Radiation Physics and Chemistry 95 (2014) 141–144

Contents lists available at ScienceDirect

Radiation Physics and Chemistry

0969-80

http://d

n Corr

E-m

azevedo

vsmdba

susanas

journal homepage: www.elsevier.com/locate/radphyschem

Characterization of a-spodumene to OSL dosimetry

R.A.P.O. d’Amorim a,n, D.A.A. de Vasconcelos b, V.S.M. de Barros b, H.J. Khoury b, S.O. Souza a

a Universidade Federal de Sergipe, Departamento de Fısica, PO Box 353, S ~ao Cristov ~ao, SE 49100-000, Brazilb Universidade Federal de Pernambuco, Departamento de Energia Nuclear, Av. Professor Luiz Freire 1000, Recife, PE 50.740-540, Brazil

H I G H L I G H T S

c The a-spodumene can be exploited for high-doses OSL dosimetry.c The a-spodumene has linear OSL response over the dose range from 30 Gy to 10 kGy.c The reproducibility of the OSL response of a-spodumene/Teflons pellets is 2.1%.

a r t i c l e i n f o

Article history:

Received 27 October 2012

Accepted 20 January 2013Available online 29 January 2013

Keywords:

Spodumene

Optically stimulated luminescence

Dosimetry

6X/$ - see front matter & 2013 Elsevier Ltd. A

x.doi.org/10.1016/j.radphyschem.2013.01.029

esponding author. Tel.: þ55 8197030094.

ail addresses: raquelalinep@hotmail.com (R.A

.radiologia@gmail.com (D.A.A. de Vasconcelo

rros@gmail.com (V.S.M. de Barros), hjkhoury

ouzalalic@gmail.com (S.O. Souza).

a b s t r a c t

The aim of this paper is to evaluate the optically stimulated luminescence (OSL) response of spodumene

(LiAlSi2O6) silicate and its potential use for gamma radiation dosimetry. A natural Brazilian crystal of a-

spodumene was used in this study. After the crystal grinding, pellets with the diameter of 6.0 mm were

prepared using a mixture of a-spodumene and polytetrafluoroethylene (Teflons) (1:2). To study the

OSL response, the samples were irradiated with gamma radiation beam of Co-60 and the response was

measured using an OSL Reader in two modes of operation: Continuous-wave and pulsed stimulation.

The results of the integrated OSL curve of the pellets irradiated with the dose of 30 Gy showed that their

response is reproducible within 72%. The variation of the OSL response upon the dose exhibits a linear

response in the range from 30 Gy to 10 kGy, with a correlation coefficient of 0.99. It is possible to

conclude that the a-LiAlSi2O6/Teflons has a great potential for OSL dosimetry of high gamma doses.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Luminescence applied to radiation dosimetry is used in differ-ent application fields. Recently the optically stimulated lumines-cence (OSL) became very popular and it is used in many areas ofthe applications of ionizing radiation. The OSL phenomenoncorresponds to the light emission of a given material, previouslyirradiated, during stimulation with photon beams of specificwavelengths. The light emitted by the material is proportionalto the received radiation dose. In optically stimulated lumines-cence (OSL) dosimetry the most used materials are quartz,feldspar and Al2O3:C (Yoshimura and Yukihara, 2006).

The OSL technique has different modes of stimulation and lightcollection. The main ones are the continuous and pulsed mode.In continuous mode, the sample is stimulated with a constantlight source and the light emitted by the sample is collected

ll rights reserved.

.P.O. d’Amorim),

s),

@gmail.com (H.J. Khoury),

simultaneously. In the pulsed mode, the sample is stimulatedwith light pulses and its luminescence is detected after thestimulation.

In this work the natural crystal of a-spodumene, lithiumaluminum silicate (LiAlSi2O6), was chosen because it is relativelyabundant in nature and present relatively high intensity ofluminescent emission. Spodumene is a pyroxene mineral consist-ing of lithium aluminium inosilicate, LiAl(SiO3)2, and is a source oflithium. The normal low-temperature form a-spodumene is in themonoclinic system whereas the high-temperature b-spodumenecrystallizes in the tetragonal system. Some of its thermolumines-cence (TL) properties are already studied for applications indating and ionizing radiation dosimetry (Brovetto et al., 1993;d’Amorim et al., 2012; Salis, 1995; Souza et al., 2004; Souza et al.,2007), but no studies discuss its OSL response. For this reason, theaim of this work is to study its OSL response and evaluate itsapplication for dosimetry.

2. Experimental

A natural crystal of a-spodumene from Minas Gerais, Brazil, wasinvestigated. After grinding the natural crystals in an agate mortar,

R.A.P.O. d’Amorim et al. / Radiation Physics and Chemistry 95 (2014) 141–144142

the grains between 75 and 180 mm were selected to prepare thea-spodumene/Teflons pellets, with 0.2 mm thick and 6.0 mm indiameter, using a mixture of a-spodumene and polytetrafluoroethy-lene (Teflons) (1:2) according to the procedure described byd’Amorim et al. (2012).

X ray diffraction (XRD) of natural grains of the a-spodumenewas performed to confirm its structure. The measurements wereperformed with a powder diffractometer PANalytical (XPert ProMPD), with CoKa radiation (l¼1.7889 A), with the tube operatingat 40 kV/30 mA in the continuous mode with steps of 21 min�1 atroom temperature. The identification of the crystalline phaseswas analyzed by the Rietveld refinement by the program FullProf(Young et al., 1995). The Rietveld refinement refers to adjustmentthe model parameters used in calculation of a diffraction patternwhich is closest to the observed (Will, 2005). The refinement iscarried out by minimizing the sum of the differences between thecalculated and observed intensities.

To study the OSL response, the pellets were irradiated withgamma source of Co-60 with a dose rate of 5.20 kGy/h in March/2012 and the response was measured using a custom made OSLReader, utilizing blue LEDs for the stimulation of the samples. Theresponse of this OSL reader was evaluated and compared to theRisø OSL reader. The OSL measurements were performed in twomodes of operation: continuous-wave (CW) and pulsed (POSL)stimulation.

To evaluate the sensibility of the OSL response of a batch of 23dosimeters, they were irradiated simultaneously with a dose of30 Gy. Five cycles of optical treatment-irradiation-reading wereperformed and then the mean value of readings (x) and therespective standard deviations (s) were calculated. The coefficientof variation (CV), also known as relative standard deviation wascalculated by the ratio of the standard deviation s to the mean m:

CV ¼sm ð1Þ

It shows the extent of variability in relation to mean of thepopulation.

3. Results and discussion

Fig. 1 presents the XRD pattern of the powder of a-LiAlSi2O6. TheXRD patterns were compared to experimental parameters referenceICSD 9668, being verified that the used sample corresponds to

Fig. 1. Y(obs) X-ray diffraction pattern from a sample of natural spodumene and

Y(cal) spectra calculated by Rietveld method. The result of refinement is the

difference Y(obs)�Y(calc).

a-spodumene. This XRD pattern also is described by d’Amorim et al.(2012). The a-spodumene belongs to the monoclinic crystal systemhaving the space group C2/c. The crystallographic parameters are:a¼9.4490 A, b¼8.3860 A, c¼5.2150 A, a¼901, b¼1101 and g¼901.

The result of the Rietveld refinement was obtained from the XRDanalysis and calculated by the least squares method and expressed asindex indicating disagreement, in percentage. This index is as valuesof permitted error (RP), obtained error (RWP), expected error (RE) andthe ratio RWP/RE (or, simply, S) (Young et al., 1995). The S value is ofcritical importance; S¼1 means that the calculated spectrum isperfectly adjusted to the experimental spectrum. Distortions ofquality measured during the sample preparation can lead to thesignificant value of high-RE, which is empirically considered excellentif below 10%.

Fig. 2 presents the results of the OSL response of the 23 pellets ofa-LiAlSi2O6/Teflons irradiated with the absorbed dose of 30 Gy. Eachpoint corresponds to the average of five measurements with theindication of the standard deviation. The solid line represents theoverall average of the measurements; the outermost dotted linescorrespond to the range (x72s) while the innermost represent therange (x71s). Considering these results, the pellets that presentedthe variation coefficient (CV) above 10% were removed from thegroup. This acceptance criterion is established by ISO 12794:2000standard (ISO, 2000), according to which the seven pellets wereremoved. Then the pellets that showed an average value out of therange of (x71s) were checked. In this case no pellets were discarded.Thus, the total number of selected pellets for the OSL study was 16.Their new average value and standard deviation were calculated. Thecorresponding CV was 2.1%.

Fig. 3 shows the OSL signal of a-LiAlSi2O6/Teflons irradiated withdifferent doses, obtained in the continuous stimulation mode during4000 s. The analysis of the OSL signal indicates that it is composed oftwo components: a fast (1), assigned to electrons that recombinedirectly with holes, and a slow (2), due to the presence of shallowtraps in the structure where electrons are retrapped for severalseconds before being recombined with holes.

Fig. 3 also shows that the decay rate becomes slower withincreasing dose. Yukihara et al. (2004) observed that the rate of OSLdecay curves increases with increasing dose in samples of Al2O3:C,irradiated with a beta source of Sr-90/Y-90. The opposite phenom-enon, observed here, may be explained by the different ionizationpower of gamma and beta radiation. Beta radiation produces locally alarge amount of ions in comparison to gamma radiation, correspond-ing to the component (1) of OSL curve. Thus, it is possible to concludethat the center responsible for component (1) is occupied firstly,becoming saturated in the region of the high ionization. Due to this

Fig. 2. OSL reproducibility of a-LiAlSi2O6/Teflons irradiated with absorbed dose

of 30 Gy with Co-60.

Fig. 3. OSL decay curves of a-LiAlSi2O6/Teflons for different absorbed doses of

Co-60, using the Continuous-wave stimulation mode (CW).

Fig. 4. OSL decay curves of a-LiAlSi2O6/Teflons for different absorbed doses of

Co-60, using the pulsed stimulation mode (POSL).

R.A.P.O. d’Amorim et al. / Radiation Physics and Chemistry 95 (2014) 141–144 143

saturation, the electrons begin to be captured by another center,which generates the component (2). The gamma radiation, however,produces ions in a much larger volume of the pellets compared to thebeta radiation of the same energy. During the irradiation, electronstend to be captured by traps that would be first filled, generating thecomponent (1). As ionization occurs sparsely owing to gammaradiation, the saturation of this kind of trap is not observed withsmaller doses, as occurs in the case of beta radiation. Thus, in thisstudy, the saturation of component of (1) occurs only at higher doses,and then the decay of the OSL curves decreases with increasing dosein the samples.

Through the OSL Pulsed mode showed in Fig. 4, the lowerdetection limit was obtained with irradiation of the 80 Gy. Thecurves of the samples irradiated with doses above 3 kGy exhibit aslight initial growth, followed by the decay of the signal. It can beseen that the decay of the OSL curves becomes increasinglysmaller with increasing dose.

Chen and McKeever (1997) explain the plurality behavior of OSLsignals by the presence of different types of traps. The initial growthof the signal is generally related to the presence of traps which arethermally unstable at the temperature at which the reading of theOSL is being performed. These traps capture the charge carriersinitially reducing the OSL intensity, but the number of chargescaptured by these traps and the number of charges released by lightstimulation reaches equilibrium. As the system approaches equili-brium, unstable traps decrease their participation in the competitionfor released charge carriers, resulting in a slight increase in OSLsignal. Reaching the equilibrium, the signal decays due to theemptying of optically active traps. This should be the same processthat is occurring with a-LiAlSi2O6/Teflons pellets irradiated withdoses above 3 kGy.

In the Pulsed mode, the luminescence of the sample is detectedafter the stimulation pulses have been applied. This evidences theinitial growth caused by the build-up of charges in a metastable level,

which can be either due to the excited state of the luminescencecenter or due to a shallow trap. Comparing the OSL curves on Figs. 3(CW) and 4 (POSL) corresponding to the doses above 3 kGy, the POSLevidences the contribution of thermally unstable traps to the lumi-nescence in the temperature of the OSL measurement. This is possibleto be checked only with the POSL because it allows discriminatingbetween light stimulation and light emission from the sample.

OSL response curves for a-LiAlSi2O6/Teflons pellets as afunction of absorbed dose can be observed in Fig. 5(a) for CWmode and (b) for POSL mode. Between 100 Gy and 10 kGy there isa range of linearity for both.

4. Conclusion

The results of the integrated OSL curves of the pellets irra-diated with the gamma dose of 30 Gy demonstrate that thereproducibility of response is within 72.1%. The results showthat the OSL response is linear over the dose range from 30 Gy to10 kGy with a correlation coefficient of 0.99. The results indicatethat the a-LiAlSi2O6/Teflons has a great potential for use in high-dose OSL gamma dosimetry.

Acknowledgements

The authors are thankful to the Brazilian Research SupportAgencies: Fundac- ~ao de Apoio �a Pesquisa e �a Inovac- ~ao Tecnologicade Sergipe (FAPITEC/SE), Coordenac- ~ao de Aperfeic-oamentode Pessoal de Nıvel Superior (CAPES), Conselho Nacional de

Fig. 5. OSL response of a-LiAlSi2O6/Teflons as function of absorbed dose of Co-60 source, (a) CW mode and (b) POSL mode.

R.A.P.O. d’Amorim et al. / Radiation Physics and Chemistry 95 (2014) 141–144144

Desenvolvimento Cientıfico e Tecnologico (CNPq), Fundac- ~ao deAmparo �a Ciencia e Tecnologia do Estado de Pernambuco(FACEPE), Instituto Nacional de Ciencia e Tecnologia de Metrolo-gia das Radiac- ~oes na Medicina (INCT/MRM), Instituto Nacional deCiencia e Tecnologia de Nanotecnologia para Marcadores Integra-dos (INCT/INAMI) and the Programa Nanobiotec-Brasil/CAPES(Rede 36).

References

Brovetto, P., Maxia, V., Salis, M., Spano, G., Lucco-Borlera, M., Mazza, D., 1993.Thermoluminescence experiments to study lattice defects in aluminosilicates.Nuovo Cimento Soc. Ital. Fis., D 15 (7), 1017–1022.

Chen, R., McKeever, S.W., 1997. Theory of Thermoluminescence and RelatedPhenomena. World Scientific, New Jersey.

d’Amorim, R.A.P.O., Teixeira, M.I., Souza, S.O., Sasaki, J.M, Caldas, L.V.E., 2012.Influence of Teflons agglutinator on TLD spodumene pellets. J. Lumin. 132,266–269.

ISO, 2000. International Organization for Standardization. Individual Thermolu-minescence Dosemeters for Extremities and Eyes. ISO 12794.

Salis, M., 1995. Lattice defects in natural a-spodumene. Nuovo Cimento Soc. Ital.Fis., D 17 (6), 649–651.

Souza, S.O., Ferraz, G.M., Watanabe, S., 2004. Effects of Mn and Fe impurities on theTL and EPR properties of artificial spodumene polycrystals under irradiation.Nucl. Instrum. Methods Phys. Res., Sect. B 218, 259–263.

Souza, S.O., Watanabe, S., Lima, A.F., Lalic, M., 2007. Thermoluminescent mechan-ism in Lilac Spodumene. Acta Phys. Pol. A 112, 1007–1012.

Will, G., 2005. Powder Diffraction: The Rietveld Method and the Two-StageMethod, first ed. Springer, Sturtz, Wurzburg.

Yoshimura, E.M., Yukihara, E.G., 2006. Optically stimulated luminescence: search-ing for new dosimetric materials. Nucl. Instrum. Methods Phys. Res., Sect. B

250, 337–341.Young, R.A., Sakthivel, A., Moss, T.S., Paiva-Santos, C.O., 1995. DBWS-9411—an

upgrade of the DBWS programs for rietveld refinement with PC and main-frame computers. J. Appl. Cryst. 28, 366–367.

Yukihara, E.G., Whitley, V.W., McKeever, S.W.S., Akselrod, A.E., Akselrod, M.S.,2004. Effect of high-dose irradiation on the optically stimulated luminescenceof Al2O3:C. Radiat. Meas. 38, 317–330.