stimulated luminescence of alpha-Al2O3: C for mixed radiation fields. Appl Radiat Isot 2001;54:675–85. [7] Mrcela I, Bokulie T, Izewska J, Budanec M, Frobe A, Kusic Z. Optically stimulated luminescence in vivo dosimetry for radiotherapy: physical characterization and clinical measurements in Co-60 beams. Phys Med Biol 2011;56:6065–82. [8] McKeever SWS, Akselrod MS, Colyott LE, Larsen NA, Polf JC, Whitley V. Characterisation of Al2O3 for use in thermally and optically stimulated luminescence dosimetry. Radiat Prot Dosimetry 1999;84:163–8. [9] Lye J, Dunn L, Kenny J, Lehmann J, Kron T, Oliver C, et al. Remote auditing of radiotherapy facilities using optically stimulated luminescence dosimeters. Med Phys. 2014;41. [10] Casey KE, Alvarez P, Kry SF, Howell RM, Lawyer A, Followill D. Development and implementation of a remote audit tool for high dose rate (HDR) Ir-192 brachytherapy using optically stimulated luminescence dosimetry. Med Phys. 2013;40. [11] Casey K, Alvarez P, Lawyer A, Kry S, Howell R, Davidson S, et al. Development and implementation of a remote audit tool for high dose rate (HDR) Ir-192 brachytherapy using optically stimulated luminescence dosimetry. Med Phys 2012;39:3721. [12] Jursinic PA. Changes in optically stimulated luminescent dosimeter (OSLD) dosimetric characteristics with accumulated dose. Med Phys. 2010;37:132–40. [13] Yukihara EG, McKeever SW. Optically stimulated luminescence (OSL) dosimetry in medicine. Phys Med Biol 2008;53:R351–79. [14] Yukihara EG, Whitley VH, Polf JC, Klein DM, McKeever SWS, Akselrod AE, et al. The effects of deep trap population on the thermoluminescence of Al2O3: C. Radiat Meas 2003;37:627–38. [15] Omotayo AA, Cygler JE, Sawakuchi GO. The effect of different bleaching wavelengths on the sensitivity of Al2O3: C optically stimulated luminescence detectors (OSLDs) exposed to 6 MV photon beams. Med Phys. 2012;39:5457–68. [16] Umisedo NK, Yoshimura EM, Gasparian PBR, Yukihara EG. Comparison between blue and green stimulated luminescence of Al2O3:C. Radiat Meas 2010;45:151–6. [17] Markey BG, Colyott LE, Mckeever SWS. Time-resolved optically stimulated luminescence from Alpha-Al2O3-C. Radiat Meas 1995;24:457–63. [18] Edmund JM, Andersen CE, Marckmann CJ, Aznar MC, Akselrod MS, BotterJensen L. CW-OSL measurement protocols using optical fibre Al2O3: C dosemeters. Radiat Prot Dosimetry 2006;119:368–74. [19] Yukihara EG, Gaza R, McKeever SWS, Soares CG. Optically stimulated luminescence and thermoluminescence efficiencies for high-energy heavy charged particle irradiation in Al2O3: C. Radiat Meas 2004;38:59–70. [20] Han SC, Choi SH, Park S, Kim CH, Jung H, Kim M, et al. Evaluation of dosimetric characteristics of reproducibility, linearity and dose dependence of optically stimulated luminescence dosimeters in Co-60 gamma-rays. Pro Med Phys 2014;25:31–6. [21] Yu M, Jang HS, Jeon DM, Cheon GS, Lee HC, Chung MJ, et al. Dosimetric evaluation of tomotherapy and four-box field conformal radiotherapy in locally advanced rectal cancer. Radiat Oncol J 2013;31:252–9. [22] Almond PR, Biggs PJ, Coursey BM, Hanson WF, Huq MS, Nath R, et al. AAPM’s TG-51 protocol for clinical reference dosimetry of high-energy photon and electron beams. Med Phys 1999;26:1847–70. [23] Scarboro SB, Cody D, Alvarez P, Followill D, Court L, Stingo FC, et al. Characterization of the nanoDot OSLD dosimeter in CT. Med Phys 2015;42:1797–807. [24] Dunn L, Lye J, Kenny J, Lehmann J, Williams I, Kron T. Commissioning of optically stimulated luminescence dosimeters for use in radiotherapy. Radiat Meas 2013;51–52:31–9. [25] Yukihara EG, Gasparian PBR, Sawakuchi GO, Ruan C, Ahmad S, Kalavagunta C, et al. Medical applications of optically stimulated luminescence dosimeters (OSLDs). Radiat Meas 2010;45:658–62. [26] Reft CS. The energy dependence and dose response of a commercial optically stimulated luminescent detector for kilovoltage photon, megavoltage photon, and electron, proton, and carbon beams. Med Phys 2009;36:1690–9. [27] BotterJensen L, Larsen NA, Markey BG, McKeever SWS. Al2O3: C as a sensitive OSL dosemeter for rapid assessment of environmental photon dose rates. Radiat Meas 1997;27:295–8. [28] Yukihara EG, Yoshimura EM, Lindstrom TD, Ahmad S, Taylor KK, Mardirossian G. High-precision dosimetry for radiotherapy using the optically stimulated luminescence technique and thin Al2O3: C dosimeters. Phys Med Biol 2005;50:5619–28. [29] Polf JC, Yukihara EG, Akselrod MS, McKeever SWS. Real-time luminescence from Al2O3 fiber dosimeters. Radiat Meas 2004;38:227–40.
Contents lists available at ScienceDirect
Physica Medica journal homepage: http://www.physicamedica.com
Original paper
Sensitivity and stability of optically stimulated luminescence dosimeters with filled deep electron/hole traps under pre-irradiation and bleaching conditions So-Yeon Park a,b,c, Chang Heon Choi a,b,c, Jong Min Park a,b,c,d, Minsoo Chun a,b,c, Ji Hye Han a,c, Jung-in Kim a,b,c,⇑ a
Department of Radiation Oncology, Seoul National University Hospital, Seoul, Republic of Korea Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Republic of Korea c Biomedical Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea d Center for Convergence Research on Robotics, Advance Institutes of Convergence Technology, Suwon, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 1 September 2016 Received in Revised form 14 March 2017 Accepted 10 May 2017 Available online 17 May 2017 Keywords: Dosimetry Optically stimulated luminescence Optimal bleaching condition nanoDot
a b s t r a c t Purpose: We aimed to evaluate the characteristics of optically stimulated luminescence dosimeters (OSLDs) with fully filled deep electron/hole traps, and determine the optimal bleaching conditions for these OSLDs to minimize the changes in dose sensitivity or linearity according to the accumulated dose. Methods: InLight nanoDots were used as OSLDs. The OSLDs were first pre-irradiated at a dose greater than 5 kGy to fill the deep electron and hole traps, and then bleached (OSLDfull). OSLDfull characteristics were investigated in terms of the full bleaching, fading, regeneration of luminescence, dose linearity, and dose sensitivity with various bleaching conditions. For comparison, OSLDs with un-filled deep electron/hole traps (OSLDempty) were investigated in the same manner. Results: The fading for OSLDfull exhibited stable signals after 10 min, for 1 and 10 Gy. The mean supralinear index values for OSLDfull were 1.001 ± 0.001 for doses from 2 to 10 Gy. Small variations in dose sensitivity were obtained for OSLDfull within standard deviations of 0.85% and 0.71%, whereas those of OSLDempty decreased by 2.3% and 4.2% per 10 Gy for unfiltered and filtered bleaching devices, respectively. Conclusions: Under the bleaching conditions determined in this study, clinical dosimetry with OSLDfull is highly stable, minimizing the changes in dose sensitivity or linearity for the clinical dose. Ó 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
1. Introduction The use of anion-deficient sapphire doped with carbon (Al2O3: C) to obtain optically stimulated luminescence (OSL) is known to have advantages for dosimetry, such as good reproducibility, low energy dependency, and low angular dependency compared to LiF thermoluminescence (TL) dosimetry for energies above MeV [1–5]. Therefore, OSL dosimeters (OSLDs) are widely used for not only in vivo dosimetry for cancer patients in the field of radiation therapy, but also personal radiation monitoring in the field of radiation protection [6–8]. Beyond in vivo dosimetry and monitoring, OSLDs are also an important tool for auditing and research in the field of radiation therapy [9–11]. ⇑ Corresponding author at: Department of Radiation Oncology, Seoul National University Hospital, Seoul 110-744, Republic of Korea. E-mail address: [email protected] (J.-i. Kim).
Several studies have explained the OSL mechanism using the energy-band model [3,12–14]. When OSLDs are irradiated, free electrons and holes are first generated, and they subsequently separate. The free electrons can potentially be captured by large numbers of shallow, dosimetric, or deep electron traps, depending on the energy band levels [3]. Further, the holes can be captured in deep hole traps and at luminescence centers. Electrons captured by shallow electron traps are released at room temperature within a few minutes, whereas electrons captured by dosimetric electron traps can be held at room temperature for more than 100 days. In addition, electrons are released from dosimetric electron traps when OSL is generated via visible-light stimulation (at wavelengths of 390–780 nm) or at 190 °C temperature [12]. For deep electron traps, delocalization conditions have been reported to involve a temperature of 900 °C and ultraviolet irradiation [12,14,15]. Furthermore, Umisedo et al. have previously demonstrated that wavelengths less than 495 nm force electrons in deep
http://dx.doi.org/10.1016/j.ejmp.2017.05.057 1120-1797/Ó 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
82
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
electron traps to move to dosimetric electron traps, which may affect the overall luminescence [16]. In OSLDs, electrons released from dosimetric electron traps recombine with holes at the luminescence center, emitting light with peaks centered at 410– 420 nm; this light is associated with the absolute dose on the OSLD and is used as a signal for dose determination [8,17]. On the other hand, the recombination of released electrons and holes in deep hole traps leads to heat emission, and hence cannot be used as a signal for dose determination. Deep hole traps compete with luminescent centers for released electrons and, therefore, reduce the overall luminescence. Omotayo et al. have reported that the signals in OSLDs are primarily affected by the bleaching time, bleaching source wavelength spectrum, stimulation wavelength, and accumulated dose [15]. By using the OSL energy-band model, it has been shown that competition between luminescence centers and deep hole traps, as well as between dosimetric and deep electron traps, causes variation in the OSLD dose linearity and dose sensitivity with accumulated dose [3,12]. For doses greater than 3 Gy, the OSL signal-todose (S/D) ratio trend exhibits a supra-linear curvature because a high dose leads to the filling of the deep electron traps; consequently, dosimetric electron traps are more likely to be filled, causing the gradient of the OSL S/D curve to increase [18,19]. In contrast to these results, other study has shown that supra-linearity was observable at a dose less than 1 Gy [12]. In various studies, the dose sensitivity has exhibited different trends in response to the accumulated dose under different conditions of bleaching time and bleaching-device wavelength spectrum [3,12,15,20]. It has been demonstrated that the change in dose sensitivity results from the complexity of the competing mechanisms involving the various free electron and hole traps. Thus, Jursinic has suggested the filling of the deep electron and hole traps to remove the effect of the OSLD supra-linearity [12]. By using the Jursinic’s approach, the dose linearity could be maintained through the preirradiation of OSLDs with dose values greater than 1 kGy. Despite the previous research on OSLDs (OSLDfull, which are OSLDs with deep electron/hole traps fully filled by pre-irradiation with dose values greater than 1 kGy), uncertainties associated with dose sensitivity still hinder the clinical use of OSLDs [7,12,21]. Two factors affecting the stable use of OSLDs are the bleaching conditions and the competition discussed in terms of the energy-band model above. In general, the dose sensitivities of OSLDfull can be changed without consideration of the bleaching condition because of the competition between the optically released electrons and deep hole traps. Therefore, it is of interest to determine whether the OSLDfull sensitivity stabilizes with respect to the accumulated dose as a result of the full filling of the free electrons and holes in the deep electron/hole traps; similarly, investigations of the optimal bleaching sources and times for such devices are also important. The present study aimed to evaluate the characteristics of OSLDfull and to determine the optimal bleaching conditions for reducing their dose measurement uncertainty. In addition, the clinical use of OSLDfull with low variation in dose sensitivity or linearity was investigated.
mode with an illumination read period of 1 s using weakstimulation light-emitting diode (LED) modes. InLight MicroStar reader uses the green stimulation LED having a median wavelength of 530 nm for weak-stimulation of OSLDs. This wavelength forces electrons to be moved from dosimetric electron traps and doesn’t have a great influence on the sensitivity change of OSLDs. Before readout, measurements to check the stability of the reader were performed 3–5 times using three of the reader modes: ‘‘DRK,” ‘‘CAL,” and ‘‘LED.” These measurements were conducted by assessing the photomultiplier tube (PMT) signal in response to stimuli. Note that ‘‘DRK” represents the PMT response without stimulus, which is an indicator of the electric noise or dark current. ‘‘CAL” is the PMT response to C-14 (T1/2 = 5730 years) encapsulated in a powdered phosphor. Finally, ‘‘LED” is the PMT response to the LED source in the reader used to stimulate OSLDs during readout; this is a reading of the counts recorded when the LED is activated [15]. After these standard measurements, each OSLD signal was recorded an average value of five readings, yielding a coefficient of variation (CV) < 0.9%. After the OSLDs had been read, a bleaching device (Hanil Nuclear, Inc., Gyeonggi-do, Korea) was used for bleaching. 36 LED chips (ATI-5730PWHB-L, ATI LED Inc., Jiangsu, China) with 15-W 6100-K color temperature (white light) measuring 38.0 3.4 cm2 were used as a bleaching source in the OSL dosimetry system. The bleaching source had a wide wavelength in the 400–750-nm range, with two peaks at approximately 450 (blue) and 550 nm (yellow). A 520 nm long-pass filter (Edmund Optics, Inc., Barrington, NJ) was applied to block LED light components with wavelengths less than 520 nm. This filter was used selectively for the experiment including the dose sensitivity, linearity and finding optimal bleaching conditions. The wavelength spectra of the bleaching source for unfiltered and filtered light were measured 2–3 times at several points (anterior, posterior, middle, left, and right sections in bleaching source) using a USB4000 spectrometer (Ocean Optics, Inc., Dunedin, FL) and there has been no difference between those spectra. Then, we have chosen the wavelengths of bleaching source measured as Fig. 1.
2.2. Pre-irradiation and experimental setup The experimental sample consisted of 60 OSLDs that were preirradiated with an accumulated dose of 5 kGy by using a Co-60 gamma-ray source. The Co-60 source (MDS Nordion, Ottawa, ON,
2. Materials and methods 2.1. OSL dosimetry system The OSL dosimeter used in the present study was an InLight nanoDot device (Landauer, Inc., Glenwood, IL), which was composed of Al2O3:C in the form of a disk of 5 mm diameter and 0.2 mm thickness within a 10 10 2 mm3 lightproof plastic case. The OSLDs were read with an InLight MicroStar reader (Landauer, Inc., Glenwood, IL) operated in the continuous-wave (CW)
Fig. 1. Wavelength spectra of unfiltered and filtered bleaching devices. A 520 nm long-pass filter was used to block the LED light components with wavelengths less than 520 nm.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Canada) had 3% uncertainty in dose delivery. This pre-irradiation process was conducted to fill the deep electron/hole traps fully (i.e., to obtain OSLDfull, following the approach of Jursinic et al. [12]). Although Jursinic recommended a pre-irradiation dose of 1 kGy, 5 kGy dose irradiation was chosen in this study to ensure the delivery of a high dose to the 60 OSLDfull devices. We also designated 30 OSLDs as the control group; these devices were not preirradiated (OSLDempty). A 6 MV photon beam delivered with a Varian ClinacÒ iX linear accelerator (Varian Medical Systems, Palo Alto, CA) has been used for all OSLD irradiations in this study past the pre-irradiation. To account for the daily output variation, absolute dose measurements were performed using a Farmer-type ionization chamber (model N30013, PTW, Freiburg, Germany) and a water phantom before each OSLD irradiation, in accordance with the American Association of Physicists in Medicine TG-51 protocol [22]. For the experiments, the OSLDs were positioned above a 10 cm solid water phantom (Standard Grade Solid Water, Gammex, Middleton, WI) in order to deliver a sufficient backscattered dose, and a 1.5 cm bolus was placed on the OSLDs to prevent air-gap effects. A source-to-surface distance (SSD) was 100 cm. All sets of OSLDs were exposed to dose values ranging from 1 to 10 Gy for each experiment. 2.3. Characteristics of OSLDfull
2.3.1. First bleaching after pre-irradiation Before further experimentation, the OSLDfull were first bleached in order to remove the large number of electrons captured in the dosimetric traps as a result of the supplied dose values of up to 5 kGy. Ten 5-kGy-irradiated OSLDfull from the experimental sample were used to determine the bleaching time necessary to remove the trapped electrons from dosimetric traps. The bleaching was performed under filtered bleaching conditions attaching 520 nm long-pass filter to LED source in bleaching device, and for each examined device, the OSLDfull signal was recorded as a function of bleaching time until the signal was stable below the low level of residual signals. For further experimentation, residual signals were defined as the signal values less than 50 which could be converted to a dose value of about 0.01 cGy corresponding to a background value for dose. After investigating the required bleaching time in the initial experiment, the remaining 40 OSLDfull were bleached in preparation for the subsequent experiments. 2.3.2. Fading The fading effect is one of the primary OSLD characteristics, and it occurs as a result of free-electron release in shallow electron traps, which occurs even at room temperature. For the experiments following the initial bleaching, two sets of five OSLDfull each were subjected to dose values of 1 Gy for one set and 10 Gy for the other. Immediately after the OSLD irradiation, the signals were recorded as functions of elapsed time until constant signals were obtained. For comparison, two sets of OSLDempty were investigated in the same manner. 2.3.3. Regeneration of the luminescence The regeneration of the luminescence should be considered as an important correction factor for OSL dosimetry because OSLD signals could increase over time even in the absence of radiation in a dark room and then background dose values could be altered. Jursinic et al. demonstrated regeneration of luminescence when in the dark room was due to filling deep electron traps and was more prominent with high accumulated dose [12]. In this study, we located one set of five OSLDfull in dark room. The readouts from the one set of five OSLDfull were recorded as a function of time.
83
For comparison, one set of five OSLDempty were investigated in the same manner. 2.4. Effects of various bleaching conditions on OSLDfull and OSLDempty In OSL dosimetry, competition between deep electron traps and deep hole traps for electrons and holes, respectively, is the primary cause of changes in the OSLD characteristics. In addition, certain bleaching conditions may affect the electrons/holes in a certain trap, causing them to move to another trap. The wavelength of blue light (shorter than 495 nm) with a long bleaching time increases the probability of partial emptying of deep electron traps by electrons [16]. Based on the theory discussed above, it was presumed that the electrons in the deep traps could affect the dose sensitivity and linearity of the OSLDs examined in this study. 2.4.1. Experimental strategy for determination of optimal bleaching conditions The optimal bleaching conditions investigated here correspond to the bleaching time and bleaching-device wavelength for which the OSLD signal is stable below an OLSD signal baseline, with the dose sensitivity and linearity remaining unchanged. In the OSLD energy-band model, it was assumed that the maximum possible number of electrons and holes was captured in a stable manner by the deep electron/hole traps (i.e., the deep traps were fully filled) under the optimal bleaching conditions. In the case of the unfiltered and filtered bleaching device, the appropriate bleaching times of OSLDfull and OSLDempty were obtained when the OSLD signals were stable for a constant signal (below a low level of residual signal). Five OSLDfull and five OSLDempty for each set were used for both filtered and unfiltered bleaching sources. The bleaching device was filtered using a 520 nm long-pass filter. Further, in order to evaluate the effect of the bleaching time and wavelength combination on the variation in dose sensitivity with the accumulated dose and dose linearity, a longer bleaching time for the unfiltered and filtered bleaching device was used for comparison. 2.4.2. Variation in dose sensitivity with accumulated dose To assess the variation in dose sensitivity with the accumulated dose, two sets which have ten OSLDfull and ten OSLDempty in each set were exposed to a dose of 5 Gy, which corresponds to the general delivered maximum dose per fraction (excluding stereotactic ablative radiotherapy doses) typically administered in clinical applications. In the case of the unfiltered bleaching device, one set underwent repeated irradiation-readout-bleaching-readout cycles with 5 Gy irradiation intervals, until the sensitivity tendencies of OSLDfull and OSLDempty were observed under the appropriate bleaching time and a longer bleaching time. The results were recorded as S/D values as functions of the accumulated doses. The other sets were bleached for the two different bleaching times with the filtered bleaching device after the OSLDs were irradiated to 5 Gy under the same irradiation setup and read in the reader. This irradiation-readout-bleaching-readout cycle was repeated several times to quantify the sensitivity behavior. The results were recorded as S/D values as functions of the accumulated dose. 2.4.3. Dose linearity To examine the dose linearity, ten OSLDfull and ten OSLDempty were employed. The doses delivered to the OSLDfull and OSLDempty were 0.1, 0.2, 0.4, 0.7, 1, 2, 4, and 10 Gy. The bleaching process was conducted using the two bleaching conditions discussed in Section 2.4.2. That is, the appropriate bleaching times for an unfiltered bleaching device determined in Section 2.4.2. and a longer bleaching time of 240 h for a filtered bleaching device. The supra-linearity index (f(D)) was used to quantify the deviation from linearity; [17]
84
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
this index is defined as the S/D value at any dose D divided by the reference signal per dose, assuming that the dose response is linear. Since the reference dose was defined as 1 Gy, f(D) is expressed as
f ðDÞ ¼ ðS=DÞ=ðS1Gy =D1Gy Þ
ð1Þ
3. Results 3.1. Characteristics of OSLDfull
3.1.1. First bleaching after pre-irradiation Fig. 2 shows the OSLDfull signal as a function of bleaching time obtained under filtered bleaching conditions, for the initial bleaching of the OSLDfull devices after the 5 kGy pre-irradiation. The initial mean OSLDfull signal (before bleaching) obtained for the ten examined devices was 7.5 105 counts. Within 600 min, the mean OSLDfull signal had decreased exponentially to less than approximately 3300 counts. Following the initial 600 min bleaching, the slope of the mean OSLD signal decreased slowly. Finally, a mean signal of 43.2 was obtained at 28,800 min (480 h). The 480 h bleaching time was sufficient to lower the OSLDfull signal below the low level of residual level after these devices had been irradiated with a dose of 5 kGy. 3.1.2. Fading The fading effects for four sets of 5 OSLDfull and 5 OSLDempty exposed to 1 or 10 Gy are shown in Fig. 3. Within 10 min, the mean signal obtained from the results for the 10 OSLDfull and 10 OSLDempty decreased exponentially with time, independent of the delivered dose. The mean signal did not vary in the period from 10 to 60 min. The 10 min fading time was sufficient for free electrons to be released from the shallow electron traps for both OSLDfull and OSLDempty. Therefore, it was possible to read out the results of OSLDfull and OSLDempty after a minimum of 10 min following irradiation. 3.1.3. Regeneration of the luminescence Fig. 4 shows the measurements of signals for five OSLDfull and five OSLDempty after several times in the dark room following bleaching to the background values. Four hours was used as a
Fig. 2. Mean OSLDfull signal as a function of bleaching time with a filtered bleaching source. Each OSLDfull was read, assuming that the loss of the OSLDfull signal with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
Fig. 3. Mean signals for OSLDfull and OSLDempty exposed to doses of 1 and 10 Gy as functions of fading time. The average of three readouts was obtained for each data point, assuming that the loss of the signal of OSLDfull and OSLDempty with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
bleaching time for these experiments. Right after the bleaching time, the mean values of five OSLDfull and five OSLDempty were 35.0 and 26.0, respectively. The curves of mean signal had linear relationship to the time after bleaching. At the time of 13 days, the mean signals of 5 OSLDfull and 5 OSLDempty were 3359.3 and 2719.0 counts in dark room, respectively which corresponded to doses of 7.3 and 6.7 cGy. 3.2. Effects of various bleaching conditions on OSLDfull and OSLDempty As shown in Fig. 5(a), the optimal bleaching time was determined to be 4 h for an unfiltered bleaching source for both OSLDfull and OSLDempty. The reduction in stored signal has decreased to a level below the residual signal and additional bleaching time has no further effect. For a filtered bleaching source, we confirmed that 240 h is the appropriate bleaching time. Fig. 6 shows the effect of bleaching time on the dose sensitivity of OSLDempty and OSLDfull for an unfiltered bleaching device, for accumulated dose values ranging from 5 to 115 Gy. With a bleaching time of 4 h, the mean sensitivity of OSLDfull was stable within a
Fig. 4. The measurements of mean signals for OSLDfull and OSLDempty in the dark room after bleaching are shown. Error bars are the standard deviation of the mean.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
85
standard deviation of 0.85%, whereas the mean sensitivity of OSLDempty decreased by approximately 2.3% per 10 Gy. The mean sensitivities of OSLDfull for all accumulated dose values were greater than those of OSLDempty by a factor of 1.42. For the first 48 h (long bleaching time) of bleaching, the sensitivity rapidly decreased by 6.0% and 10.5% for the OSLDempty and OSLDfull sets, respectively. Subsequently, the sensitivity decreased evenly by 3.1% and 5.4% per 10 Gy for the OSLDempty and OSLDfull sets, respectively. For the filtered bleaching device, the sensitivity of OSLDfull did not change; however, the sensitivity of OSLDempty varied, as shown in Fig. 7. The mean sensitivity of OSLDfull was stable with a maximum standard deviation of 0.71% relative to the first sensitivity value of OSLDfull (which was used as a reference). In contrast, the mean sensitivity of OSLDempty decreased by approximately 4.2% per 10 Gy with the same filtered bleaching source. Fig. 8 shows the f(D) of OSLDfull and OSLDempty devices for dose responses ranging from 0.1 to 10 Gy. The mean values of f(D) for OSLDfull were 1.001 ± 0.001 for all doses under both 4 h bleaching with the unfiltered bleaching device and 240 h with the filtered bleaching device. The mean values of f(D) for OSLDempty were 1.001 ± 0.000 for doses less than 1 Gy showing a linear response to doses while those for OSLDempty were 1.000, 1.006, 1.012, and 1.021 for 2, 4, 7, and 10 Gy, respectively, under 4 h bleaching with the unfiltered bleaching device. In the case of bleaching for 240 h with the filtered bleaching device, the mean values of f(D) for OSLDempty were 1.000 ± 0.001 for doses less than 1 Gy while the values of f(D) for OSLDempty were 1.001, 1.005, 1.011, and 1.019 for 2, 4, 7, and 10 Gy, respectively.
4. Discussion
Fig. 5. Mean signals observed for OSLDempty and OSLDfull exposed to 5 Gy as functions of bleaching time, for (a) an unfiltered bleaching device and (b) a filtered bleaching device. Each OSLD was read five times, assuming that the loss of the OSLD signal with sequential readouts was negligible for these experiments. The error bars indicate the standard deviations.
Fig. 6. Change in mean sensitivity of OSLDempty and OSLDfull with respect to the accumulated dose and bleaching time for an unfiltered bleaching device. The OSLDempty which was already exposed to 10 Gy was used. In this study, the sensitivity was defined as the signal per dose, given that the delivered dose was 5 Gy at every cycle. For accumulated doses ranging from 5 to 65 Gy, a bleaching time of 4 h was used, whereas for doses beyond 65 Gy, a bleaching time of 48 h was used. The error bars indicate the standard deviations.
The aim of this work was to assess the characteristics of OSLDfull and to find the optimal bleaching conditions for achieving a highly accurate and stable OSL dosimetric system. Jursinic has reported that the dose sensitivity of OSLDempty decreased by 4% per 10 Gy for accumulated doses greater than 20 Gy and the probability of electron transference from the deep electron traps to dosimetric electron traps increased at wavelengths less than 495 nm [3]. In this study, the dose sensitivity of OSLDempty shown in Fig. 5
Fig. 7. Change in mean sensitivity of OSLDempty and OSLDfull with respect to accumulated dose for a filtered bleaching source. In this study, the sensitivity was defined as the signal per dose, given that the delivered dose was 5 Gy at every cycle. For accumulated doses ranging from 5 to 30 Gy, a bleaching time of 240 h was used, whereas for doses beyond 30 Gy, a bleaching time of 480 h was used. The error bars indicate the standard deviations.
86
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Fig. 8. Dose linearity of 10 OSLDfull and 10 OSLDempty to doses ranging from 0.1 to 10 Gy for two types of bleaching conditions: 4 h with an unfiltered bleaching device (4 h, unfiltered) and 10 days with a filtered bleaching device (240 h, filtered). The supra-linearity index is defined as the ratio between the signal per dose at a certain dose and the signal per dose at 1 Gy, assuming that the dose response for a 1 Gy dose is linear.
decreased beyond 10 Gy because the OSLDempty already had an accumulated dose of approximately 10 Gy. A long bleaching time with a filtered bleaching device also changed the dose sensitivity of OSLDempty; this result is shown in Fig. 7. The changes in dose sensitivity with respect to accumulated dose originated from the failure to fill the deep electron/hole traps fully, regardless of the bleaching conditions employed. Scarboro et al. have demonstrated the characteristics of OSLDs in patient-specific computed tomography (CT) dosimetry by investigating important dosimetry behaviors such as signal depletion, signal fading, dose linearity, angular dependence, and energy dependence [23]. However, Scarboro et al. did not discuss changes in dose sensitivity, because the characteristics of OSLDempty were investigated under a low accumulated dose. Note that OSLDempty could be applied for the measurement of low dose at an accumulated dose of less than 20 Gy in applications such as cone beam computed tomography (CBCT) and CT dosimetry [23]. In the present study, the sensitivity of OSLDfull did not change with the accumulated dose when the OSLDfull were bleached for the optimal bleaching time of 4 h with an unfiltered bleaching source, as shown in Fig. 6. Further, the OSLDfull sensitivity remained constant when the optimal bleaching time of 240 h with a filtered source was applied for bleaching, despite employing a very long bleaching time (greater than the 240 h bleaching time for which a linear response to dose was obtained). The optimal bleaching time of 240 h is not suitable in the clinic. When the spectra of filtered bleaching source were measured, the absolute intensity under filtered condition was five times lower than those under unfiltered condition which was the reason why efficiency of filtered bleaching was low. Further study using a long-pass filter having a small thickness will be performed in the future. The dose sensitivity of OSLDfull decreased for 48 h bleaching time with the unfiltered bleaching device. Overall, it was empirically demonstrated that a filtered bleaching device with a longer wavelength does not affect the electrons in the deep electron traps, irrespective of the bleaching time, whereas an unfiltered bleaching device could affect the electrons. Therefore, the use of an unfiltered bleaching device is inappropriate. However, it was experimentally determined that the dose sensitivity of OSLDfull was unchanged, as shown in Fig. 6. We are unable to explain this phenomenon theo-
retically, which is a limitation of this study and analysis of the energy level of OSLDfull will be performed in the future. The results of the investigation of the bleaching-wavelength effect are similar to those of a previous study, which reported a quite stable sensitivity for OSLDs exposed to 1 and 10 Gy irradiations with 26 W fluorescent lamps, a long-pass filter, and a bleaching time of 2000 min [15]. The authors of that study concluded that OSLDs may be re-used with an accuracy of 3% and 5% for 7 and 70 Gy accumulated dose, respectively. The OSLDs used in that study did not have fully filled deep electron/hole traps, and the changes in dose sensitivity with respect to the accumulated dose were due to competition between the OSLD luminescence centers and deep hole traps, as well as between the dosimetric electron traps and deep electron traps. These changes may constitute large uncertainties in the corresponding dosimetric results. Dose linearity has also been reported in several studies. For OSLDempty, the response to D is supra-linear above 3 Gy [3], 3.5 Gy [28], and 1 Gy [29], which is consistent with our results (Fig. 8). Further, Jursinic observed that the dose-response behavior of OSLDfull is linear up to a dose of 10 Gy [3], which is consistent with our results (Fig. 8). Therefore, it has been demonstrated that OSLDs with fully filled deep electron/hole traps have a linear dose response up to a dose of 10 Gy. Pre-irradiation of OSLD with doses larger than 5 kGy using a linear accelerator may be impractical and the use of OSLDfull could be limited in clinic. In a limited situation, the uncertainty of OSLDempty should be checked for in vivo dosimetry and auditing. The uncertainty of OSLDempty could be increased by two characteristics; dose sensitivity and dose linearity. In this study, it has been demonstrated that maximum variations of dose sensitivity and dose linearity of OSLDempty were 6.0 and 1.9% and periodic calibration of OSLDempty should be required. Fading behavior has been investigated in several previous studies [3,7,15,24–26]. For the current experimental conditions, the fading time of OSLDfull is in close agreement with that for OSLDempty. Specifically, it was previously reported that the transient signal due to free-electron release from shallow electron traps began to disappear within a period of 8 min [3]. Furthermore, Omotayo et al. observed a similar fading time of 10 min when OSLDempty were exposed to 1 or 10 Gy, although the initial slopes of the decay curves depended on the delivered dose [15]. In addition, Dunn et al. performed a fading experiment with OSLDs exposed to 2 Gy and demonstrated that the time to readout should be greater than 16 min [24]. In our study, the OSLDfull signal was shown to be stable within a standard deviation of 0.4% for a fading time ranging from 10 to 60 min (Fig. 3). As shown in Fig. 3, the signal is stable within a standard deviation of 2% for 2.5 days [12,27], and a 4–5% reduction over 9 months from the initial signal has been reported [24]. Our study suggests that a highly stable OSLD dosimetry system can be constructed with fully filled deep traps and appropriate bleaching conditions. Fig. 9 shows the proposed procedure for the setup of a highly stable OSLDfull system for use in radiation therapy. The following guidelines should be followed: 1. OSLDfull should be bleached with filtered bleaching source to remove electrons from only dosimetric electron traps. 2. Characteristics of OSLDfull in terms of the fading and regeneration of luminescence conditions has similar tendency with those of OSLDempty. Fading time suggested in our study is 10 min and maximum doses increased by regeneration of luminescence in dark room is 7.3 cGy after 13 days. 3. OSLDfull has dose linearity for exposed doses over a range of 10 Gy. 4. The dose sensitivity of OSLDfull can be stable under the optimal bleaching conditions.
S.-Y. Park et al. / Physica Medica 38 (2017) 81–87
Fig. 9. Flowchart for the setup of a highly stable OSLDfull system for use in radiation therapy.
5. Conclusions We investigated the characteristics of OSLDs with fully filled deep electron/hole traps (OSLDfull) and found the optimal bleaching conditions for OSL dosimetry by considering the effects of the bleaching time and the bleaching wavelength on the dose sensitivity and linearity of OSLDfull. Clinical dosimetry with OSLDfull having fully filled deep traps under appropriate bleaching conditions was shown to be highly stable and accurate, with no change in either the dose sensitivity or linearity. Acknowledgements This work was supported by a National Research Foundation of Korea (NRF) grant, funded by the Korea government (MISP) (No. 2014M2B2A4031164), and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI14C3459). The authors would like to thank the following groups and individuals for their assistance: Korea Multi-Purpose Accelerator Complex (KOMAC) for initiating this work, Yoonseok Ka in the Nano Science and Technology program at Seoul National University for measuring the wavelength spectrum of the bleaching source, Hanil Nuclear for providing the bleaching device, Nascimento Luana Freitas in SCKCEN Belgian Nuclear Research Centre for reviewing this manuscript carefully and Hyunseok Lee in Korea Institute of Nuclear Safety and Hyunseok Kim in Convergence Science and Technology at Seoul National University for giving assistance in this study. References [1] McKeever SW, Moscovitch M. On the advantages and disadvantages of optically stimulated luminescence dosimetry and thermoluminescence dosimetry. Radiat Prot Dosimetry 2003;104:263–70. [2] Kerns JR, Kry SF, Sahoo N, Followill DS, Ibbott GS. Angular dependence of the nanoDot OSL dosimeter. Med Phys. 2011;38:3955–62.
87
[3] Jursinic PA. Characterization of optically stimulated luminescent dosimeters, OSLDs, for clinical dosimetric measurements. Med Phys. 2007;34:4594–604. [4] Jursinic PA. Angular dependence of dose sensitivity of nanoDot optically stimulated luminescent dosimeters in different radiation geometries. Med Phys. 2015;42:5633–41. [5] Lehmann J, Dunn L, Lye JE, Kenny JW, Alves AD, Cole A, et al. Angular dependence of the response of the nanoDot OSLD system for measurements at depth in clinical megavoltage beams. Med Phys. 2014;41:061712. [6] Lee SY, Lee KJ. Development of a personal dosimetry system based on optically stimulated luminescence of alpha-Al2O3: C for mixed radiation fields. Appl Radiat Isot 2001;54:675–85. [7] Mrcela I, Bokulie T, Izewska J, Budanec M, Frobe A, Kusic Z. Optically stimulated luminescence in vivo dosimetry for radiotherapy: physical characterization and clinical measurements in Co-60 beams. Phys Med Biol 2011;56:6065–82. [8] McKeever SWS, Akselrod MS, Colyott LE, Larsen NA, Polf JC, Whitley V. Characterisation of Al2O3 for use in thermally and optically stimulated luminescence dosimetry. Radiat Prot Dosimetry 1999;84:163–8. [9] Lye J, Dunn L, Kenny J, Lehmann J, Kron T, Oliver C, et al. Remote auditing of radiotherapy facilities using optically stimulated luminescence dosimeters. Med Phys. 2014;41. [10] Casey KE, Alvarez P, Kry SF, Howell RM, Lawyer A, Followill D. Development and implementation of a remote audit tool for high dose rate (HDR) Ir-192 brachytherapy using optically stimulated luminescence dosimetry. Med Phys. 2013;40. [11] Casey K, Alvarez P, Lawyer A, Kry S, Howell R, Davidson S, et al. Development and implementation of a remote audit tool for high dose rate (HDR) Ir-192 brachytherapy using optically stimulated luminescence dosimetry. Med Phys 2012;39:3721. [12] Jursinic PA. Changes in optically stimulated luminescent dosimeter (OSLD) dosimetric characteristics with accumulated dose. Med Phys. 2010;37:132–40. [13] Yukihara EG, McKeever SW. Optically stimulated luminescence (OSL) dosimetry in medicine. Phys Med Biol 2008;53:R351–79. [14] Yukihara EG, Whitley VH, Polf JC, Klein DM, McKeever SWS, Akselrod AE, et al. The effects of deep trap population on the thermoluminescence of Al2O3: C. Radiat Meas 2003;37:627–38. [15] Omotayo AA, Cygler JE, Sawakuchi GO. The effect of different bleaching wavelengths on the sensitivity of Al2O3: C optically stimulated luminescence detectors (OSLDs) exposed to 6 MV photon beams. Med Phys. 2012;39:5457–68. [16] Umisedo NK, Yoshimura EM, Gasparian PBR, Yukihara EG. Comparison between blue and green stimulated luminescence of Al2O3:C. Radiat Meas 2010;45:151–6. [17] Markey BG, Colyott LE, Mckeever SWS. Time-resolved optically stimulated luminescence from Alpha-Al2O3-C. Radiat Meas 1995;24:457–63. [18] Edmund JM, Andersen CE, Marckmann CJ, Aznar MC, Akselrod MS, BotterJensen L. CW-OSL measurement protocols using optical fibre Al2O3: C dosemeters. Radiat Prot Dosimetry 2006;119:368–74. [19] Yukihara EG, Gaza R, McKeever SWS, Soares CG. Optically stimulated luminescence and thermoluminescence efficiencies for high-energy heavy charged particle irradiation in Al2O3: C. Radiat Meas 2004;38:59–70. [20] Han SC, Choi SH, Park S, Kim CH, Jung H, Kim M, et al. Evaluation of dosimetric characteristics of reproducibility, linearity and dose dependence of optically stimulated luminescence dosimeters in Co-60 gamma-rays. Pro Med Phys 2014;25:31–6. [21] Yu M, Jang HS, Jeon DM, Cheon GS, Lee HC, Chung MJ, et al. Dosimetric evaluation of tomotherapy and four-box field conformal radiotherapy in locally advanced rectal cancer. Radiat Oncol J 2013;31:252–9. [22] Almond PR, Biggs PJ, Coursey BM, Hanson WF, Huq MS, Nath R, et al. AAPM’s TG-51 protocol for clinical reference dosimetry of high-energy photon and electron beams. Med Phys 1999;26:1847–70. [23] Scarboro SB, Cody D, Alvarez P, Followill D, Court L, Stingo FC, et al. Characterization of the nanoDot OSLD dosimeter in CT. Med Phys 2015;42:1797–807. [24] Dunn L, Lye J, Kenny J, Lehmann J, Williams I, Kron T. Commissioning of optically stimulated luminescence dosimeters for use in radiotherapy. Radiat Meas 2013;51–52:31–9. [25] Yukihara EG, Gasparian PBR, Sawakuchi GO, Ruan C, Ahmad S, Kalavagunta C, et al. Medical applications of optically stimulated luminescence dosimeters (OSLDs). Radiat Meas 2010;45:658–62. [26] Reft CS. The energy dependence and dose response of a commercial optically stimulated luminescent detector for kilovoltage photon, megavoltage photon, and electron, proton, and carbon beams. Med Phys 2009;36:1690–9. [27] BotterJensen L, Larsen NA, Markey BG, McKeever SWS. Al2O3: C as a sensitive OSL dosemeter for rapid assessment of environmental photon dose rates. Radiat Meas 1997;27:295–8. [28] Yukihara EG, Yoshimura EM, Lindstrom TD, Ahmad S, Taylor KK, Mardirossian G. High-precision dosimetry for radiotherapy using the optically stimulated luminescence technique and thin Al2O3: C dosimeters. Phys Med Biol 2005;50:5619–28. [29] Polf JC, Yukihara EG, Akselrod MS, McKeever SWS. Real-time luminescence from Al2O3 fiber dosimeters. Radiat Meas 2004;38:227–40.