- Email: [email protected]
Determination of the energy response of closely tissue-equivalent diamond dosimeters for radiotherapy dosimetry
Applied Radiation and Isotopes 71 (2012) 23–24
Contents lists available at SciVerse ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Determination of the energy response of closely tissue-equivalent diamond dosimeters for radiotherapy dosimetry R.P. Hugtenburg a,b,n, F.M. Saeedi c,d, A.E.R. Baker c a
College of Medicine, Swansea University, Singleton Park, Swansea SA2 8PP, UK Department of Medical Physics and Clinical Engineering, Singleton Hospital, Abertawe Bro Morgannwg UHB, Swansea, Wales SA2 8QA, UK c Department of Physics and Astronomy, University of Birmingham, Birmingham B15 2TT, UK d Oncology Center, National Guard Hospital-Jeddah, Saudi Arabia b
a r t i c l e i n f o
abstract
Available online 28 March 2012
A monoenergetic X-ray syncrotron source was used to determine the energy response of a diamond detector in the range 5–25 keV, clarifying the elemental composition of the detector. The response is shown to be influenced by the detector housing and electrical contacts. A model for the energy response of the detector is determined that is valid in the 5 keV–15 MeV with an accuracy of 5% and therefore can be used to correct the dosimeter response to low-energy and scattered radiation. Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved.
Keywords: Diamond detector X-ray synchrotron Energy response
1. Introduction Solid-state detectors are increasingly used for dosimetry purposes in megavoltage photon radiotherapy, in order to avoid issues relating to the use of tissue-equivalent ionisation chambers, including a lack of electronic equilibrium (Saeedi et al., 2004) and poor sensitivity in small radiation fields, used in stereotactic radiosurgery and intensity modulated radiotherapy. The sensitivity of solid-state dosimeters is generally excellent, however many dosimeters, including silicon diodes, depart significantly from tissue-equivalence, particular in their sensitivity to low-energy and scattered X-rays (Yin et al., 2002, 2004). In recent years liquid-filled ionisation chambers and diamond-based detectors have been considered as closely tissue-equivalent detectors that do not suffer in the absence of electronic equilibrium and therefore can be used to measure dose in small fields and very close to the surface of a phantom. In this work we report on the tissue-equivalence of a commercially available diamond dosimeter based on natural diamonds. Measurements utilising a monoenergetic source are compared with the response recorded in previous work with quasi-monoenergetic photon sources and Monte Carlo methods, in order to develop models for the energy dependence of the detector in the range 0.005–15 MeV. It has been shown in previous work (Yin et al., 2004) that corrections to the diamond detector dose response are needed to achieve high orders of dosimetric accuracy in the kilovoltage
n Corresponding author at: Department of Medical Physics and Clinical Engineering, Singleton Hospital, Abertawe Bro Morgannwg UHB, Swansea, Wales SA2 8QA, UK. E-mail address: [email protected] (R.P. Hugtenburg).
range, particularly below 100 keV, where the photoelectric effect predominates. The response is strongly influenced by the lower effective-Z of carbon relative to water and the presence of metallic electrical contacts. The dosimetry of X-ray synchrotron beams has recently been considered for several other detectors with good tissue-equivalence, including radiochromic film (Nariyama, 2011) and 3D gels (Abdul Rahman et al., 2010).
2. Methods A natural diamond-based detector (PTW Freiburg/ IPTP Dubna, serial number 7017) was compared with a parallel-plate ionisation chamber (PTW type 23342) in a monoenergetic synchrotron X-ray beam, with both detectors embedded in RW3 Solid-Water (PTW). The diamond detector is calibrated by the manufacturer using a 60 Co gamma-ray source. Measurement of dose with the detector has been shown to be highly stable and reproducible. Diamond detectors typically have a small dose-rate non-linearity. Corrections were made for a slightly sub-linear dose-rate response in the detector with an exponent, D ¼ 0:975. In previous experimental studies of the diamond detector, response to therapeutic X-rays of energy 45 kVp and 100 kVp from a superficial X-ray unit (Hugtenburg et al., 2001) and 6 MV and 15 MV X-rays from a therapeutic linear accelerator have been measured. It is generally observed that the energy dependence of the diamond detector is slowly varying above 100 keV and below 10 MeV, where the Compton scattering process predominates. The parallel-plate ionisation chamber (PTW type 23342) is designed for the dosimetry of superficial X-rays and is described by the manufacturer as having tissue-equivalence of better than 2% in the range 8–32 keV.
0969-8043/$ - see front matter Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apradiso.2012.03.016
24
R.P. Hugtenburg et al. / Applied Radiation and Isotopes 71 (2012) 23–24
Table 1 Fractional composition (by weight) of the diamond detector used in Eq. (1). Element
C
Cu
Ag
Au
Fraction
0.997
0.0005
0.001
0.0015
Fig. 1. Experimental data and modelling of the response the diamond detector to the parallel-plate ionisation chamber. Data from Yin et al. (2004) is merged with current data measured on an X-ray synchrotron. The solid lines show models based on attenuation and absorption for the composite material detector defined in table 1 as well as a pure carbon detector. The measured data is repeated in the inset figure, where the experimental error is comparable to the symbol size.
X-ray synchrotron station 16.3 at Daresbury laboratories (Collins et al., 1998) was used to provide monoenergetic photons in the energy range from 5 to 30 keV. The second generation Daresbury (SRS) X-ray synchrotron produced highly monoenergetic X-rays of comparable intensity to therapeutic beams, through the use of a Si monochromator (DE=E o103 ). The X-ray synchrotron beam profile was narrower than the detectors in the vertical direction, so a motorised stage was used to move the detectors slowly across the beam, collimated to 8 mm 0.1 mm to generate an 8 mm 8 mm field. The field uniformity was shown to be of the order of 5%, determined with radiochromic film (MD-55-2, GafChromic film). A model for the detector response, based on linear combinations of the mass-absorption coefficient of the detector constituents and attenuation of the beam in the detector and upstream housing, is proposed, as demonstrated in the following equation: R¼
ðmen =rÞdd ðmtÞpe ðmtÞdd e e ðmen =rÞwater
in previous experimental studies with a quasi-monoenergetic X-ray set (Yin et al., 2004), and utilising the 60Co dose calibration provided by the manufacturer. Agreement of typically better than 5%, between the experiment and the model, is achieved over this range, based on the given constituent composition of the detector. The dose response in regions far from the atomic orbital edges is shown to be reproduced well by the model described in Eq. (1). The largest departures from the model occur in the energy range below 25 keV, in which two effects predominate; namely the increased divergence of the absorption cross-sections of carbon and oxygen, and rapid changes in absorption in the vicinity of the K-edge of Cu (at 9.0 keV), the K-edge of Ag (25.5 keV) and the L-edges of Au (14.3, 13.7 and 11.9 keV). Energy deposition in the ionisation chamber was modelled as an air-equivalent material. The figure inset highlights the new experimental data measured with the diamond detector. Previous work (Hugtenburg et al., 2007) has shown that modelling in the vicinity of K-edges can be significantly in error ( 4 5%) due to large variations in the cross-section. The effects were shown to be mitigated by choosing a finer grid for the interpolation of cross-sections in the case of Monte Carlo modelling. Use of a more sophisticated Monte Carlo model of the detector, which includes geometrically precise models for the constituent components of the diamond detector, has been considered, with some improvement to the reproduction of the response in the vicinity of K- and L-edges observed. The ionisation chamber has not been modelled in detail and instead is considered as a wateror air-equivalent homogeneous material, air-equivalence proving to offer better agreement with the response of the ionisation chamber.
4. Conclusion The energy response of a diamond detector is examined with a monoenergetic X-ray synchrotron source, enabling the consideration of the influence of the constituent metallic elements, Cu, Ag and Au. The experimental data extends and clarifies issues relating to the modelling of the energy response of diamond detectors to a wider range of energies. The model of the energy response can be used to correct for distortions to the detector response for X-ray therapeutic beams at large field sizes, relating to changes in the response of the detector to scattered photons. References
ð1Þ
where the ‘dd’ subscript refers to coefficients and thickness associated with the diamond detector composite material established in this work, while the ‘pe’ subscript refers to the polyethylene housing of the diamond. The thickness of the polyethylene housing upstream of the detector was, t pe ¼ 1 mm, the thickness of the diamond detector to the effective point of measurement was t dd ¼ 0:25 mm. The attenuation and massabsorption coefficients for the diamond detector ðmen =rÞdd were determined from a combination of the four constituent elements, as shown in Table 1. The detector dimensions and constituents were identified using a combination of methods, including scanning of the detector with X-ray fluorescence, radiographic studies and manufacturer descriptions.
3. Results Fig. 1 shows the energy dependence of the diamond detector in the 5–205 keV, combining the present data with data recorded
Abdul Rahman, A.T., Bruer-Krisch, E., Brochard, T., Adamovics, J., Bradley, D., Doran, S., 2010. Verification of synchrotron microbeam radiation therapy using a purpose-built optical CT microscope. J. Phys.: Conf. Ser. 250, 012083. Collins, S.P., Cernik, R.J., Fell, B., Tang, C.C., Harris, N.W., Miller, M.C., Oszlanyi, G., 1998. Station 16.3: a high-resolution single-crystal diffraction facility at the SRS. J. Synchrotron Rad. 5, 1263–1269. Hugtenburg, R.P., Johnston, K., Chalmers, G.J., Beddoe, A.H., 2001. Application of diamond detectors to the dosimetry of 45 and 100 kVp therapy beams: comparison with a parallel-plate ionization chamber and Monte Carlo. Phys. Med. Biol. 46, 2489–2501. Hugtenburg, R.P., Saeedi, F., Bradley, D.A., 2007. Modeling dose response to synchrotron X-rays in solid-state and biological systems. Radiat. Eff. Defects Solids 162, 765–769. Nariyama, N., 2011. Macroscopic and microscopic X-ray dose mappings with CCDbased GafChromic film dosimeters. Applied Radiation and Isotopes, http://dx. doi.org/10.1016/j.apradiso.2011.12.009, in press. Saeedi, F., Yin, Z., Hugtenburg, R.P., 2004. Dosimetry in conditions of electron disequilibrium. Radiat. Phys. Chem. 71, 979–980. Yin, Z., Hugtenburg, R.P., Beddoe, A.H., 2002. Response of silicon diode dosemeters to scattered radiation from megavoltage photon beams. Radiat. Prot. Dosimetry 101, 415–418. Yin, Z., Hugtenburg, R.P., Beddoe, A.H., 2004. Response corrections for solid-state detectors in megavoltage photon dosimetry. Phys. Med. Biol. 49, 3691–3702. Yin, Z., Hugtenburg, R.P., Green, S., Beddoe, A.H., 2004. Dose responses of diamond detectors to monoenergetic X-rays. Nuclear Instruments and Methods in Physics Research B 213, 646–649.
Contents lists available at SciVerse ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Determination of the energy response of closely tissue-equivalent diamond dosimeters for radiotherapy dosimetry R.P. Hugtenburg a,b,n, F.M. Saeedi c,d, A.E.R. Baker c a
College of Medicine, Swansea University, Singleton Park, Swansea SA2 8PP, UK Department of Medical Physics and Clinical Engineering, Singleton Hospital, Abertawe Bro Morgannwg UHB, Swansea, Wales SA2 8QA, UK c Department of Physics and Astronomy, University of Birmingham, Birmingham B15 2TT, UK d Oncology Center, National Guard Hospital-Jeddah, Saudi Arabia b
a r t i c l e i n f o
abstract
Available online 28 March 2012
A monoenergetic X-ray syncrotron source was used to determine the energy response of a diamond detector in the range 5–25 keV, clarifying the elemental composition of the detector. The response is shown to be influenced by the detector housing and electrical contacts. A model for the energy response of the detector is determined that is valid in the 5 keV–15 MeV with an accuracy of 5% and therefore can be used to correct the dosimeter response to low-energy and scattered radiation. Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved.
Keywords: Diamond detector X-ray synchrotron Energy response
1. Introduction Solid-state detectors are increasingly used for dosimetry purposes in megavoltage photon radiotherapy, in order to avoid issues relating to the use of tissue-equivalent ionisation chambers, including a lack of electronic equilibrium (Saeedi et al., 2004) and poor sensitivity in small radiation fields, used in stereotactic radiosurgery and intensity modulated radiotherapy. The sensitivity of solid-state dosimeters is generally excellent, however many dosimeters, including silicon diodes, depart significantly from tissue-equivalence, particular in their sensitivity to low-energy and scattered X-rays (Yin et al., 2002, 2004). In recent years liquid-filled ionisation chambers and diamond-based detectors have been considered as closely tissue-equivalent detectors that do not suffer in the absence of electronic equilibrium and therefore can be used to measure dose in small fields and very close to the surface of a phantom. In this work we report on the tissue-equivalence of a commercially available diamond dosimeter based on natural diamonds. Measurements utilising a monoenergetic source are compared with the response recorded in previous work with quasi-monoenergetic photon sources and Monte Carlo methods, in order to develop models for the energy dependence of the detector in the range 0.005–15 MeV. It has been shown in previous work (Yin et al., 2004) that corrections to the diamond detector dose response are needed to achieve high orders of dosimetric accuracy in the kilovoltage
n Corresponding author at: Department of Medical Physics and Clinical Engineering, Singleton Hospital, Abertawe Bro Morgannwg UHB, Swansea, Wales SA2 8QA, UK. E-mail address: [email protected] (R.P. Hugtenburg).
range, particularly below 100 keV, where the photoelectric effect predominates. The response is strongly influenced by the lower effective-Z of carbon relative to water and the presence of metallic electrical contacts. The dosimetry of X-ray synchrotron beams has recently been considered for several other detectors with good tissue-equivalence, including radiochromic film (Nariyama, 2011) and 3D gels (Abdul Rahman et al., 2010).
2. Methods A natural diamond-based detector (PTW Freiburg/ IPTP Dubna, serial number 7017) was compared with a parallel-plate ionisation chamber (PTW type 23342) in a monoenergetic synchrotron X-ray beam, with both detectors embedded in RW3 Solid-Water (PTW). The diamond detector is calibrated by the manufacturer using a 60 Co gamma-ray source. Measurement of dose with the detector has been shown to be highly stable and reproducible. Diamond detectors typically have a small dose-rate non-linearity. Corrections were made for a slightly sub-linear dose-rate response in the detector with an exponent, D ¼ 0:975. In previous experimental studies of the diamond detector, response to therapeutic X-rays of energy 45 kVp and 100 kVp from a superficial X-ray unit (Hugtenburg et al., 2001) and 6 MV and 15 MV X-rays from a therapeutic linear accelerator have been measured. It is generally observed that the energy dependence of the diamond detector is slowly varying above 100 keV and below 10 MeV, where the Compton scattering process predominates. The parallel-plate ionisation chamber (PTW type 23342) is designed for the dosimetry of superficial X-rays and is described by the manufacturer as having tissue-equivalence of better than 2% in the range 8–32 keV.
0969-8043/$ - see front matter Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apradiso.2012.03.016
24
R.P. Hugtenburg et al. / Applied Radiation and Isotopes 71 (2012) 23–24
Table 1 Fractional composition (by weight) of the diamond detector used in Eq. (1). Element
C
Cu
Ag
Au
Fraction
0.997
0.0005
0.001
0.0015
Fig. 1. Experimental data and modelling of the response the diamond detector to the parallel-plate ionisation chamber. Data from Yin et al. (2004) is merged with current data measured on an X-ray synchrotron. The solid lines show models based on attenuation and absorption for the composite material detector defined in table 1 as well as a pure carbon detector. The measured data is repeated in the inset figure, where the experimental error is comparable to the symbol size.
X-ray synchrotron station 16.3 at Daresbury laboratories (Collins et al., 1998) was used to provide monoenergetic photons in the energy range from 5 to 30 keV. The second generation Daresbury (SRS) X-ray synchrotron produced highly monoenergetic X-rays of comparable intensity to therapeutic beams, through the use of a Si monochromator (DE=E o103 ). The X-ray synchrotron beam profile was narrower than the detectors in the vertical direction, so a motorised stage was used to move the detectors slowly across the beam, collimated to 8 mm 0.1 mm to generate an 8 mm 8 mm field. The field uniformity was shown to be of the order of 5%, determined with radiochromic film (MD-55-2, GafChromic film). A model for the detector response, based on linear combinations of the mass-absorption coefficient of the detector constituents and attenuation of the beam in the detector and upstream housing, is proposed, as demonstrated in the following equation: R¼
ðmen =rÞdd ðmtÞpe ðmtÞdd e e ðmen =rÞwater
in previous experimental studies with a quasi-monoenergetic X-ray set (Yin et al., 2004), and utilising the 60Co dose calibration provided by the manufacturer. Agreement of typically better than 5%, between the experiment and the model, is achieved over this range, based on the given constituent composition of the detector. The dose response in regions far from the atomic orbital edges is shown to be reproduced well by the model described in Eq. (1). The largest departures from the model occur in the energy range below 25 keV, in which two effects predominate; namely the increased divergence of the absorption cross-sections of carbon and oxygen, and rapid changes in absorption in the vicinity of the K-edge of Cu (at 9.0 keV), the K-edge of Ag (25.5 keV) and the L-edges of Au (14.3, 13.7 and 11.9 keV). Energy deposition in the ionisation chamber was modelled as an air-equivalent material. The figure inset highlights the new experimental data measured with the diamond detector. Previous work (Hugtenburg et al., 2007) has shown that modelling in the vicinity of K-edges can be significantly in error ( 4 5%) due to large variations in the cross-section. The effects were shown to be mitigated by choosing a finer grid for the interpolation of cross-sections in the case of Monte Carlo modelling. Use of a more sophisticated Monte Carlo model of the detector, which includes geometrically precise models for the constituent components of the diamond detector, has been considered, with some improvement to the reproduction of the response in the vicinity of K- and L-edges observed. The ionisation chamber has not been modelled in detail and instead is considered as a wateror air-equivalent homogeneous material, air-equivalence proving to offer better agreement with the response of the ionisation chamber.
4. Conclusion The energy response of a diamond detector is examined with a monoenergetic X-ray synchrotron source, enabling the consideration of the influence of the constituent metallic elements, Cu, Ag and Au. The experimental data extends and clarifies issues relating to the modelling of the energy response of diamond detectors to a wider range of energies. The model of the energy response can be used to correct for distortions to the detector response for X-ray therapeutic beams at large field sizes, relating to changes in the response of the detector to scattered photons. References
ð1Þ
where the ‘dd’ subscript refers to coefficients and thickness associated with the diamond detector composite material established in this work, while the ‘pe’ subscript refers to the polyethylene housing of the diamond. The thickness of the polyethylene housing upstream of the detector was, t pe ¼ 1 mm, the thickness of the diamond detector to the effective point of measurement was t dd ¼ 0:25 mm. The attenuation and massabsorption coefficients for the diamond detector ðmen =rÞdd were determined from a combination of the four constituent elements, as shown in Table 1. The detector dimensions and constituents were identified using a combination of methods, including scanning of the detector with X-ray fluorescence, radiographic studies and manufacturer descriptions.
3. Results Fig. 1 shows the energy dependence of the diamond detector in the 5–205 keV, combining the present data with data recorded
Abdul Rahman, A.T., Bruer-Krisch, E., Brochard, T., Adamovics, J., Bradley, D., Doran, S., 2010. Verification of synchrotron microbeam radiation therapy using a purpose-built optical CT microscope. J. Phys.: Conf. Ser. 250, 012083. Collins, S.P., Cernik, R.J., Fell, B., Tang, C.C., Harris, N.W., Miller, M.C., Oszlanyi, G., 1998. Station 16.3: a high-resolution single-crystal diffraction facility at the SRS. J. Synchrotron Rad. 5, 1263–1269. Hugtenburg, R.P., Johnston, K., Chalmers, G.J., Beddoe, A.H., 2001. Application of diamond detectors to the dosimetry of 45 and 100 kVp therapy beams: comparison with a parallel-plate ionization chamber and Monte Carlo. Phys. Med. Biol. 46, 2489–2501. Hugtenburg, R.P., Saeedi, F., Bradley, D.A., 2007. Modeling dose response to synchrotron X-rays in solid-state and biological systems. Radiat. Eff. Defects Solids 162, 765–769. Nariyama, N., 2011. Macroscopic and microscopic X-ray dose mappings with CCDbased GafChromic film dosimeters. Applied Radiation and Isotopes, http://dx. doi.org/10.1016/j.apradiso.2011.12.009, in press. Saeedi, F., Yin, Z., Hugtenburg, R.P., 2004. Dosimetry in conditions of electron disequilibrium. Radiat. Phys. Chem. 71, 979–980. Yin, Z., Hugtenburg, R.P., Beddoe, A.H., 2002. Response of silicon diode dosemeters to scattered radiation from megavoltage photon beams. Radiat. Prot. Dosimetry 101, 415–418. Yin, Z., Hugtenburg, R.P., Beddoe, A.H., 2004. Response corrections for solid-state detectors in megavoltage photon dosimetry. Phys. Med. Biol. 49, 3691–3702. Yin, Z., Hugtenburg, R.P., Green, S., Beddoe, A.H., 2004. Dose responses of diamond detectors to monoenergetic X-rays. Nuclear Instruments and Methods in Physics Research B 213, 646–649.