ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

Thermoluminescence characteristics of different dimensions of Ge-doped optical fibers in radiation dosimetry

Mahfuza Beguma,b, A.K.M.Mizanur Rahmana,b, H.A. Abdul-Rashida Z.Yusoffa,Mahbuba Begumb, K.A. Mat-Sharifc, Y.M. Amind D.A. Bradleye

aFaculty of Engineering, Multimedia University 63100 Cyberjaya, Selangor Darul Ehsan, Malaysia email: [email protected]

bBangladesh Atomic Energy Commission Bangladesh

cTelekom Research and Development Lingkaran Teknokrat Timur 63000 Cyberjaya, Selangor Darul Ehsan, Malaysia

cDepartment of Physics Faculty of Science University of Malaya 50603 Kuala Lumpur, Malaysia

eDepartment of Physics University of Surrey Guildford GU2 7XH, UK

Abstract

Important thermoluminescence (TL) properties of five (5) different core sizes Ge- doped optical fibers have been studied to develop new TL material with better response. These are drawn from same preform applying different speed and tension during drawing phase. The results of the investigations are also compared with most commonly used standard TLD-100 chips (LiF:Mg,Ti) and commercial multimode Ge- doped optical fiber (Yangtze Optical Fiber, China). Scanning Electron Microscope (SEM) and EDX analysis of the fibers are also performed to map Ge distribution across the deposited region. Standard Gamma radiation source in SSDL (Secondary Standard Dosimetry Lab) was used for irradiation covering dose range from 1Gy to 10 788

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

Gy. The essential dosimetric parameters that have been studied are TL linearity, reproducibility and fading. Prior to irradiation all samples ~0.5 cm length are annealed at temperature of 400°C for 1 hour period to standardize their sensitivities and background. Standard TLD-100 chips are also annealed for 1 hour at 400°C and subsequently 2 hours at 100°C to yield the highest sensitivity. TL responses of these fibers show linearity over a wide gamma radiation dose that is an important property for radiation dosimetry. Among all fibers used in this study, 100 µm core diameter fiber provides highest response that is 2.6 times than that of smallest core (20 µm core) optical fiber. These fiber-samples demonstrate better response than commercial multi-mode optical fiber and also provide low degree of fading about 20% over a

period of fifteen days for gamma radiation. Effective atomic number (Zeff) is found in the range (13.25 to 13.69) that is higher than soft tissue (7.5) however within the range of human-bone (11.6-13.8). All the fibers can also be re-used several times as a detector after annealing. TL properties of the Ge-doped optical fibers indicate promising applications in ionizing radiation dosimetry.

Keywords: Ge-doped optical fiber; Radiation dosimetry; Effective atomic number; SEM-EDX analysis

1.- INTRODUCTION

Ionizing radiation dosimetry is important for the protection of patients and workers against detrimental effects of excessive exposure of radiation. Different types of materials and techniques have been investigated to assess the magnitude of ionizing radiation. Among them, “thermoluminescence dosimeters (TLD)” are increasingly being used in different fields such as medical institutions, environmental monitoring etc. The most common commercially available TL phosphor is lithium fluoride, LiF:Mg,Ti (6Li-7.5%, 7Li-92.5%). The existence of a trace amount of impurities (magnesium) in the lattice structure of LiF provides enhanced thermoluminescence response than its purest form [Khan Faiz M., 1994].

789

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

In recent years, several researchers have investigated TL response of silica optical fibers to develop new TL material with additional advantages than conventional TLDs. Different types of radiation including photons, electrons, protons, beta, alpha particles etc have used for the studies and obtained promising TL characteristics in ionizing radiation dosimetry [Noramaliza M Noor et al., 2012; Yaakob et al., 2011; Abdul Rahman AT et al., 2011; Hashim S et al., 2010; Issa Fatma et al., 2011].

Optical fiber is usually made of two concentric layers “core” and “cladding” with different index of refraction. TL properties of SiO2 optical fiber depends on the trapping process that is caused by the occurrence of structural defects in the material [Abdul Rahman et al., 2011].

Types and concentration of impurities in the material also effects TL response. Silica optical fibers also have several properties to overcome some limitations of commonly used TL materials such as water-resistant [Yaakob Nor H et al., 2011; Abdul Rahman AT et al., 2010; Hashim S et al., 2009; Ong CL et al., 2009]. This property paves the way for their use in intercavitary and interstitial measurements and also has high spatial resolution due to its small size and also low cost. These characteristics of optical fibers make it very promising for dosimetric applications.

The objective of this work was to investigate important dosimetric characteristics of five different core sizes of Ge-doped silica optical fibers including photon dose response, reproducibility and fading for gamma irradiation. The results of the study are also compared with commercially available standard TLD-100 chips (LiF:Mg,Ti) and commercial multimode Ge-doped optical fiber (Yangtze Optical Fiber, China) that was fabricated for communication purposes. SEM (Scanning Electron Microscopy) and EDX (Energy Dispersive X-ray Spectroscopy) analysis of the fibers have also performed to study the relative elemental composition since TL response strongly depends on it. This technique is also used to determine effective atomic number Zeff of the fiber -dosimeters.

790

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

2.- MATERIALS AND METHODS

Five different core sizes of Ge doped optical fibers have been purposely fabricated by Modified Chemical Vapour Deposition (MCVD) process with sufficient dopant concentration and pulled from the same “pre-form” with different tension and speed values during the drawing phase. Core and cladding diameters of the fiber-samples are respectively (20/120) µm, (40/241) µm, (60/362) µm, (80/483) µm and (100/604) µm.

2.1.-Sample preparation

The fibers were cut into approximately 5mm long pieces using optical fiber cleaver (Cl- 03 Max, ILSIN, Korea) and diamond-cutter (S90R, Thorlabs) and cleaned with acetone. The weight of each fiber was determined carefully using an electronic balance (Mettler Toledo, Switzerland) to normalize each measurement. The samples were handled using a vacuum tweezers to avoid scratches and contamination at the surface of optical fiber- samples that may affect TL response.

2.2.-Annealing process

Annealing was carried out for each group of optical fibers immediately before each irradiation of gamma radiation. During the process the fibers were kept in an alumina container and annealed in an oven (Lindberg/Blue M, USA) at a temperature of 400ºC for 1hour period for erasing background signal and low-temperature glow peaks. TLD -100 chips were also annealed for 1hr at 400ºC and subsequently 2hr at 100ºC to remove all residual TL signal [Hossain I et al., 2013]. After heating, to avoid thermal stress the samples were slowly cooled inside the oven to finally equilibrate at a temperature of 20ºC. All fibers were located in gelatine capsule to provide homogeneity in the irradiation, routine storage and easy handling [Espinosa G et al., 2006].

791

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

2.3.-Irradiation of dosimeters

Ge-doped optical samples with five different core sizes, commercial multimode fibers and standard TLD-100 chips were exposed to Gamma-radiation from Cobalt-60 standard radiation source at SSDL-Nuclear Malaysia over the dose range from 1Gy to 10 Gy. Perspex solid water phantom (Polymethyl methacrylate, PMMA) with dimension of (30×15×30) cm3 was used to place the samples during irradiation maintaining source to surface distance (SSD) of 100 cm and field size- (10×10) cm2.

2.4.-Readout system

TL response of the fibers was measured using Harshaw Model TLD reader under nitrogen (N2) gas atmosphere to eliminate spurious light signals from triboluminescence and also to diminish oxidation of the heating element. During TL measurement different parameters were set. These were preheat temperature 80ºC for 10s, acquired temperature rate 10ºC /s, acquisition time 13.33 s and maximum anneal temperature 400ºC to sweep out any residual signal. In this case, the luminescence intensity produced by heating the sample inside the TLD-reader is directly proportional to the absorbed dose of radiation.

2.5.-SEM-EDX analysis of optical fibers

SEM-EDX analysis technique was carried out to map the relative distribution of dopant concentration inside five different core-diameter Ge-doped optical fibers. The effective atomic number, Zeff of the fibers was also determined using this technique. It is important for predicting radiation interaction properties of dosimetric material. For the sample preparation, optical fibers were cut into one (1) cm length by using optical clever to remove surface irregularity that can affect the results and also attached to the surface area of a stainless steel sample holder using carbon-tape. In the process, high energy (30 792

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru kV) electrons interacted with optical fiber samples and produced characteristic X-rays with fixed wavelength for each element in the material and provided information about relative distribution of the compositional elements. The values of Zeff for the samples were determined using the following Mayneord equation.

1 m m m m m Zeff  (a1Z1  a2Z2  a3Z3 ......  anZn ) (1)

Where a1, a2, a3……………..an are the weight fraction contributions of each element in the optical fibers to the total number of electrons in the mixture. The value of m adopted for photon practical purposes is 2.94 [Khan Faiz M., 1994].

3.-RESULTS

3.1.-Dose response

Dose response linearity is an important characteristic for TL dosimetric application. TL response of Ge-doped optical fibers with five different core sizes was investigated and compared with most commonly used standard TLD-100 chips and commercial multimode optical fiber (MMF). All the fibers provided linear TL response with different doses of gamma radiation ranging from 1Gy to 10Gy. It was also found that largest core (100µm) size optical fiber showed TL response of about 2.6, 2.0, 1.6 and 1.1 times than that of 20µm, 40µm, 60 µm and 80 µm core size optical fibers respectively and also demonstrated 26% TL response regarding standard TLD-100 chip shown in Figure1 and Figure 2. Among the fibers commercial MMF had least response. For each dose, ten (10) fiber-samples were read and averaged by discarding outlying and low response fibers.

793

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

Figure 1.- TL dose-response of five different core-size Ge-doped optical fibers and commercial fiber.

Figure 2.- Comparison of TL response of largest core optical fiber, commercial fiber and standard TLD-100 chips.

794

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

3.2.-Fading

Fading depends on several parameters such as storage temperature, time, radiation type and annealing procedure. For fading measurement, all the fibers were irradiated and kept in light-tight box at room temperature (24°C-27°C) to reduce thermally or optically stimulated release of trapped electrons. TL response of the fibers was measured at various intervals over 15 days after irradiation. It was found that five different core size optical fibers provided approximately 20% fading for the period shown in Figure 3.

Figure 3.- Fading of fiber-samples over 15 days for gamma irradiation.

3.3.-Reproducibility

Reproducibility of the samples was assessed by annealing, irradiation and readout of each type of samples over five (5) repeated cycles. Seven pieces of each type of samples were

795

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru exposed to the same dose (3Gy) for each cycle. An average value was considered for the study. From the Figure 4 it was found that optical fiber samples could be re-used several times without any noticeable change in TL response after every re-use.

Figure 4.- Reproducibility of fiber-samples through repeated cycles of annealing, irradiation and readout

796

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

3.4.-Dopant concentration

Type and concentration of impurities inside radiation-dosimeter can significantly affect TL signal. SEM-EDX analysis of the fiber samples was performed to map elemental composition and to assess the dopant concentration in the samples. Figure 5 represented bright and dark zones in the form of rings where bright portion corresponded to the presence of germinium enriched area and the dark rings were due to silica. Dopant concentration of the fiber samples was found in the range from 3.23 mol% to 3.85 mol%.

100 100 micron Si 100 micron 100000 Element Weight % Sigma % O K 59.02 0.12 O Si K 37.39 0.11 Ge K 3.59 0.05

50 2500 Ge 0 100 80 micron 80 micron 100000 Element Weight % Sigma % Si O K 55.69 0.14 O Si K 40.08 0.13 Ge K 4.23 0.06

50 2500 Ge 100 0 60 micron 100000 60 micron Element Weight % Sigma % Si O K 57.24 0.13 O Si K 38.66 0.12 Ge K 4.1 0.06

50

2500 Ge 1 Intensity (a.u.) 1 Intensity

100  0 100000 40 micron 40 micron Element Weight % Sigma %

Si O K 58.77 0.14 Relative (a.u.) Intensity

Ge K O Si K 37.44 0.13 Ge K 3.78 0.06

50 2500 Ge 100 1000000 20 micron 20 micron Element Weight % Sigma % Si O K 57.88 0.14 O Si K 38.28 0.13 Ge K 3.85 0.06

50

2500 Ge 0 -140ahhsshhshfshhfsah-120 -100 -80 -60 -40 -20 0 20 40 60 0 1 2 9 10 11 12 Diameter (micron) Energy (keV)

Figure 5.- SEM-EDX analysis of five different core-size Ge-doped optical fibers.

3.5.-Effective atomic number, Zeff

Effective atomic number of TL phosphor is required for determining correct amount of radiation that is exposed to human body. In the present study, Zeff of fiber-samples were

797

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru estimated using Mayneord equation. The elemental composition of the fibers (germinium, silicon and oxygen) was determined using SEM-EDX analysis technique and found in the range from 13.25 to 13.69 shown in Table-1.

Table 1.- Effective atomic number (Zeff) of different core sizes Ge-doped optical fibers.

Sl. no Optical fiber (core/cladding) Effective atomic number (Zeff) [µm] 1. 100/604 13.25 2. 80/483 13.69 3. 60/362 13.57 4. 40/241 13.40 5. 20/120 13.43

4.-DISCUSSION

TL characteristics and elemental composition of the fiber-samples were investigated for the development of optical fiber as a TL material.

Linear relationship between thermoluminescence response and absorbed dose is a desirable property for TL material. In the case, all fibers satisfied this TL characteristic providing good linear response over the dose range. In every case larger core optical fiber represented better response than smaller core fibers largly due to more traps present in the larger core fibers.

In fading, trapped electrons are released randomly before read out. For accurate absorbed dose measurement fading study is required without which may otherwise lead to great inaccuracy. In the case all fiber samples demonstrated almost similar lower degree of

798

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru fading (about 20%) over fifteen (15) days. TL material that contains low fading can store dosimetric information longer period of time after exposure.

During repeated annealing, irradiation and readout process, ideally TL phosphors should not suffer any significant change in TL response as long as it is not exposed to high dose of radiation. From the study, it is apparent that five different core size optical fibers could be reused up to 5 times without being detrimental effect on TL response.

In silica optical fibers dopants have important effect on TL response. It can greatly enhance TL response by providing increase number of traps. In practice non-uniformity in the distribution of dopant in optical fibers may be occurred belonging to the same production batch that causes variation in TL yields. Five different core optical fibers represented almost similar dopant concentration in the range (3.23 mol% to 3.85 mol%). Exact amount of dopant in the fibers is useful to obtain better TL response from tailor- made doped silica optical fibers

Ideally effective atomic number Zeff should be human tissue equivalent (7.42) [Abdul Rahman AT et al., 2012; Wagiran H et al., 2011; Furetta C et al., 2001] that is independent of the incident photon energy. Tissue equivalency is very important for certain type of radiation interaction such as X-ray energy (20 keV-100 keV) where photoelectric interaction is predominant and depends on the third power of the atomic number (Z3) of interacting material. In this energy region, higher atomic number of TLD can provide over-response of radiation. The value of Zeff of the samples (13.25 to 13.69) is higher than soft tissue (7.42) however within the range of human-bone (11.6-13.8).

5.-CONCLUSION

Results of the study represented several essential TL characteristics of five different core size germanium doped optical fibers including linearity of dose, reproducibility and

799

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru fading. All the fibers provided linear dose response and also could be re-used several times without any noticeable change in TL response. Lower degree of fading about 20% was also found for the fiber-samples over 15 days after irradiation. Largest core optical fiber (100 µm) represented best response among the fibers that was about 2.6 and 4.0 times than that of smallest core (20 µm) and commercial Ge-doped optical fiber respectively. SEM-EDX analysis was also performed to map relative presence of dopant concentration in the fibers. Effective atomic number was estimated in the range from 13.25 to 13.69 that is higher than soft tissue even if within the range of human-bone. All these characteristics of five different core size optical fibers are very promising for ionizing radiation dosimetry in different fields.

REFERENCES Abdul Rahman AT; Hugtenburg RP; Abdul Sani Siti Fairuz; Alalawi AIM; Issa Fatma.(2012). An investigation of the thermoluminescence of Ge-doped SiO2 optical fibers for application in interface radiation dosimetry. Applied Radiation and Isotopes 70: 1436-1441.

Abdul Rahman AT; Nisbet A; Bradley D A .(2011). Dose-rate and the reciprocity law: TL response of Ge-doped SiO2 optical fibers at therapeutic radiation doses. Nuclear Instruments and Methods in Physics Research A652:891-895.

Abdul Rahman AT; Bradley DA; Doran SJ; Thierry Brochard; Brauer-Krisch Elke; Bravin A. (2010). The thermoluminescence response of Ge-doped silica fibers for synchrotron microbeam radiation therapy dosimetry. Nuclear Instruments and Methods in Physics Research A 619: 167-170.

Espinosa G; Golzarri1 J I; Bogard J; Garcı´a-Macedo J. (2006). Commecial optical fibre as TLD material. Radiation Protection Dosimetry119 (1-4): 197–200.

Furetta C; Prokic M; Salamon R; Prokic V; Kitis G. (2001). Dosimetric characteristics of tissue equivalent thermoluminescent solid TL detectors based on lithium borate. Nuclear Instruments and Methods in Physics Research A456: 411-417.

Hossain I; Wagiran H; Yaakob NH. (2013). Thermoluminescence of Ge- and Al-doped SiO2 optical fibers subjected to 0.2-4.0 Gy external photon radiotherapeutic dose. Journal of Applied Spectroscopy 80 (4): 635-638.

Hashim S; bradley DA; Peng N; Ramli AT; Wagiran H. (2010). The thermoluminescence response of oxygen-doped optical fibres subjected to photon and electron

800

ISSSD 2014 April 13 to 16th, 2014. Cusco, Peru

irradiations. Nuclear Instruments and Methods in Physics Research A619: 291- 294.

Hashim S; Al-Ahbabi S; Bradley DA; Webb M; Jeynes C; Ramli AT; Wagiran H.( 2009). Thermoluminescence response of doped SiO2 optical fibres subjected to photon and electron irradiations. Applied Radiation and Isotopes 67: 423-427.

Issa Fatma; Latip Nur Atiqah ABD; Bradley DA; Nisbet Andrew. (2011). Ge-doped optical fibers as thermoluminescence dosimeters for kilovoltage X-ray therapy irradiations. Nuclear Instruments and Methods in Physics Research A 652: 834- 837. K.A. Mat-Sharif; M.I. Zulkifli; S.Z. Muhamad Yassin; N. Tamchek; S.M. Aljamimi; A. Yusoff; Y.M. Amin; S.A. Siti Shafiqah and H.A. Abdul-Rashid. (2013). Effect of GeCl4/SiCl4 Flow ration on Germanium incorporation in MCVD process. Proceedings of IEEE 4th International Conference on Photonic, Melaka, Malaysia: 284-287. Khan Faiz M. THE PHYSICS OF RADIATION THERAPY. Maryland, USA, Williams & Wilkins, (Second edition,1994).

Noramaliza M Noor; Nasiha A Shukor ; Hussein M; Nisbet; Bradley DA.( 2012). Comparison of the TL fading characteristics of Ge-doped optical fibres and LiFdosimeters. Applied Radiation and Isotopes 70:1384–1387.

Ong CL; Kandaiya S; Kho HT; Chong MT. (2009). Segments of a commercial Ge-doped optical fiber as a thermoluminescent dosimeter in radiotherapy. Radiation Measurements 44:158–162.

Wagiran H; Hossain I; Asni H; Ramli AT. (2011). Thermoluminescence energy response of a germinium-doped optical fiber obtained using a Monte Carlo N-partical Code simulation. Journal of the Korean Physical Society 59 (2): 337-340.

Yaakob NH; Wagiran H; Hossain I; Ramli AT; Bradley DA; Hashim S; Ali H. (2011) Electron irradiation response on Ge and Al-doped SiO2 optical fibers. Nuclear Instruments and Methods in Physics Research A637: 185-189.

Yaakob Nor H; Wagiran Husin; Imam Hossain Md; Ramli Ahmad T; Bradley DA; Ali Hasan. (2011). Low-dose photon irradiation response of Ge and Al-doped optical fibers. Applied Radiation and Isotopes 69: 1189-1192.

801