Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem
Development of tailor-made silica fibres for TL dosimetry D.A. Bradley a,b,n, Siti F. Abdul Sani a, Amani I. Alalawi a,c, S.M. Jafari a,d, Noramaliza M. Noor e, A.R. Hairul Azhar f, Ghafour Amouzad Mahdiraji g, Nizam Tamchek h, S. Ghosh i, M.C. Paul i, Khalid S. Alzimami j, A. Nisbet a,k, M.J. Maah l a
Centre for Nuclear and Radiation Physics, Department of Physics, University of Surrey, Guildford, Surrey, GU2 7XH, UK Department of Physics, University of Malaya, 50603 Kuala Lumpur, Malaysia c Physics Department, Faculty of Applied Sciences, Umm Al-Qura University, Makkah, P. O. Box 175, Saudi Arabia d Radiology Department, Faculty of Medicine, Kabul Medical University, Kabul, Afghanistan e Department of Imaging, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia f Faculty of Engineering, Multimedia University, 2010 Cyberjaya, Selangor, Malaysia g Photonics Research Group, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia h Department of Physics, Faculty of Science, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia i Fiber Optics and Photonics Division, Central Glass and Ceramic Research Institute, Kolkata 700032, India j Department of Radiological Sciences, King Saud University, Riyadh 11432, Saudi Arabia k Department of Medical Physics, The Royal Surrey County Hospital NHS Trust, Guildford, Surrey GU2 7XX, UK l Department of Chemistry, University of Malaya, 50603 Kuala Lumpur, Malaysia b
H I G H L I G H T S
Optical fibres tailor-made for TL dosimetry. Sensitive to diagnostic as well as therapy doses in medicine. Preform and fibre pulling facilities. Relative TL and EPR measurements.
art ic l e i nf o
a b s t r a c t
Article history: Received 2 July 2013 Accepted 30 March 2014
The Ge dopant in commercially available silica optical fibres gives rise to appreciable thermoluminscence (TL), weight-for-weight offering sensitivity to MV X-rays several times that of the LiF dosimeter TLD100. The response of these fibres to UV radiation, X-rays, electrons, protons, neutrons and alpha particles, with doses from a fraction of 1 Gy up to 10 kGy, have stimulated further investigation of the magnitude of the TL signal for intrinsic and doped SiO2 fibres. We represent a consortium effort between Malaysian partners and the University of Surrey, aimed at production of silica fibres with specific TL dosimetry applications, utilizing modified chemical vapour deposition (MCVD) doped silica–glass production and fibre-pulling facilities. The work is informed by defect and dopant concentration and various production dependences including pulling parameters such as temperature, speed and tension; the fibres also provide for spatial resolutions down to o10 mm, confronting many limitations faced in use of conventional (TL) dosimetry. Early results are shown for high spatial resolution ( 0.1 mm) single-core Ge-doped TL sensors, suited to radiotherapy applications. Preliminary results are also shown for undoped flat optical fibres of mm dimensions and Ge-B doped flat optical fibres of sub-mm dimensions, with potential for measurement of doses in medical diagnostic applications. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Silica fibres Defects Thermoluminescence Dosimetry
1. Introduction
n Corresponding author at: Centre for Nuclear and Radiation Physics, Department of Physics, University of Surrey, Guildford, Surrey, GU2 7XH, UK. E-mail address:
[email protected] (D.A. Bradley).
In support of state-of-the-art radiotherapy there is need for dosimetry systems (i) that are sensitive down to small fractions of a Gy; (ii) that offer high spatial resolution, down to sub mm for small fields, and; (iii) that can be used to make accurate high precision (better than 3%) measurements that involve the steep
http://dx.doi.org/10.1016/j.radphyschem.2014.03.042 0969-806X/& 2014 Elsevier Ltd. All rights reserved.
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
dose gradients on edges of radiation fields. The ability to make in vivo dose evaluations represents yet another strong desire. Commercially available phosphor-based thermoluminescent (TL) dosimeters (TLD) enjoy wide-ranging applications, including an ability to measure the doses associated with diagnostic techniques. Their performance is also supported by knowledge of the production dependences. However, in regard to applications in radiotherapy, they are relatively large (typically mm) and potentially hygroscopic, failing in part to meet demands for detailed measurement of dose distribution within tissues. Conversely, optical fibre dosimeters offer the potential for sensitive in vivo radiotherapy assessment at high spatial resolution ( 0.1 mm). They can also provide a radiation measurement environment close to that of a Bragg–Gray cavity (the presence of the dosimeter minimally disturbing dose deposition local to the dosimeter), being important in accurate evaluation of the absorbed dose in tissue. Nevertheless, until recently, optical fibre dosimeters have had limited success in measuring diagnostic doses. Additionally, various control issues involved in producing a well-characterized silica-based material are yet to be defined in detail.
Fig. 1. Response of Ge-doped fibres of 8, 11 and 50 mm core dopant diameters, to 80 kVp irradiation. The solid lines are least-square fits to the data (note that the y-axis to the right represents the TL yield of 11 mm core dopant diameter fibres). The standard error of the mean error bars is smaller than the size of the data points.
2. University of Surrey/Malaysian consortium The present collaboration [the University of Surrey and a Malaysian consortium: University of Malaya, Multimedia University (MMU), Universiti Putra Malaysia, Telekom R&D (TRD), Universiti Teknologi Malaysia and the Malaysian Nuclear Agency, among others] is developing optical fibres as a 1-D system, based primarily on TL readout technology. The work utilizes the method of modified chemical vapour deposition (MCVD) for doped silica–glass production, supported by fibre-pulling facilities that can provide for a number of morphologies (cylindrical- and flat-fibres for example). The aim is to design and produce tailored fibres, characterized in terms of radiation response and dependence on production parameters also examining the production possibilities that arise from access to the consortium facilities. These allow for tailor-made fibres, producing enhanced TL yield per unit dose. The work demands physical interpretation at the defect level (not only for media whose signal is dependent solely on the presence of intrinsic defects but also for media whose signal is modified by deliberate inclusion of extrinsic defects). Such analysis can be done through a variety of techniques, including UV–vis, x-ray absorption near edge structure (XANES) analysis (as discussed later) and x-ray spectroscopy (XPS).
3. TL response Previous studies of the irradiation response of Ge-doped silica (SiO2) telecommunication fibres include exposure to UV light, x-ray beams, a synchrotron microbeam facility, electrons, protons, neutrons and alpha particles, measuring doses from a fraction of 1 Gy up to 10 kGy (Bradley et al., 2012). Figs. 1 and 2 show previously unpublished results, obtained with such fibres; the details of the investigations are described by Bradley et al. (2012) and Abdul Rahman et al. (2012). In brief these concern TL response over a wide range of dose (1– 30 Gy), each data point representing the mean TL yield of 20 individual fibres. Fibres with core dopant diameters of 8, 11 and 50 μm, irradiated by kilovoltage photon beams (80 kVp and 250 kVp) were used. It is apparent from these that the energy response of the fibres can be used as a means of determining the effective incident energy of a bremsstrahlung source or indeed the energy of a monochromatic photon source. The results also show that the Ge dopant in commercially available
Fig. 2. Response of Ge-doped fibres of 8, 11 and 50 mm core dopant diameters to 250 kVp irradiation. The solid lines are least-square fits to the data (note that the y-axis to the right represents the TL yield of 11 mm core dopant diameter fibres). The standard error of the mean error bars is smaller than the size of the data points.
silica optical fibres gives rise to appreciable TL, weight-for-weight fibres offering sensitivity to megavoltage x-rays several times that of the popular LiF material TLD100 (Yusoff et al., 2005). Compared to TL phosphors, the fibres offer improved positional sensitivity, 100 μm cf. mm. The response of these fibres encourages present conduct of more comprehensive investigation, seeking a fuller understanding of the magnitude of the TL signal for undoped SiO2 as well as for doped SiO2 in a range of dopant concentrations, providing a basis for exploitation of the TL for dosimetry. While it is apparent that the fibres have considerable utility as dosimeters in radiotherapy, the potential exploitation of these media as diagnostic dosimeters (down to doses of o0.1 mGy) has remained largely unexplored. The latter is a particular focus of interest for future studies by this group.
4. TL phenomena in SiO2 4.1. Structural defects, transmission losses and TL In high-intensity radiation environments (e.g. particular locations in accelerator and nuclear reactors halls) the radiation interactions in optical communication fibres result in structural defects that lead to transmission losses. Such recognition has lead quite naturally to discussion of optical fibres as dosimeters, the irradiation dose being reflected in the defect-dependent TL signal. As an instance, in Fig. 3 we show a comparison of TL responses of
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
Predicted TL yield from 8 m length of fibre compared against measured TL yield from 5mm length of CorActive fibre
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
18
3
Al-2 (23 µm)
16
Ge CorActive (50 µm) Al-3 (23 µm)
14 12
AlC-3 (23 µm)
10
AlC-2 (23 µm)
8
Ge India (23 µm)
6 4
Fig. 4. Schematic of amorphous silica structural makeup, showing two adjacent tetrahedra and the bond angle.
2 0 0
50 Number of fibres
100
Fig. 3. Relative response of Al and Ge-doped fibres of 23 and 50 mm core dopant diameter irradiated using a 6 MV x-ray linac source. The Al and Ge-doped fibres of 23 mm core dopant diameter were fabricated at the Central Glass and Ceramic Research Institute, Kolkata. The Ge-doped 50 mm core dopant diameter fibres were obtained from CorActive (Canada). The fibres used herein were all unscreened, reflecting in the widely varying TL yields.
various fibres irradiated to a photon dose of 3 Gy produced at 6 MV, scaled from 5 mm lengths typically used by this group in Ge-doped fibre TL studies to present choice of an 8 m length of fibre. The TL response and dB loss in light transport are essentially different measures of the same thing, both being directly defect concentration dependent. The relative TL response of the 50- and 23 mm core dopant diameter examples shown in the figure should simply be the ratio of the square of the core dopant diameters (¼ 4.7 for the Ge-doped fibres, as also apparent in the recorded results), reflecting the total amount of dopant per unit length in the two fibres. In present results, radiation losses are within the approximate range 1 to 3 dB m 1, the Al-doped fibres being the most sensitive. The quoted dB losses are in line with expectation, as indicated for instance by Lu et al. (1999). Here it is to be noted that Al acts as a getter for oxygen, the oxygen defects giving rise to the observed enhanced response.
structure. In the micro-crystallites model the distribution of bond angles falls within the types of structure included in the continuous random network model and hence within the limit of small crystallites the two models converge. In amorphous silica, defects in the short and mid-range order are disruptions in the normal SiO2 tetrahedron coordination and normal two-fold coordination of the oxygen atoms. The absence of oxygen atoms in the tetrahedra is referred to as oxygen vacancy defects (or E0 centres) while the absence of silicon atoms is referred to as silicon vacancy defects. One common type of defect in silica is the broken or dangling bond, creating atoms with dangling orbitals populated by unpaired electrons. The main defects that fall into the dangling bond category are the non-bridging-oxygen (NBO) centre (O3RSi–O ), the peroxy radical, and variations of the E0 centre, i.e. the oxygen vacancy. One important diamagnetic defect in silica is the neutral oxygen vacancy where the Si–Si bond is formed, being usually a precursor to one of the E0 centre types of defect. One important feature of the intrinsic defect is that their number is temperature dependent. For silica, there are two main defects within this category, the previously mentioned oxygen vacancy centre and the self-trapped exciton, the latter being a defect resulting from the interaction between an excited electron and the corresponding hole left in the valence band.
5. Defect sensitive analytical techniques 4.2. Silica structural defects Clearly, defects are central to the TL phenomena. In crystals a band-gap exists between the valence and conduction bands, with no electron- or hole-states allowed within that gap. Disturbance of the crystal structure by the presence of a defect can be associated with one or more additional energy levels within the forbidden gap. Unlike the perfect crystal energy bands that extend throughout the crystal, the additional levels are localized at the crystal defect. In distinction to the long-range order of crystalline media such as quartz (a crystalline form of silica), the silica from which the fibres are composed are amorphous. However, silica x-ray diffraction measurements indicate the structural order present in crystalline media to be preserved in such glassy material over short and intermediate range scales. The atomic coordination and first and second neighbour distances of silica are very similar in the amorphous and crystalline forms and thus also among the basic building blocks of the two solid states of SiO2.
5.1. Electron paramagnetic resonance (EPR), the TL technique and diamagnetic defects With regard to the broken or dangling bond being paramagnetic as well as thermoluminescent (Fig. 5a), the defects are detectable using the electron paramagnetic resonance (EPR) technique (Fig. 5b). For more on this see Jafari et al., 2014. Fig. 5 shows example TL and EPR signals for a particular type of jewellery glass bead (GB) of mass 10 mg and an optical fibre (OF) of mass 0.2 mg. Comparison of the TL and EPR signals due to identical irradiations of the OF and GB (Fig. 5c and d) reveals comparable thermoluminescence but widely diverging paramagnetic defects, the EPR of signal being practically insignificant (Fig. 5b). Of further note is that the TL study was carried out for a dose of 3 Gy while the EPR study was obtained using a dose of 1 kGy, the TL signal clearly being the more practically useful dosimetric signal for both dosimeter types. Diamagnetic defects are not detectable by EPR but can be studied by mapping out their energy levels using absorption, luminescence, or photoionization spectroscopy.
4.3. The continuous random network and microcrystalline models 5.2. XANES study Two models exist for the structure of amorphous silica—the continuous random network model and the microcrystalline model. In the former, each oxygen atom is shared by two SiO2 tetrahedra (Fig. 4). The only difference to the crystalline form is the bond angle, the value of the angle being allowed to vary from one tetrahedron corner to another, forming a random network
Fig. 6 shows an example of a synchrotron-based x-ray absorption near-edge structure (XANES) study of Cu-doped silica, with a Cu þ formulation and a Cu2 þ formulation, comparison being made with results for a thin-film of neutral Cu and CuSO4 (i.e. Cu2 þ ) (Yusoff et al., 2005). It is apparent that this form of photon
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
4
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 5. (a) Radiation response of SiO2–GeO2 doped fibres, doped core and cladding diameter of 50 μm and 124.7 7 0.1 μm respectively (CorActive, Canada) and two types of Glass Bead (White and Frosted White, Mill Hill, Japan), exposed to 1–2500 cGy 6MV x-rays. (b) The relatively weak EPR signal obtained from a frosted white glass (SiO2) jewellery bead and an insignificant EPR signal obtained from 5 mm OF irradiated to a dose of 1 kGy from a Co-60 source. (c) Relative TL yield (at 3 Gy). (d) Relative EPR signal (at 1 kGy)
Fig. 7. XANES spectra from a sample of Ge-doped silica.
Fig. 6. XANES spectra from two samples of Cu-doped silica: curve (b) Cu þ formulation and curve (c) Cu2 þ formulation. Comparison is made between results for a thin-film of neutral Cu [curve (a)] and CuSO4 (i.e. Cu2 þ ) [curve (d)]. From Yusoff et al. (2005).
absorption study, which depends on the tuneable capability of the synchrotron source, has a powerful ability to determine the charge state of the medium. Fig. 7 shows a recent pilot measurement of one of the Ge-doped silica fibres fabricated by this collaboration, being the first such collaboration-fibre measurement, the full analysis of which has yet to be worked through. The absorption edge, at 11,114 eV, is similar to quartz-type GeO2 (Witkowska et al., 2006). The clear intent of present study is to investigate how the dopant is incorporated within the silica matrix, a matter which among other properties predicates TL yield.
6. The collaboration fibre production plans The work of the collaboration is intended to pave the way for tailored dosimeter fibres, informed by dopant concentration and
pulling parameter effects that include temperature, speed and tension, with potential for measurement of doses from the extremes of environmental radiation levels through to radiation sterilization/processing doses. The ability to also provide for spatial resolutions down to o10 μm confronts the many limitations faced in use of conventional thermoluminscence (TL) dosimetry. On the basis of the foregoing, the collaboration is currently engaged in development of a 1-D TLD system for a range of ionizing radiation dosimetry applications, in particular designing, fabricating and characterizing Ge-doped SiO2 fibres. The work requires support in the form of XANES absorption spectroscopy and a number of other analytical techniques to provide for a fuller appreciation of the TL characteristics of the fibres. As previously mentioned, the collaboration represented herein is utilizing the MCVD doped silica-glass production and fibre-pulling facilities available to it within the consortium. For present purposes we provide a general set of statements concerning the methodology used in the fibre production phase (Phase 1) as depicted in Fig. 8, the samples being produced using chelate and the modified chemical vapour deposition (MCVD) technique. More detailed description of the means of fibre
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Start fibre design
Fabricate optical fibre/flat fibre
Absorption spectroscopy/analysis
Repeat for different dopant concentration, dimensions
TLD characterization
Fig. 9. Photomicrograph of the side on view of a Ge–B doped flat fibre, together with spatial dimensions, in terms of the outermost value (V1), the flat portion dimension (V2) and the thickness (H1). The flat fibre has been created through a vacuum applied to the fibre whilst hot.
Data Analysis
Fig. 8. The upper schematic shows the design process being followed by the consortium. The section identified by the words ‘absorption spectroscopy/analysis’ includes XANES, TOF-SIMS and SEM-EDX analyses. The lower schematic indicates our present proposal for production and analyses of various samples, it being further indicated that optically stimulated luminescence (OSL) studies and radio luminescence (RL) studies are also being conducted within the consortium, discussion of these being outside of the scope of the present paper, the focus of which is TL study. Investigation has begun of various dopants, dopant concentrations (indicated by ‘mol%’) and fibre dimensions (indicated by ‘Dia’).
production is provided by Salasiah et al. (2013). Samples are now becoming available for analysis, minus the core/cladding of optical communication fibres, with diameters from 10 to 100 μm and Ge concentrations of between 2 and 6 mol%. The doped preform is a result of MMU/TRD research, with drawing of the fibres being carried out by the UM team (diameters being defined by optical microscopy and a Zumbach unit). For low dopant concentrations use is made of electron probe micro-analyzer (EPMA) analysis and/ or time-of-flight secondary ion mass spectroscopy (TOFSIM) to then be associated with a well-defined set of parameters including temperature, speed and tension used in fibre production and optical characterization. The fibres have low dopant homogeneity, challenged by the high ratio ( 200) between the preform and final fibre diameters, (from an initial 25 mm to a final 0.125 mm). TOFSIM results are expected to guide improvement in fibre fabrication and final TLD quality. Phase 2 of the research is proposed to involve absorption spectroscopy characterization of the fabricated doped-optical fibres, scanning across the K-edge of Ge, from 10.5 keV to 11.5 keV, the XANES analysis of the fabricated doped fibre allowing investigation of the charge state of the Ge, whether it is incorporated into the silica in a singly, doubly or triply ionized state or perhaps as a mixture of these. This is of considerable importance, providing underpinning understanding of the luminescence yield of the doped medium, it having been shown in previous studies that the charge state has a controlling role in defining the luminescence yield of a particular medium.
Fig. 10. A comparison of TL yield from undoped flat fibres, doped cylindrical fibres and TLD-100 chips, normalized to a dose of 3 Gy. Present results concern undoped flat fibres ( 1 mm wide 1 mm thick 10 mm length mass 20 mg).
7. Early results and discussions Here we report initial findings from fibres produced by the collaboration facilities. Present results concern undoped flat fibres ( 1 mm wide 1 mm thick 10 mm length mass 20 mg) and Ge-B doped flat fibres (of dimensions 320 μm wide 32 μm thick 5 mm length; mass 0.2 mg). Fig. 9 shows a photomicrograph of one of the Ge–B doped fibres. Fig. 10 shows results for the undoped fibres, irradiated to a dose of 3 Gy using 6 MV photons and to a dose of 0.1 Gy using 140 kVp photons. The data have been normalized to a dose of 3 Gy to allow a direct comparison of the TL yield. Apparently the TL yields from the current batch of FFs, irradiated at 140 kVp, offer a sensitivity which while less than that of the TLD-100 chips is nevertheless somewhat comparable and hence capable of measuring doses at diagnostic levels. For 6 MV irradiations the response of the TLD100 chips close to an order of magnitude greater than that of the 50 μm Ge core dopant diameter fibres, which in turn are at least an order of magnitude greater than that of the 9 μm Ge core dopant diameter fibres is further apparent. The latter two fibres are of course relatively light by at least one order of magnitude compared to the FF, accounting in large part for the difference in TL yield, further moderated by the fact that the cylindrical fibres are doped. It is first to be acknowledged that the response of the undoped FFs is closely linked with the SiO2 vacancy defects and the dangling bonds previously addressed. In addition to TL from the foregoing, the fibres, whether doped or undoped, are further affected by drawing-induced defect centres, all of the fibres produced by the University of Malaya collaboration having been made with Suprasil F500 (Heraeus Quarzglas GmbH, Germany), a reduced OH content (typically 0.02 ppm) fused silica with a great
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
6
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
many defects when compared against crystal or Quartz glass. As such, it is apparent that defect centres in fibres made from the Suprasil preform can be manipulated, to an extent presently unknown, by controlling the pulling parameters, reflected by changes in the TL response for fibres of the same size and type. By controlling the pulling parameters (temperature, pulling speed, and pressure), there is potential to make the fibre of higher or lower TL response, a matter currently under investigation within the collaboration. In Fig. 11 results of a screening process for a batch of 258 samples of Ge-B-doped Flat Fibres irradiated to a dose of 500 mGy using an Eldorado Cobalt-60 machine located at Nuclear Malaysia, the dose rate being 0.05 Gy/min are shown. The fibres were irradiated on the surface of a polymethyl methacrylate, PMMA (Perspex) solid water phantom of dimensions 30 cm 15 cm 30 cm at a distance of 100 cm from the radiation source, the field size being 10 cm 10 cm. Prior to the process of discarding the samples showing a more disparate TL yield, the mean optical fibres response per unit mass was 34.02Eþ 03 nC with a standard deviation of 10.17Eþ03 nC. Selection was made to retain a group of 36 optical fibres, reflecting homogeneous dopant concentrations to within 75%. This group of optical fibres gave a mean response of 27.07Eþ03 nC and 1.47Eþ03 nC standard deviation. In passing, it is to be noted with respect to Fig. 10 that the inhomogeneity of dopant along the length of a fibre accounts for significant light loss in optical fibre communications. It is possible that a TL screening technique could be used in sampling fibres in optical-fibre quality control (QC). Given that the screening dose of 0.5 Gy was used to obtain a response of 30 103 nC per unit mass, compared against a background from unirradiated fibres of 0.6270.09 nC per unit mass, this points to a signal to background ratio of 10: 1 for a dose of 0.1 mGy and a lower-limit of detection approaching 0.011 mGy. Such levels are certainly sufficient for use of the doped FFs in measuring chest x-ray doses ( 0.1 mGy). Finally in seeking to obtain even higher sensitivity fibres, mention should be made here of the use of high atomic number fibre coatings, e.g. Au (Z¼ 79), to generate large numbers of photoelectrons, with path lengths of the order of several hundred microns for the photon energies of interest, and hence large TL yields. Using commercially available Ge-doped fibres, members of this group have demonstrated agreement between experimental measurements made at 250 kVp (for coating thicknesses obtained up to 100 nm) for the doseenhancement obtained in coating fibres with Au and Monte Carlo simulations (Alalawi et al., 2013). The question then arises at to the optimum thickness of a coating in order to generate the maximum dose enhancement factor (DEF). The situation is one of interplay between photoelectron generation within the coating and the stopping of electrons within that medium. Fig. 12 shows results of
Fig. 12. Results of simulation using the Monte Carlo code DOSXYZ, examining the dose enhancement factor (DEF) for increasing thicknesses of Au coating upon which 250 kVp photons are incident.
simulation using the Monte Carlo code DOSXYZ, examining the situation for increasing thicknesses of Au coating upon which 250 kVp photons are incident. The results indicate for the particular situation that a Au coating of just below 10 mm is predicted to produce a DEF of 10 over that was obtained in the absence of the coating. The practicalities of providing a single-sided coating to a FF would certainly be rather less demanding than that for a cylindrical fibre, the FF also availing itself to simple measurements of diagnostic doses based on the monitoring of entrance dose.
8. Conclusion Present research promises definition of high yield TL fibres sufficient to satisfy a range of dosimetric needs. Characterization is being sought to ensure that the mechanism of TL yield in optical fibres is well understood, allowing a favourable well controlled production situation to be established. The intended end point is to specify dosimeters, not only for clinical dosimetry but also for their application in industrial/energy–industry settings.
Acknowledgements Fig. 6 was reprinted from Radiation Physics and Chemistry, 74, 2005, 459–481, Yusoff A.L., Hugtenburg R.P. and Bradley D.A., Review of development of a silica-based thermoluminescence dosimeter, with permission from Elsevier. The authors are grateful for a University of Malaya – Ministry of Higher Education of Malaysia UM–MOHE High Impact Research Grant UM.C/625/1/HIR/33. The authors also thank staff at the Royal Surrey County Hospital and Department of Physics, University of Surrey for help in performing the irradiations. References
Fig. 11. The data represented by the diamond symbol represent a set of dosimeters providing a uniform TL yield to within 7 5% of the mean, obtained from a wider unscreened collection of dosimeters (shown using the square symbol). A screening dose of 0.5 Gy was used to obtain a response of 30 103 nC per unit mass, compared against a background from unirradiated fibres of 0.62 7 0.09 nC per unit mass, pointing to a signal to background ratio of 10:1 for a dose of 0.1 mGy and a lower-limit of detection approaching 0.01 mGy.
Abdul Rahman, A.T., Hugtenburg, R.P., Siti Fairus Abdul, Sani, Alalawi, A.I.M., Issa, Fatma, Thomas, R., Barry, M.A., Nisbet, A., Bradley, D.A., 2012. An investigation of the thermoluminescence of Ge-doped SiO2 optical fibres for application in interface radiation dosimetry. Appl. Radiat. Isot. 70, 1436–1441. Alalawi Amani I., Hugtenburg, R.P.J. , Abdul Rahman, A.T. Barry, M.A., Nisbet, A., Alzimami Khalid S., Bradley, D.A.. Measurement of dose enhancement close to high atomic number media using optical fibre thermoluminescence dosimeters. Radiat. Phys. Chem., 2013 95, 145–147. http://dx.doi.org/10.1016/j.radphyschem. 2013.05.017. Bradley, D.A., Nisbet, A., Abdul Rahman, A.T., Issa, F., Mohd Noor, N., Alalawi, A, Hugtenburg, R.P., 2012. Review of doped silica glass optical fibres: their properties and potential applications in radiation therapy dosimetry. Appl. Radiat. Isot. 71, 2–11, http://dx.doi.org/10.1016/j.apradiso.2012.02.001. Jafari, S.M., Bradley, D.A., Goldstone, C.A., Sharp, P.H.G., Alalawi, Amani I., Jordan, J.T., Clark, C.H., Nisbet, A., Spyrou, N.M., 2014. Low-cost commercial glass beads as dosimeters in radiotherapy. Radiat. Phys. Chem. 97, 95–101. Lu, Ping, Bao, Xiaoyi, Kulkarni, Narayan, Brown, Kellie, 1999. Gamma ray radiation induced visible light absorption in P-doped silica fibers at low dose levels. Radiat. Meas. 30, 725–733.
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
D.A. Bradley et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎ Salasiah, M., Nordin, A.J., Fathinul Fikri, A.S., Hishar, H., Tamchek, N., Taiman, K., Ahmad Bazlie, A.K., Abdul-Rashid, H.A., Mahdiraji, G.A., Mizanur, R., Noor Noramaliza, M., Radiation dose to radiosensitive organs in PET/CT myocardial perfusion examination using versatile optical fibre. Full paper 0204013; In: Kalli Kyriacos, Kanka Jiri, Mendez Alexis (Eds.), Proceedings of SPIE Conference Micro-structured and Specialty Optical Fibres II, Vol. 8775, 2013, p. 87750U-1.
7
Witkowska, A., Sikora, B., Trzebiatowski, K., Rybicki, J., 2006. Germanate anomaly in heavy metal oxide glasses: an EXAFS analysis. J. Non-Cryst. Solids 352, 4356–4361. Yusoff, A.L., Hugtenburg, R.P., Bradley, D.A., 2005. Review of development of a silicabased thermoluminescence dosimeter. Radiat. Phys. Chem. 74, 459–481.
Please cite this article as: Bradley, D.A., et al., Development of tailor-made silica fibres for TL dosimetry. Radiat. Phys. Chem. (2014), http: //dx.doi.org/10.1016/j.radphyschem.2014.03.042i
