Systematic development of new thermoluminescence and optically stimulated luminescence materials

Journal of Luminescence 133 (2013) 203–210

Contents lists available at SciVerse ScienceDirect

Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin

Systematic development of new thermoluminescence and optically stimulated luminescence materials E.G. Yukihara a,n, E.D. Milliken a, L.C. Oliveira a, V.R. Orante-Barro´n b, L.G. Jacobsohn c, M.W. Blair d a

Physics Department, 145 Physical Sciences II, Oklahoma State University, Stillwater, OK 74078, USA ´n en Polı´meros y Materiales, Universidad de Sonora, Hermosillo, Sonora 83000, Me ´xico Departamento de Investigacio c Center for Optical Materials Science and Engineering Technologies (COMSET), and School of Materials Science and Engineering, Clemson University, Clemson, SC, USA d Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM 87545, USA b

a r t i c l e i n f o

abstract

Available online 13 December 2011

This paper presents an overview of a systematic study to develop new thermoluminescence (TL) and optically stimulated luminescence (OSL) materials using solution combustion synthesis (SCS) for applications such as personal OSL dosimetry, 2D dose mapping, and temperature sensing. A discussion on the material requirements for these applications is included. We present X-ray diffraction (XRD) data on single phase materials obtained with SCS, as well as radioluminescence (RL), TL and OSL data of lanthanide-doped materials. The results demonstrate the possibility of producing TL and OSL materials with sensitivity similar to or approaching those of commercial TL and OSL materials used in dosimetry (e.g., LiF:Mg,Ti and Al2O3:C) using SCS. The results also show that the luminescence properties can be improved by Li co-doping and annealing. The presence of an atypical TL background and anomalous fading are discussed and deserve attention in future investigations. We hope that these preliminary results on the use of SCS for production of TL and OSL materials are helpful to guide future efforts towards the development of new luminescence materials for different applications. & 2011 Elsevier B.V. All rights reserved.

Keywords: Thermoluminescence Optically Stimulated Luminescence Solution Combustion Synthesis Radioluminescence

1. Introduction Thermoluminescence (TL) and Optically Stimulated Luminescence (OSL), also called Photo-Stimulated Luminescence (PSL), are techniques widely used in radiation dosimetry [1–4], luminescence dating [1,5,6], and computed radiography [7]. They rely on the stimulated recombination of electrons and holes created by ionizing radiation and trapped at defects in the crystalline lattice of the host material, leading to luminescence whose intensity is related to the energy deposited in the detector by ionizing radiation (i.e., absorbed dose). In TL the stimulation is provided by controlled heating of the detector [3,4]. In OSL, stimulation is provided by controlled illumination [1,2]. In spite of the widespread use of TL and OSL, a demand exists for new materials with tailored properties for specific applications, including OSL neutron dosimetry, 2D dose mapping and temperature sensing, as discussed below. There are a limited number of OSL materials for personal dosimetry application, particularly for neutron dosimetry. Only two materials are commercially used in OSL dosimetry, Al2O3:C and BeO, and this limited availability has been pointed out as a

n

Corresponding author. Tel.: þ1 405 744 6535; fax: þ1 405 744 1112. E-mail address: [email protected] (E.G. Yukihara).

0022-2313/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jlumin.2011.12.018

weak point of the OSL technique [2,8]. Moreover, these materials do not have a high cross-section for neutron interaction, which means that they cannot be used as neutron detectors [8]. This problem has been partially solved by preparing detectors made of a mixture of OSL material and neutron converters [9,10] such as 6 Li or 10B, which convert neutrons into charged particles [11]. Although this solution is commercially satisfactory [12], higher neutron sensitivities could be achieved using new OSL materials containing 6Li or 10B as part of the crystalline structure, which would require the development of new OSL materials based on compounds such as Li2B4O7 or MgB4O7. Two-dimensional dose mapping in medical dosimetry, particularly in quality assurance and dose verification in radiotherapy, is another area of potential application of OSL materials. Although 2D dosimetry has been performed using TL [13–16], an all-optical technique such as OSL would be a better choice for this type of application, as evidenced by the use of the OSL in computed radiography [7]. The main problem with using photostimulable phosphors used in computed radiography, such as BaXBr (X¼F, Cl, Br) and CsBr, for dosimetry is their high effective atomic number [17,18] (Zeff  50) and signal fading ( 450% in 36 h) [7,19]. One-dimensional dose mapping using Al2O3:C OSL detectors has been used in computed tomography [20–22], but the luminescence lifetime of the main luminescence centers in Al2O3:C (35 ms) is too long for 2D dosimetry readout by spot-scanning

204

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

laser. OSL systems based on BeO or SrS have been described [23,24], but these systems present problems such as limited spatial resolution or high effective atomic number of the detector material (e.g. Zeff ¼34.6 in the case of SrS) [2]. More recently, renewed interest has been expressed in the use of TL for temperature sensing, in particular as passive temperature sensors in biological agent defeat tests [25], but the lack of suitable materials is also one of the main obstacles. The concept is based on the fact that charges trapped at different energy levels within the conduction band are affected differently by the temperature experienced by the particles, and this can be quantified by measuring the TL curves of particles previously irradiated: depending on the time–temperature profile, the TL peaks would be erased differently. For this application, materials with multiple TL peaks that are light-insensitive are required. Unfortunately, a survey of existing TL materials reveals most of them to be light sensitive [3]. LiF:Mg,Ti is an exception, but this material is known for having a complex defect structure, which causes the TL properties to be dependent on the entire temperature history, both before and after irradiation [3]. Because of the complex nature of the TL and OSL processes, which require the presence of both recombination and trapping centers introduced by intrinsic or extrinsic defects in the host material, development of new materials has been serendipitous. Often, the nature of the recombination centers is known because of its characteristic emission spectrum, but that of trapping centers responsible for the TL/OSL signal is not. Recently, two developments increased the chances of more precisely engineering the TL and OSL properties of materials. The first development is the demonstration that chemical routes such as solution combustion synthesis (SCS) [26–29] may offer a more efficient way to synthesize TL/OSL materials [30–33] and investigate the role of dopants in the TL and OSL process. The second is the understanding that the energy levels introduced by lanthanide (Ln) dopants and their role in the TL (and possibly OSL) process can be predicted based on a few parameters [34–37]. Based on these developments and motivated by the lack of suitable TL and OSL materials for different applications, we initiated a systematic study to develop new TL and OSL materials with properties tailored for the specific applications discussed above. Our approach uses SCS as the main synthesis method, accompanied by

characterization of the crystal structure and luminescence properties of the materials produced. The objective of this work is to present an overview of these efforts by showing the range of materials synthesized by SCS, typical radioluminescence (RL) spectra to show the incorporation of luminescence centers, as well as TL and OSL of some of the samples that exhibited high sensitivity to ionizing radiation. We also discuss the effect of Li co-doping in the RL and TL properties and some unexpected results related to background of the TL measurements and fading. This work does not intend to be an exhaustive study of any single material; see for example references [38,39]. Instead, we focus on general observations that we hope can be useful for other investigators working on the development of new TL and OSL materials.

2. Material requirements Table 1 summarizes the most important requirements for the specific applications discussed above. In all cases, it is expected that the trapped charge population is stable at room temperature. In personal OSL dosimetry, additional requirements include a light sensitive trapped charge population, emission in the blueUV range of the spectrum, tissue equivalency, and predominance of single trapping centers. Emission in the blue-UV range of the spectrum allows for detection of light at shorter wavelengths than stimulation (blue or green), in addition to being a better match for the spectral response of photomultiplier tubes (PMTs). In OSL dosimetry, emission in wavelengths shorter than the stimulation wavelength makes it easier to separate between the stimulation light and the OSL emission using optical filters [2]. Tissue equivalency means that the host material has an effective atomic number similar to water or tissue (Zeff  7.5–7.6), so that the detector has a response with dependence on photon energy similar to the material of interest [17]. Predominance of a single trapping center means that that the signal is not associated with overlapping components with different dosimetric properties (e.g., thermal stability). Moreover, luminescence centers characterized by radiation transitions with long lifetime are useful because of the possibility of increasing the signal-to-noise ratio using a time-resolved luminescence technique called pulsed OSL

Table 1 Examples of desirable properties for new TL/OSL materials for different applications. Application

Desirable properties

All

 Trapped charge population stable at room temperature

Personal OSL dosimetry

     

Trapped charge population sensitive to light Emission in the blue-UV region Tissue equivalency (Zeff  7.5) Single trapping center associated with the OSL signal Long luminescence lifetime (4100 ms) in case of POSL applications Intrinsic neutron sensitivity, i.e. having Li or B in its composition (for neutron dosimetry)

2D OSL dosimetry

     

Trapped charge population sensitive to light Short luminescence lifetime (o 100 ms) Emission in the blue-UV region Tissue equivalency (Zeff  7.5) Single trapping center associated with OSL signal Small grain sizes ( mm or less)

Temperature sensing (TL)

 Multiple TL peaks over a wide range of temperatures  Simple TL kinetics (first order)  Trapped charge population insensitive to light

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

(POSL), in which the optical stimulation and the luminescence detection occur asynchronously [40]. For 2D OSL dosimetry, the materials need to have an OSL resultant from luminescence centers characterized by short luminescence lifetime, so that it is feasible to read a twodimensional detector in a reasonable period. In comparison with computed radiography, applications in 2D dosimetry are less stringent in terms of sensitivity and resolution. Most of the applications would be in quality control for radiotherapy, so the doses involved are high and there is no patient involved. Also, dose information with spatial resolution higher than 0.1 mm is hardly justifiable. On the other hand, requirements in terms of precision and accuracy would be higher, since one is interested in absolute or relative dose measurements as a function of position. To achieve that, OSL materials should have a low effective atomic number and an OSL signal stable at room temperature to reduce the need for correction factors. For temperature sensing applications, effective atomic number is not a constraint, but the materials should have multiple trapping centers that are light insensitive (and if possible characterized by simple recombination kinetics).

3. Materials and methods Table 2 shows the samples obtained by SCS in this work. Because of the range of applications discussed above, we focused on wide band-gap materials with a range of effective atomic numbers. The materials were prepared using oxidizers and fuels combined to obtain an elemental stoichiometric coefficient of unity (fe ¼ 1) [41]. Typical quantities and the corresponding volume of purified water (Type I, Milli-Q, Millipore Corporation, Billerica, MA, USA) are indicated in Table 2. The dopants were introduced as nitrates, and their concentrations refer to the molar percentage relative to the metal nitrate of the host with the potential to be

205

substituted. The borates were prepared with a 35% excess of boric acid to account for losses during combustion and annealing. All reagents are ACS grade obtained from Alfa–Aesar (Ward Hill, MA, USA) or Sigma–Aldrich (Sigma-Aldrich Co, LLC, St. Louis, MO, USA). The aqueous mixture of reagents was dried at 200 1C for  1.5 h on a hot plate inside a fume hood. The temperature was then increased to 500 1C, causing the mixture to undergo combustion after a few minutes. The resultant powder was crushed using an agate mortar and pestle and placed in alumina crucibles for annealing. The samples were annealed in a temperature controlled tube furnace (Marshall model 1123, ThermCraft Inc., Winston Salem, NC, USA) or muffle furnace (Omegalux LMF-3550, Omega Engineering, Inc., Stamford, CT, USA) at temperatures up to 1100 1C for up to 10 h, depending on the sample. The postcombusted and annealed powder was then crushed again using an agate mortar and pestle. Other materials were used for comparison of the TL, OSL and RL properties. For TL we used commercial LiF:Mg,Ti (TLD-100, Thermo Fisher Scientific, Inc., Franklin, MA, USA), for OSL we used commercial Al2O3:C (Landauer, Inc., Glenwood, IL, USA), and for RL we used the scintillators Lu2SiO5:Ce (LSO, Single Crystal Growth Laboratory, Los Alamos National Laboratory, Los Alamos, NM, USA) and Gd2SiO5:C (GSO, Hitachi Chemical Co., Ltd., ) (see [42]). The crystalline structures of the samples were characterized by X-ray diffraction (XRD) using a Phillips Analytical X-ray diffractometer (model PW3020) with CuKa radiation and scanning the 2y in 0.02 degree step size and 0.5 step time. RL spectra were obtained by exciting the samples with a 40 kV X-ray tube (MagnumTM, Ag transmission target, Moxtek Inc.) delivering a dose rate of approximately 150 mGy/s at the sample position, and detecting the luminescence using a CCD fiber spectrometer (model USB-2000, Ocean Optics Inc., Dunedin, FL, USA) via an optical fiber (1 mm core diameter, transmission between 200–1100 nm). Each RL spectrum was measured using  10 mg of powder deposited in stainless steel cups. The spectra

Table 2 Materials produced by SCS, bandgap energy, effective atomic number, typical reagent quantities and solution volume, and dopants investigated. The effective atomic number was calculated as in Bos [18]. Host

Eg (eV)

Zeff

Reagents

ZrO2 Y2O3

5 5.6

36.3 36.1

MgAl2O4

5.8

11.2

Y3Al5O12

7.5

30.6

11.9 g Zr(NO3)2  6H2O; 3.5 g urea, 50 ml Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb 7.6 g Y(NO3)3  6H2O, 3 g urea, 50 ml Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, (Ce,Sm), (Ce,Dy), (Ce,Tm), (Eu,Sm), (Eu,Dy), (Eu,Tm), (Tb,Sm), (Tb,Dy), (Tb,Tm), (Tb,Eu) 7.8 g Mg(NO3)2  6H2O, 23.2 g Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb Al(NO3)3  9H2O, 12.2 g urea, 50 ml 8.1 g Y(NO3)3  6H2O, 13.5 g Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, (Ce,Sm), (Ce,Eu), (Ce,Yb), (Tb:Sm), (Tb,Eu) , Al(NO3)3  9H2O, 8.6 g urea, 50 ml (Tb,Yb), (Ce,Eu,Yb), (Tb,Eu,Yb) Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Li 4.3 g La(NO3)3  6H2O, 2.7 g Mg(NO3)2  6H2O, 3.2 g H3BO3; 2.1 g glycine, 100 ml 16.7 g Ca(NO3)2  4H2O, 7.1 g urea, 50 ml Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Li 5.2 g Mg(NO3)2  6H2O, 6.7 g H3BO3, 2.0 g Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Tm, Yb , (Ce, Pr), (Ce, Nd), (Ce, Dy), (Ce, Tm), (Tb, Pr), (Tb, urea, 100 ml Nd), (Tb, Dy), (Tb, Tm), Li, Na, K 15.5 g Al(NO3)3  9H2O, 6.1 g urea, 50 ml Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Li ,Zn, Zr, Si, Ag, Fe, Mn, Ni, Cu, (Zr,Si), (Zr,P), (Zr, Tb), Li 0.74 g Ca(NO3)2  4H2O, 14.1 g Ce, Eu, Li Al(NO3)2  9H2O, 6.0 g urea, 50 ml 7.0 g LiNO3, 12.6 g H3BO3, 8.0 g NH4NO3, Dy, Ce, Mn, Cu, Ag, (Cu, Ag), Ni, Cr 5.71 glycine, 50 ml 3.1 g Ca(NO3)2  4H2O, 10.0 g Ce, Eu, Li Al(NO3)3  9H2O, 5.4 g urea, 50 ml 13.1 g Mg(NO3)2  6H2O, 5.1 g urea, 50 ml Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Li, Al, (Ce,Gd), (Nd,Gd), (Ce,Ca), (Ce, La), (Ce, Dy), (Ce, Eu), (Ce, Er), (Ce, Tm), (Ce, Yb), ( Ce, Tb), La , Ca, Na, Ba, K, Al , Fe, Cr, Mn, P, Si, Co, Zn, Zr, In, (In,Gd), (In, Nd), Ti , Cu, Ag, (Cu, Ag), Ni, (Ce:Gd), (Nd:Gd), (Nd, Ho), (Nd, Tm), (Ce,Ca), (Fe, Mn), (Nd, Ho), (Nd, Dy), (Nd, Gd), (Nd,Tm), (Nd, Ho), (Nd, Er) 7.5 g Al(NO3)3  9H2O, 5.2 g urea, 1.4 g Ce LiNO3, 50 ml

LaMgB5O10

41.5

CaO MgB4O7

7.1

18.3 8.5

Al2O3

9.5

11.3

CaAl12O19

12.4

Li2B4O7

7.5

7.3

CaAl2O4

7.4

14.8

MgO

7.8

10.8

LiAlO2

10.7

Dopants investigated (co-doping indicated in parenthesis)

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

all dopants are incorporated into the MgO lattice [38]. However, we have not carried out extensive XRD investigations of doped samples yet. The XRD patterns of undoped ZrO and LaMgB5O10 presented a mixture of phases and are not shown here. RL measurements demonstrate the incorporation of the dopants as luminescence centers in the materials. Fig. 2 shows the RL emission spectra for three lanthanide-doped compounds which displayed the strongest RL intensities. The main characteristic emission lines from the trivalent lanthanides can be observed, although the intensities varied with the host material. The emission band from Ce3 þ is also observed in CaO and MgO. The RL intensity for various lanthanide materials are compared in Table 3. TL curves with a variety of shapes and peaks located at temperatures in the dosimetric range were observed from these samples, some of them with intensity comparable to or higher than commercial TL materials. Fig. 3 compares the curves for materials which exhibit strong TL with that from LiF:Mg,Ti. It is worth mentioning that for shape of the TL curve of MgO:Ln, Li and MgB4O7:Ln, Li changes depending on the type of lanthanide used for doping. In the case of CaO, the most intense TL ( 4106 counts per 0.2 s) was observed for undoped samples, with doping generally decreasing the TL intensity (results not shown here). At this point concentration quenching curves for RL and TL were obtained for MgO:Ce,Li only [38]. OSL investigations have been focused on materials with low effective atomic number, from which MgO showed the best results. Examples of OSL curves for MgO:Ln1%,Li3% and various dopants are compared in Fig. 4 with the OSL from commercial

(511)

(400)

(422) (006)

*

50

(201)

60

*

(018)

(024)

(110)

(107)

40

(217)

30

(205)

20

(102)

(104)

60

*

(116)

50

(110)

(513) (702) (810)

40

(113)

30

(012)

20

(220)

(311)

(611) (622)

60

(111)

(220)

(311)

(200) (111)

*

40

(210) (211) (020) (311) (220) (302) (321) (402) (421) (512) (232)

Intensity (arb. units)

* 20

(440)

(222) (400) (411) (332) (134)

(211)

Intensity (arb. units)

Fig. 1 shows the XRD pattern for various undoped samples produced by SCS, demonstrating that the materials are single phase. Data on MgO have been presented elsewhere [38]. These results assure we obtained the desired host lattice. In the case of doped MgO:Ce,Li, XRD reveals an additional CeO2 phase, indicating that not

(206)

4. Results and discussion

(203)

were not corrected for the response of the system, which peaks at 500 nm and reaches 10% efficiency at 250 nm and 730 nm [38]. TL and OSL measurements were carried out using a Risø TL/OSL reader (model TL/OSL-DA-15, Risø National Laboratory, Røskilde, Denmark). The TL or OSL signals were detected using a PMT (model 9235QB, Electron Tubes, Inc.). For the TL measurements we used Schott BG-39 filters (6 mm thickness, transmission between  340 and 610 nm, Schott AG, Mainz, Germany) in front of the PMT. The samples were heated at 5 1C/s in high purity nitrogen gas atmosphere. For OSL measurements, the samples were stimulated with blue LEDs (centered at 470 nm, irradiance of  30 mW/cm2) using Hoya U-340 filters (7.5 mm thickness, transmission between 290 and 370 nm, Hoya Corporation USA, Santa Clara, CA, USA) in front of the PMT. For TL and OSL measurements, the samples were irradiated with  0.5 Gy using a 90Sr/90Y beta source. More details on the Risø readers can be found in Bøtter-Jensen et al. [43] and references therein. All RL, TL and OSL data were obtained using 10 mg of powder in stainless steel cups.

(114)

206

20

60

20

30

40 50 2θ (deg.)

60

20

30

40

50

60

40 50 2θ (deg.)

(302)

30

(310) (222)

(444) (640) (642) (800)

(611)

(422) (521)

(420)

40 50 2θ (deg.)

(321) (400)

(211)

(512)

(204) (224)

(312) (004)

30

*

20

(220) (113)

60

(200)

40

(102)

20

(201) (211)

60

(101)

50

(110) (111)

40

(112)

30

(211) (202)

(200)

Intensity (arb. units)

20

60

Fig. 1. XRD patterns of samples produced by SCS, accompanied by information on annealing temperature and duration, powder diffraction card number, crystal system and space group: (a) Y2O3 (1100 1C for 2 h, 01-083-0927, cubic, Ia-3), (b) CaO (900 1C for 2 h, 01-077-2376, cubic, Fm-3m), (c) MgAl2O4 (900 1C for 2 h, 01–071-6329, cubic, Fd-3m), (d) MgB4O7 (900 1C for 2 h, 00-017-0927, orthorhombic, Pbca), (e) Al2O3 (900 1C for 2 h, 00-042-1468, rhombohedral, R-3c), (f) CaAl12O19 (1200 1C for 4 h, 00-0380470, hexagonal, P63/mmc), (g) LiB4O7 (860 1C for 40 min, 01–084-2191, tetragonal, 141cd), (h) Y3Al5O12 (900 1C for 2 h, 01-071-1853, cubic, Ia-3d), (i) LiAlO2 (1200 1C for 4 h, 00–038-1464, tetragonal, P4212). Miller indices are presented only for the most intense peaks. The asterisk (n) indicates an artifact introduced by the sample holder.

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

260

1300

650

1200

600

1100

550

1000

500

900

450

800

400

700

350

600

300

500

250

400

200

300

150

200

100

100

50

207

Yb 240 220 200

RL intensity (cps)

180 160

Tm Er Ho Dy Tb

140 Gd 120 Eu 100 Sm 80 60

Nd Pr

40 20

Ce undoped

0 200

400

600

800

0 200

400 600 800 Wavelength (nm)

0 200

400

600

800

Fig. 2. RL spectra from (a) MgO:Li3%, (b) CaO and (c) Y2O3 undoped or doped with different lanthanides. The spectra are offset vertically for better visualization. For comparison, the maximum intensity from LSO:Ce and YSO:Ce scintillators in powder measured in the same conditions are  140 cps and 180 cps, respectively. The spectra were not corrected for the detection response of the system [38]. Table 3 RL intensities for various lanthanide-doped materials produced by SCS; the values are in counts per second and correspond to the maximum intensity of the indicated emission band. All lanthanide concentrations are 0.1%, except when indicated otherwise. The data were obtained using powder (10 mg) and identical measurement conditions. The data are only for qualitative comparison, since the x-ray energy deposited in different materials varies with the mass energy absorption coefficients of each material [17] and the spectra were not corrected for the detection response of the system [38]. The position of the lines varies depending on the material.

Al2O3 CaO MgB4O7:Li1% LaMgB5O10 Y2O3 Y3Al5O12 MgO:Li3% (Ln1%)

Ce

Pr  620 nm

Nd  396 nm

Sm  600 nm

Eu  600 nm

Gd 313 nm

Tb 544 nm

Dy  570 nm

Ho 548 nm

Er  407 nm

Tm 454 nm

30a 80 o1 o1 o1 70 80

o1 60 o1 1 50 25 1

13 o1 o1 o1 o1 o 20 20

13 210 4 25 70 60 18

30 100 1 o1 140 60 60

10 90 2 1 50 30 6

100 250 17 14 175 35 6

25 390 35 35 250 o 20 16

o1 25 o1 o1 o 20 o1 o1

6 o 20 o1 o1 o 20 o1 o1

13 80 14 22 o 20 o1 6

‘‘o 1’’ indicates undetected or weak emission. a

The emission is probably due to F centers, since undoped samples also show similar emission band.

Al2O3:C. MgO samples which presented the most intense OSL signals are Gd- , Nd- and Tm-doped samples, in which the emission from the respective trivalent lanthanides occurs in the transmission range of the optical filters (  290–370 nm) used in the OSL measurements (see Fig. 2a). For some samples, Li co-doping substantially improved the RL or TL intensities, or both. Fig. 5 shows the effect of Li co-doping on the RL and TL of MgO doped with Dy (Fig. 5a) and Eu (Fig. 5b). Note that Li co-doping not only increases the intensity of the RL and TL, but also changes the structure of the TL curve in comparison to the undoped samples or samples doped only with lanthanide. Improved photoluminescence properties due to the effect of Li have been observed before for MgO [44,45]. We speculate that Li may be acting as a charge compensator: Li þ substituting for Mg2 þ creates a defect with net negative charge (LiMg)  that may compensate for the

incorporation of the trivalent lanthanide in the Mg site, which creates a defect with positive effective charge (LnMg) þ . Therefore, in the presence of Li, the incorporation of trivalent lanthanide into the crystal lattice may be more effective. Moreover, the ionic radius ˚ and Mg2 þ (0.86A) ˚ are similar, therefore favoring the of Li þ (0.90A) substitution. However, changes in TL curves with Li co-doping show that Li also introduces or favors the formation of other defects acting as trapping centers. It should be pointed out that Li co-doping did not result in a consistent improvement in the luminescence properties of CaO. Fig. 6 shows the effect of Li co-doping on Ce-doped MgB4O7. In this material, Li co-doping increased the TL intensity, but did not affect the RL intensity from the trivalent lanthanides. Annealing was also observed to improve the RL and TL of some samples, particularly in the case of Y3Al5O12 and MgO. This is

208

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

0.4 1.0

Sm (1100°C/6h) 0.3

0.8 Nd (0.5%) 0.6

0.2 Ce

TL intensity (106 counts per 0.2s)

0.4

0.1 0.2 0.0

0.0 100

200

300

400

100

200

300

400

0.0 400 100 Temperature (°C)

200

300

400

5

1.0

Dy Tm

4

0.8

3

0.6

2

0.4 Tb

1

0.2 Ce

0 100

200

300

Fig. 3. TL curves for different lanthanide(Ln)-doped compounds synthesized by SCS: (a) MgO; (b) Li2B4O7 and (c) MgB4O7. Panel (d) shows the TL from commercial LiF:Mg,Ti (TLD-100) for comparison.

OSL intensity (106 counts per 0.2 s)

3.0 2.5 Al2O3:C 2.0 MgO:Gd1%,Li3%

1.5

MgO:Nd1%,Li3%

1.0

MgO:Tm1%,Li3% 0.5

MgO:Sm1%,Li3%

0.0 0

10

20

30 Time (s)

40

50

60

Fig. 4. OSL curves of MgO samples synthesized by SCS compared to the OSL curve of commercial Al2O3:C measured in the same conditions. It should be pointed out the experimental conditions used here are not optimum for Al2O3:C. Based on the emission spectrum of Al2O3:C and transmittance spectra of optical filters, we estimate that OSL measurements of Al2O3:C using Hoya U-340 filters (7.5 mm thickness) are 70% lower than identical measurements using filters centered at the Al2O3:C emission band (Kopp 5113, 8 mm thickness).

exemplified in Fig. 7 for Y3Al5O12:Ce,Yb, showing that the TL peak at 200 1C increases as the annealing temperature is increased from 900 1C to 1100 1C, whereas high temperature peaks responsible for TL above 300 1C are reduced. Other investigators also emphasized the importance of annealing in improving the luminescence properties of materials produced by SCS [46,47]. In the case of MgO:Ce,Li, annealing to increasing temperatures improve the overall RL and TL intensities and change the relative intensities of the TL peaks [38].

In spite of the promising results obtained using SCS for the development of new TL and OSL materials, a few observations deserve attention. In MgAl2O4 we observed an atypical TL background, as exemplified in Fig. 8. The TL curves in Fig. 8 were obtained using un-irradiated samples, after the samples have already been annealed to 900 1C for 2 h in a furnace and heated once to 450 1C at 5 1C/s in the TL reader. The TL curves should exhibit a low background (  300–500 cps) characteristic of PMT dark counts, increasing above  400 1C due to blackbody radiation, as illustrated Fig. 8 for a MgB4O7:Ce,Li sample. It is not clear whether this atypical emission in MgAl2O4 is due to incomplete reaction during the combustion process or annealing that is not optimized. In any case, it is worth determining whether or not this effect is related to the material synthesis technique. Another important observation is the presence of anomalous fading in some samples. This is exemplified in Fig. 9 for Y3Al5O12: Ce,Yb. This sample presents a TL peak at 200 1C which should be relatively stable at room temperature, but which exhibits a substantial decrease in intensity even after a short interval following irradiation (2 h). Other studies have shown that, in YAG prepared by co-precipitation, the rate of fading increases with the concentration of Ce and Yb, leading to the suggestion that the anomalous fading is caused by tunneling between the Yb2 þ (trapping center) and the Ce4 þ (recombination center) [37]. It would also be important to understand whether this anomalous fading is restricted to Y3Al5O12 or related to the material synthesis technique.

5. Conclusions This study demonstrates the possibility of producing TL and OSL materials with sensitivity similar to or approaching those of commercial TL and OSL materials used in dosimetry (e.g., LiF:Mg,Ti and Al2O3:C) using SCS. Of equal importance, the SCS method offers an efficient way for testing the influence of

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

5

16

YAG: Ce0.1%,Yb0.1%

Dy1%, Li3%

RL intensity (cps)

12 10 8 6 4 Dy1%

0.8

Dy1%, Li3%

TL intensity (105 counts/0.2s)

TL intensity (105 counts per 0.2s)

14

0.6

0.4 undoped Dy1% 0.2

2

4

1100°C/2h 1100°C/10h

3

2 900°C/2h 1

0.0

0

100 200 300 Temperature (ºC)

400 500 600 700 800 Wavelength (nm)

400

0 100

200 300 Temperature (°C)

1.5

40 30 20 Eu1%

Fig. 7. Effect of annealing on the TL of yttrium aluminum garnet (YAG), Y3Al5O12, produced by SCS.

Eu1%, Li3%

4

1.0

MgAl2O4:Ln0.1%

0.5

undoped Eu1%

0.0

0 400 500 600 700 800 Wavelength (nm)

300 100 200 Temperature (ºC)

400

Fig. 5. Effect of Li co-doping on the RL and TL properties of (a) MgO:Dy and (b) MgO:Eu. The data were obtained using 10 mg of powder in identical conditions. The samples were annealed at 900 1C for 2 h.

TL intensity (104 counts per 0.2s)

RL intensity (cps)

50

10

400

Eu1%,Li3% TL intensity (105 counts per 0.2s)

60

209

Tb 3 Gd Eu Sm

2

Pr Nd 1

Ce MgB4O7:Ce0.1%, Li1%

0 100

TL intensity (105 counts/0.2s)

8

Ce0.1% Li1% MgB4O7

400

Fig. 8. A typical background emission from un-irradiated MgAlO4 samples previously annealed at 900 1C for 2 h and then heated to 450 1C at 5 1C/s, compared with a normal background from MgB4O7:Ce0.1%, Li1%, both prepared by SCS.

6 Ce0.1% Li0.1% 4

2 Ce0.1% undoped 0 100

200 300 Temperature (°C)

200 Temperature (°C)

300

400

Fig. 6. Effect of Li co-doping on the TL of MgB4O7. The samples were annealed at 900 1C for 2 h.

different dopants, allowing the introduction of luminescence centers with emission in the wavelength of interest for different applications. The luminescence properties can be improved by Li co-doping and annealing at an appropriate temperature. Atypical or anomalous effects (background and fading) have been observed in some samples and deserve more attention in future investigations. This study represents a new paradigm in TL/OSL research whereby new materials can be discovered and designed by systematic investigation rather than by serendipity. More in-depth studies focused on specific materials are required to further develop useful TL and OSL materials. There are a large number of synthesis parameters to be investigated, including dopant concentrations, fuel–oxidizer ratios, annealing temperatures, codopants, and so on. Nevertheless, we hope these preliminary results on the use of SCS for production of TL and OSL materials are helpful to guide future efforts towards the development of new materials needed for the applications discussed here.

210

E.G. Yukihara et al. / Journal of Luminescence 133 (2013) 203–210

4 YAG:Ce0.1%,Yb0.1%



TL (105 counts per 0.2s)

3

2  + 2h

1

0 100

200 300 Temperature (°C)

400

Fig. 9. Example of anomalous fading for yttrium aluminum garnet (YAG) Y3Al5O12 irradiated with 0.5 Gy of beta radiation, showing the TL curve immediately after irradiation, with or without a 2 h period in the dark before TL readout. The samples were annealed at 1100 1C for 2 h.

Acknowledgements The authors thank Jim Puckette (Boone Pickens School of Geology, Oklahoma State University, Stillwater, OK) for the use of the Phillips Analytical X-ray diffractometer, and Gregoire Denis for discussions and suggestions. This work was supported by the Oklahoma Center for the Advancement of Science and Technology (OCAST) through OHRS award project number HR09-104, and by the US Defense Threat Reduction Agency (DTRA) through contract HDTRA1-10-1-0007. References [1] L. Bøtter-Jensen, S.W.S. McKeever, A.G. Wintle, Optically Stimulated Luminescence Dosimetry, Elsevier, Amsterdam, 2003. [2] E.G. Yukihara, S.W.S. McKeever, Optically Stimulated Luminescence: Fundamentals and Applications, John Wiley & Sons, Chichester, West Sussex, UK, 2011. [3] S.W.S. McKeever, M. Moscovitch, P.D. Townsend, Thermoluminescence Dosimetry Materials: Properties and Uses, Nuclear Technology Publishing, Ashford, 1995. [4] R. Chen, S.W.S. McKeever, Theory of Thermoluminescence and Related Phenomena, World Scientific Publishing Co., Singapore, 1997. [5] M.J. Aitken, Thermoluminescence Dating, Academic Press, Orlando, 1985. [6] M.J. Aitken, An Introduction to Optical Dating, Oxford University Press, Oxford, 1998. [7] P. Leblans, D. Vandenbroucke, P. Willems, Materials 4 (2011) 1034.

[8] S.W.S. McKeever, M. Moscovitch, Radiat. Prot. Dosim. 104 (2003) 263. [9] J.C.R. Mittani, A.A.R. d. Silva, F. Vanhavere, M.S. Akselrod, E.G. Yukihara, Nucl. Instrum Methods., Phys. Res. B 260 (2007) 663. [10] E.G. Yukihara, J.C.R. Mittani, F. Vanhavere, M.S. Akselrod, Radiat. Meas. 43 (2008) 309. [11] G.F. Knoll, Radiation Detection and Measurements, John Wiley & Sons, Chichester, West Sussex, UK, 2000. [12] C. Passmore, M. Kirr, Radiat. Prot. Dosim. 144 (2011) 155. [13] B. Marczewska, P. Bilski, P. Olko, M.P.R. Waligo´rski, Radiat. Meas 38 (2004) 833. [14] L. Czopyk, M. Klosowski, P. Olko, J. Swakon, M.P.R. Waligorski, T. Kajdrowicz, G. Cuttone, G.A.P. Cirrone, F. Di Rosa, Radiat. Prot. Dosim 126 (2007) 185. [15] B. Marczewska, P. Bilski, L. Czopyk, P. Olko, M.P.R. Waligo´rski, S. Zapotoczny, Radiat. Prot. Dosim 120 (2006) 129. [16] P. Olko, B. Marczewska, L. Czopyk, M.A. Czermak, M. K"osowski, M.P.R. Waligo´rski, Radiat. Prot. Dosim 118 (2006) 213. [17] F.H. Attix, Introduction to Radiological Physics and Radiation Dosimetry, Wiley-VCH, Weinheim, 2004. [18] A.J.J. Bos, Nucl. Instrum Methods. Phys. Res. B 184 (2001) 3. [19] E. Ariga, S. Ito, S. Deji, T. Saze, K. Nishizawa, Med. Phys. 34 (2007) 166. [20] E.G. Yukihara, C. Ruan, P.B.R. Gasparian, W.J. Clouse, C. Kalavagunta, S. Ahmad, Phys. Med. Biol. 54 (2009) 6337. [21] C. Ruan, E.G. Yukihara, W.J. Clouse, P.B.R. Gasparian, S. Ahmad, Med. Phys. 37 (2010) 3560. [22] J.A. Bauhs, T.J. Vrieze, A.N. Primak, M.R. Bruesewitz, C.H. McCollough, RadioGraphics 28 (2008) 245. [23] A. Jahn, M. Sommer, J. Henniger, Radiat. Meas. 45 (2010) 674. [24] K. Idri, L. Santoro, E. Charpiot, J. Herault, A. Costa, N. Ailleres, R. Delard, J.R. Vaille, J. Fesquet, L. Dusseau, IEEE Trans. Nucl. Sci 51 (2004) 3638. [25] M.L. Mah, M.E. Manfred, S.S. Kim, M. Prokic´, E.G. Yukihara, J.J. Talghader, IEEE Sensors 10 (2010) 311. [26] J.J. Kingsley, N. Manickam, K.C. Patil, Bull. Mater. Sci. 13 (1990) 179. [27] J.J. Kingsley, K.C. Patil, Mater. Lett. 6 (1988) 427. [28] L.E. Shea, J. McKittrick, O.A. Lopez, J. Am. Ceram. Soc. 79 (1996) 3257. [29] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos, Mater. Lett. 10 (1990) 6. [30] V.S.M. Barros, W.M. Azevedo, H.J. Khoury, M.E.A. Andrade, P.LInhares Filho, Radiat. Meas. 45 (2009) 435. [31] V.S.M. Barros, W.M. Azevedo, H.J. Khoury, P.L. Filho, Radiat. Meas. 43 (2008) 345. [32] M.W. Blair, L.G. Jacobsohn, B.L. Bennett, S.C. Tornga, E.G. Yukihara, E.A. McKigney, R.E. Muenchausen, Phys. Status Solidi A 206 (2009) 904. [33] M.W. Blair, L.G. Jacobsohn, S.C. Tornga, O. Ugurlu, B.L. Bennett, E.G. Yukihara, R.E. Muenchausen, J. Lumin. 130 (2010) 825. [34] P. Dorenbos, J. Alloys Compd. 488 (2009) 568. [35] P. Dorenbos, A.J.J. Bos, Radiat. Meas. 43 (2008) 139. [36] A.J.J. Bos, P. Dorenbos, A. Bessie re, B. Viana, Radiat. Meas. 43 (2008) 222. [37] F. You, A.J.J. Bos, Q. Shi, S. Huang, P. Dorenbos, J. Phys.: Condens. Matter 23 (2011) 6 215502. [38] V.R. Orante-Barro´n, L.C. Oliveira, J.B. Kelly, E.D. Milliken, G. Denis, L.G. Jacobsohn, J. Puckette, E.G. Yukihara, J. Lumin. 131 (2011) 1058. [39] L.C. Oliveira, G. Denis, E.D. Milliken, J. Puckette, E.G. Yukihara, J. Lumin. (in preparation). [40] M.S. Akselrod, S.W.S. McKeever, Radiat. Prot. Dosim. 81 (1999) 167. [41] S.R. Jain, K.C. Adiga, V.R.P. Verneker, Combust. Flame 40 (1981) 71. [42] E.G. Yukihara, L.G. Jacobsohn, M.W. Blair, B.L. Bennett, S.C. Tornga, R.E. Muenchausen, J. Lumin. 130 (2010) 2309. [43] L. Bøtter-Jensen, K.J. Thomsen, M. Jain, Radiat. Meas. 45 (2010) 253. [44] F. Gu, C. Li, H. Cao, W. Shao, Y. Hu, J. Chen, A. Chen, J. Alloys Compd. 453 (2008) 361. [45] F. Gu, C.Z. Li, H.B. Jiang, J. Cryst. Growth 289 (2006) 400. [46] L.G. Jacobsohn, S.C. Tornga, B.L. Bennett, R.E. Muenchausen, O. Ugurlu, T.K. Tseng, J. Choi, P.H. Holloway, Radiat. Meas. 45 (2010) 611. [47] E. Zych, Opt. Mater. 16 (2001) 445.