- Email: [email protected]
Online neutron fluence measurement at University Hospital Essen neutron therapy facility using gallium arsenide LEDs
Radiation Measurements 46 (2011) 1761e1764
Contents lists available at ScienceDirect
Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas
Online neutron fluence measurement at University Hospital Essen neutron therapy facility using gallium arsenide LEDs R. Hentschel a, *, B. Mukherjee b, J. Lambert b, W. Deya a, J. Farr b a b
Strahlenklinik, University Hospital Essen, Hufelandstrasse 55, 45122 Essen, Germany Westdeutsches Protonentherapiezentrum Essen (WPE) gGmbH, Hufelandstrasse 55, 45122 Essen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 October 2010 Accepted 26 May 2011
The detector and sensor group of the West German Proton Therapy Centre (WPE) has developed a novel real-time neutron fluence monitor based on tiny, inexpensive, commercially available GaAs-LEDs. The linear detection range for d(14)þBe neutrons was evaluated to be 5.0 108e2.0 1011 neutron.cm2. However, this monitor can be used universally for neutrons of any energy distribution. Using scaling factors, fluence calibration curves for 1 MeV and 14 MeV DþT fusion neutrons have been calculated. The sensitivity of the detector increases with increasing neutron energy. This makes it suitable for the detection of high-energy neutrons, providing an extra advantage for use at a proton therapy facility where there is a high proportion of high-energy neutrons. The detector is practically not sensitive to photons. A prototype of the online GaAs-LED based neutron fluence monitor has been tested successfully at University Hospital Essen neutron therapy facility and will be implemented at WPE in the near future. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Neutron fluence Fast neutron Proton therapy GaAs LED
1. Introduction The West German Proton Therapy Centre Essen (WPE) is expected to start clinical operation in 2011. Its detector and sensor group is developing devices which are not available commercially, however could be necessary or desirable for the determination of all aspects connected with the operation of a proton therapy facility. In that framework also an online neutron fluence monitor has been developed. The neutron fluence monitor has been tested at University Hospital Essen neutron therapy facility. Since 1972, the Strahlenklinik of the University Hospital Essen is operating this neutron therapy facility based on a TCC CV28 medical cyclotron. Neutrons of an average energy of 6 MeV are produced via [d(14)þBe] reaction by bombarding a thick beryllium target with 14 MeV deuterons (Rassow et al., 1978). The patient treatment at the facility has now been laid off. However, the neutron facility is further used for radiobiology experiments and collaborative research projects with the detector and sensor group of the neighbouring West German Proton Therapy Centre Essen
* Corresponding author. Tel.: þ49 201 7234184; fax: þ49 201 7234197. E-mail address: [email protected] (R. Hentschel). 1350-4487/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.radmeas.2011.05.062
(WPE) in the fields of radiation measurement, dosimetry and instrumentation. 2. The GaAs-LED based neutron fluence monitor developed by detector and sensor group Fast neutrons interact with semiconductor materials dislodging the lattice atoms from their original stable positions, thereby creating vacancies. The combinations of vacancies and interstitial atoms are known as Frenkel-pairs. Displacement damage caused due to non-ionising-energy-loss (NIEL) of fast neutrons in bulk Gallium Arsenide (GaAs) light emitting diodes (LED) results in the reduction of their light output (Bates et al., 1997). The neutron energy dissipation in the LED is an explicit function of the kerma in GaAs and independent of neutron energy spectrum (Williams et al., 1994). Accompanying photons have practically no influence on the light output (Mukherjee et al., 2007). We have used this phenomenon to construct a fluence monitor for high-energy neutrons. Usually, single LEDs with known light emission are irradiated in a neutron field and the reduction of the light emission is determined offline after the irradiation. We developed an online monitor enabling the light reduction be seen immediately. The circuit diagram of the monitor is given in Fig. 1. A bulk yellow GaAs-LED LED2 (Type: LN48YPX, Manufacturer: Panasonics Corporation,
1762
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
Fig. 1. Circuit diagram of the GaAs-LED based neutron monitor.
Japan) is the sensitive element of the neutron monitor. The sensor diode LED2 is connected to a constant current supply made with the fixed-voltage regulator VC1. The diode light output is fed to the light-dependent-resistor LDR2 via an optical fibre. The resistance of LDR2 is compared to that of an identical light-dependent-resistor LDR4 in a Wheatstone bridge. LDR4 is irradiated by light emitted from diode LED4, in such a way that the Wheatstone bridge is balanced. Therefore, the light output of LED4 can be regulated by voltage controller VC2. Diodes LED1 and LED3 signalise the correct function of LED2 and LED4 respectively. The measuring device M is a common digital multimeter with a resolution of 1 mV and datalogging function. The monitor is supplied by an AC adapter PS delivering 12 V DC.
Fig. 3. Monte Carlo calculated neutron spectrum from [d(14)þBe] reaction (a), in addition line spectra of 1 MeV neutrons (b) and 14 MeV DþT fusion neutrons (c).
Fig. 2. Schematic setup of the real-time neutron fluence monitoring experiment.
Fig. 4. Warm-up behaviour of the GaAs-LED based neutron fluence monitor.
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
Fig. 5. Output R from two 150 neutron irradiations with a break (no beam) between them.
1763
(RSC). The detector light output was fed to the Wheatstone bridge (WB) of the neutron fluence monitor via a 1 mm diameter optical PMMA fibre cable (OFC). The electronic equipment consisting of the neutron fluence monitor and the digital multimeter DMM including data logger was located behind the radiation shielding of the therapy vault. The LND was placed at the isocentre of the rotatable therapy gantry, 125 cm from the beryllium target T. The beryllium target T was bombarded with a 14 MeV deuteron beam with an intensity of 50 0.5 mA from the TCC CV28 medical cyclotron to produce neutrons with a fluence rate of 1.2 108 neutron.cm2.s1 at the location of LND. The fluence rate and the neutron spectrum at that location, see Fig. 3, has been calculated by simulation of the neutron therapy gantry (Morand et al., 2009) using MCNPX V2.6.0 Monte Carlo code (Pelowitz, 2008). The neutron spectrum is characterized by a broad Gaussian distribution with a mean energy of 6.0 MeV. The light output of the GaAs-LED was measured while it was lit with a constant forward current of 9.34 0.01 mA. Fig. 4 demonstrates the warm-up behaviour of the neutron fluence monitor. The monitor needs a warm-up time of about 1 h in order to ensure sufficient output stability. After that time the fluctuations do not exceed 30 mV. Fig. 5 is a plot of two neutron irradiations with a 15 min break between them. Reduction of LED light gives a positive output signal in our case. An irradiation of the LED with more than 1.0 108 neutron.cm2 gives enough output voltage to be distinguished from accidental fluctuations. After stopping the neutron irradiations, some annealing of the LED chip and reconstruction of light yield may be observed. 4. Results and data analysis
Fig. 6. The differential neutron kerma coefficient of GaAs (kGaAs) is plotted as a function of neutron energy (En); the fitting polynomial is shown in the inset.
3. Test of the neutron fluence monitor at Essen cyclotron based fast neutron generator The experimental setup for the calibration of the real-time neutron fluence monitor is depicted in Fig. 2. The sensitive GaAsLED (LND) was connected to the constant current supply of the monitor (CCS) using a standard twisted pair RF-shielded cable
The differential neutron kerma coefficient of GaAs is shown in Fig. 6 (Ougouag et al., 1990). The accumulated neutron fluence as a function of the measured output signal R is plotted in Fig. 7. The GaAs-LED based neutron fluence monitor has a linear response within the fluence range of 5.0 108e2.0 1011 neutron.cm2. Additionally, we have rescaled the experimentally estimated calibration curve for d(14)þBe neutrons for 1 MeV and 14 MeV DþT fusion neutrons (Fig. 7). The fluence calibration factors for 1 MeV and 14 MeV neutrons were calculated to be 12.2 and 0.33 respectively.
Fig. 7. The neutron fluence calibration curves for d(14)þBe neutrons (a) and rescaled for 1 MeV (b) and 14 MeV (c) neutrons. Note that the lowest detectable neutron fluence level drops with increasing neutron energy.
1764
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
5. Conclusion
References
We have developed a novel real-time neutron fluence monitor based on tiny, inexpensive, commercially available GaAs-LEDs. The linear detection range for d(14)þBe neutrons was evaluated to be 5.0 108e2.0 1011 neutron.cm2. Using scaling factors we have calculated fluence calibration curves for 1 MeV and 14 MeV DþT fusion neutrons. This method can be universally used for neutrons of any energy distribution. The sensitivity of the detector increases with increasing neutron energy. This makes it suitable for the detection of high-energy neutrons, providing an extra advantage for use at a proton therapy facility where there is a high proportion of high-energy neutrons. The GaAs-LED neutron fluence monitor operates in integral mode, making it suitable for use in pulsed neutron radiation. Hence, it could be used as an emergency dosimeter at a high-energy particle accelerator environment to detect intense, short pulses of neutrons during beam spills.
Bates, R., Da Via, C., O’Shea, A., Pickford, A., Raine, C., Smith, K.,1997. Radiation induced damage in GaAs particle detectors. IEEE. Trans. Nucl. Sci. 44, 1705e1707. Mukherjee, B., Simrock, S., Khachan, J., Rybka, D., Romaniuk, R., 2007. Application of low-cost gallium arsenide light-emitting-diodes as kerma dosemeters and fluence monitor for high-energy neutrons. Radiat. Prot. Dosimetry. 126, 256e261. Morand, J., Hentschel, R., Wittig, A., Moss, R., Hachem, S., Liu, Y.H., Sauerwein, W., 2009. Analytical description of the d(14)þBe neutron beam source at the Essen fast neutron therapy facility. Nucl. Technol. 168, 456e461. Ougouag, A.M., Williams, J.G., Danjaji, M.B., Yang, S.Y., Meason, J.L., 1990. Differential displacement kerma cross sections for neutron interactions in Si and GaAs. IEEE. Trans. Nucl. Sci. 37, 2219e2227. Pelowitz, D.P. (Ed.), 2008. MCNPX User’s Manual. Version 2.6.0, LA-CP-07e1473. Rassow, J., Huedepohl, G., Maier, E., Meissner, P., 1978. CIRCE: Cyclotron isocentric neutron therapy facility, Radiological Centre Essen, in: Burger, G., Ebert, H.G. (Eds.), Proceedings of the Third Symposium on Neutron Dosimetry in Biology and Medicine, EUR 5848, Commission of the European Communities, Luxembourg, pp. 327e337. Williams, J.G., Griffin, P.J., Kelly, J.G., Figueroa, T., 1994. Estimation of 1-MeV equivalent neutron fluence from dosimetry responses without spectrum unfolding. IEEE. Trans. Nucl. Sci. 41, 2147e2151.
Contents lists available at ScienceDirect
Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas
Online neutron fluence measurement at University Hospital Essen neutron therapy facility using gallium arsenide LEDs R. Hentschel a, *, B. Mukherjee b, J. Lambert b, W. Deya a, J. Farr b a b
Strahlenklinik, University Hospital Essen, Hufelandstrasse 55, 45122 Essen, Germany Westdeutsches Protonentherapiezentrum Essen (WPE) gGmbH, Hufelandstrasse 55, 45122 Essen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 October 2010 Accepted 26 May 2011
The detector and sensor group of the West German Proton Therapy Centre (WPE) has developed a novel real-time neutron fluence monitor based on tiny, inexpensive, commercially available GaAs-LEDs. The linear detection range for d(14)þBe neutrons was evaluated to be 5.0 108e2.0 1011 neutron.cm2. However, this monitor can be used universally for neutrons of any energy distribution. Using scaling factors, fluence calibration curves for 1 MeV and 14 MeV DþT fusion neutrons have been calculated. The sensitivity of the detector increases with increasing neutron energy. This makes it suitable for the detection of high-energy neutrons, providing an extra advantage for use at a proton therapy facility where there is a high proportion of high-energy neutrons. The detector is practically not sensitive to photons. A prototype of the online GaAs-LED based neutron fluence monitor has been tested successfully at University Hospital Essen neutron therapy facility and will be implemented at WPE in the near future. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Neutron fluence Fast neutron Proton therapy GaAs LED
1. Introduction The West German Proton Therapy Centre Essen (WPE) is expected to start clinical operation in 2011. Its detector and sensor group is developing devices which are not available commercially, however could be necessary or desirable for the determination of all aspects connected with the operation of a proton therapy facility. In that framework also an online neutron fluence monitor has been developed. The neutron fluence monitor has been tested at University Hospital Essen neutron therapy facility. Since 1972, the Strahlenklinik of the University Hospital Essen is operating this neutron therapy facility based on a TCC CV28 medical cyclotron. Neutrons of an average energy of 6 MeV are produced via [d(14)þBe] reaction by bombarding a thick beryllium target with 14 MeV deuterons (Rassow et al., 1978). The patient treatment at the facility has now been laid off. However, the neutron facility is further used for radiobiology experiments and collaborative research projects with the detector and sensor group of the neighbouring West German Proton Therapy Centre Essen
* Corresponding author. Tel.: þ49 201 7234184; fax: þ49 201 7234197. E-mail address: [email protected] (R. Hentschel). 1350-4487/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.radmeas.2011.05.062
(WPE) in the fields of radiation measurement, dosimetry and instrumentation. 2. The GaAs-LED based neutron fluence monitor developed by detector and sensor group Fast neutrons interact with semiconductor materials dislodging the lattice atoms from their original stable positions, thereby creating vacancies. The combinations of vacancies and interstitial atoms are known as Frenkel-pairs. Displacement damage caused due to non-ionising-energy-loss (NIEL) of fast neutrons in bulk Gallium Arsenide (GaAs) light emitting diodes (LED) results in the reduction of their light output (Bates et al., 1997). The neutron energy dissipation in the LED is an explicit function of the kerma in GaAs and independent of neutron energy spectrum (Williams et al., 1994). Accompanying photons have practically no influence on the light output (Mukherjee et al., 2007). We have used this phenomenon to construct a fluence monitor for high-energy neutrons. Usually, single LEDs with known light emission are irradiated in a neutron field and the reduction of the light emission is determined offline after the irradiation. We developed an online monitor enabling the light reduction be seen immediately. The circuit diagram of the monitor is given in Fig. 1. A bulk yellow GaAs-LED LED2 (Type: LN48YPX, Manufacturer: Panasonics Corporation,
1762
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
Fig. 1. Circuit diagram of the GaAs-LED based neutron monitor.
Japan) is the sensitive element of the neutron monitor. The sensor diode LED2 is connected to a constant current supply made with the fixed-voltage regulator VC1. The diode light output is fed to the light-dependent-resistor LDR2 via an optical fibre. The resistance of LDR2 is compared to that of an identical light-dependent-resistor LDR4 in a Wheatstone bridge. LDR4 is irradiated by light emitted from diode LED4, in such a way that the Wheatstone bridge is balanced. Therefore, the light output of LED4 can be regulated by voltage controller VC2. Diodes LED1 and LED3 signalise the correct function of LED2 and LED4 respectively. The measuring device M is a common digital multimeter with a resolution of 1 mV and datalogging function. The monitor is supplied by an AC adapter PS delivering 12 V DC.
Fig. 3. Monte Carlo calculated neutron spectrum from [d(14)þBe] reaction (a), in addition line spectra of 1 MeV neutrons (b) and 14 MeV DþT fusion neutrons (c).
Fig. 2. Schematic setup of the real-time neutron fluence monitoring experiment.
Fig. 4. Warm-up behaviour of the GaAs-LED based neutron fluence monitor.
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
Fig. 5. Output R from two 150 neutron irradiations with a break (no beam) between them.
1763
(RSC). The detector light output was fed to the Wheatstone bridge (WB) of the neutron fluence monitor via a 1 mm diameter optical PMMA fibre cable (OFC). The electronic equipment consisting of the neutron fluence monitor and the digital multimeter DMM including data logger was located behind the radiation shielding of the therapy vault. The LND was placed at the isocentre of the rotatable therapy gantry, 125 cm from the beryllium target T. The beryllium target T was bombarded with a 14 MeV deuteron beam with an intensity of 50 0.5 mA from the TCC CV28 medical cyclotron to produce neutrons with a fluence rate of 1.2 108 neutron.cm2.s1 at the location of LND. The fluence rate and the neutron spectrum at that location, see Fig. 3, has been calculated by simulation of the neutron therapy gantry (Morand et al., 2009) using MCNPX V2.6.0 Monte Carlo code (Pelowitz, 2008). The neutron spectrum is characterized by a broad Gaussian distribution with a mean energy of 6.0 MeV. The light output of the GaAs-LED was measured while it was lit with a constant forward current of 9.34 0.01 mA. Fig. 4 demonstrates the warm-up behaviour of the neutron fluence monitor. The monitor needs a warm-up time of about 1 h in order to ensure sufficient output stability. After that time the fluctuations do not exceed 30 mV. Fig. 5 is a plot of two neutron irradiations with a 15 min break between them. Reduction of LED light gives a positive output signal in our case. An irradiation of the LED with more than 1.0 108 neutron.cm2 gives enough output voltage to be distinguished from accidental fluctuations. After stopping the neutron irradiations, some annealing of the LED chip and reconstruction of light yield may be observed. 4. Results and data analysis
Fig. 6. The differential neutron kerma coefficient of GaAs (kGaAs) is plotted as a function of neutron energy (En); the fitting polynomial is shown in the inset.
3. Test of the neutron fluence monitor at Essen cyclotron based fast neutron generator The experimental setup for the calibration of the real-time neutron fluence monitor is depicted in Fig. 2. The sensitive GaAsLED (LND) was connected to the constant current supply of the monitor (CCS) using a standard twisted pair RF-shielded cable
The differential neutron kerma coefficient of GaAs is shown in Fig. 6 (Ougouag et al., 1990). The accumulated neutron fluence as a function of the measured output signal R is plotted in Fig. 7. The GaAs-LED based neutron fluence monitor has a linear response within the fluence range of 5.0 108e2.0 1011 neutron.cm2. Additionally, we have rescaled the experimentally estimated calibration curve for d(14)þBe neutrons for 1 MeV and 14 MeV DþT fusion neutrons (Fig. 7). The fluence calibration factors for 1 MeV and 14 MeV neutrons were calculated to be 12.2 and 0.33 respectively.
Fig. 7. The neutron fluence calibration curves for d(14)þBe neutrons (a) and rescaled for 1 MeV (b) and 14 MeV (c) neutrons. Note that the lowest detectable neutron fluence level drops with increasing neutron energy.
1764
R. Hentschel et al. / Radiation Measurements 46 (2011) 1761e1764
5. Conclusion
References
We have developed a novel real-time neutron fluence monitor based on tiny, inexpensive, commercially available GaAs-LEDs. The linear detection range for d(14)þBe neutrons was evaluated to be 5.0 108e2.0 1011 neutron.cm2. Using scaling factors we have calculated fluence calibration curves for 1 MeV and 14 MeV DþT fusion neutrons. This method can be universally used for neutrons of any energy distribution. The sensitivity of the detector increases with increasing neutron energy. This makes it suitable for the detection of high-energy neutrons, providing an extra advantage for use at a proton therapy facility where there is a high proportion of high-energy neutrons. The GaAs-LED neutron fluence monitor operates in integral mode, making it suitable for use in pulsed neutron radiation. Hence, it could be used as an emergency dosimeter at a high-energy particle accelerator environment to detect intense, short pulses of neutrons during beam spills.
Bates, R., Da Via, C., O’Shea, A., Pickford, A., Raine, C., Smith, K.,1997. Radiation induced damage in GaAs particle detectors. IEEE. Trans. Nucl. Sci. 44, 1705e1707. Mukherjee, B., Simrock, S., Khachan, J., Rybka, D., Romaniuk, R., 2007. Application of low-cost gallium arsenide light-emitting-diodes as kerma dosemeters and fluence monitor for high-energy neutrons. Radiat. Prot. Dosimetry. 126, 256e261. Morand, J., Hentschel, R., Wittig, A., Moss, R., Hachem, S., Liu, Y.H., Sauerwein, W., 2009. Analytical description of the d(14)þBe neutron beam source at the Essen fast neutron therapy facility. Nucl. Technol. 168, 456e461. Ougouag, A.M., Williams, J.G., Danjaji, M.B., Yang, S.Y., Meason, J.L., 1990. Differential displacement kerma cross sections for neutron interactions in Si and GaAs. IEEE. Trans. Nucl. Sci. 37, 2219e2227. Pelowitz, D.P. (Ed.), 2008. MCNPX User’s Manual. Version 2.6.0, LA-CP-07e1473. Rassow, J., Huedepohl, G., Maier, E., Meissner, P., 1978. CIRCE: Cyclotron isocentric neutron therapy facility, Radiological Centre Essen, in: Burger, G., Ebert, H.G. (Eds.), Proceedings of the Third Symposium on Neutron Dosimetry in Biology and Medicine, EUR 5848, Commission of the European Communities, Luxembourg, pp. 327e337. Williams, J.G., Griffin, P.J., Kelly, J.G., Figueroa, T., 1994. Estimation of 1-MeV equivalent neutron fluence from dosimetry responses without spectrum unfolding. IEEE. Trans. Nucl. Sci. 41, 2147e2151.