Journal of Sol-Gel Science and Technology 19, 325–328, 2000 c 2000 Kluwer Academic Publishers. Manufactured in The Netherlands. °
New Lutetium Silicate Scintillators ERIC BESCHER∗, S.R. ROBSON AND J.D. MACKENZIE Department of Materials Science, University of California Los Angeles, Los Angeles, CA 90095, USA
[email protected]
B. PATT AND J. IWANCZYK Photon Imaging, 19355 Business Center Drive, Northridge, CA 91324, USA E.J. HOFFMAN School of Medicine, University of California Los Angeles, Los Angeles, CA 90095, USA
Abstract. Cerium-doped lutecium orthosilicate (LSO) is the most promising scintillator discovered in almost five decades. It exhibits a unique combination of important properties for x and gamma-ray spectroscopy: high density, fast decay, and large light yield. However, the practical use of LSO is hindered by difficulties related to its fabrication as a single crystal by the Czochralski method. We report on the usefulness of the sol-gel process in obtaining lutecium silicate scintillators. Upon appropriate drying and firing, lutetium silicate crystals can be grown in a silica matrix. The bulk, polycrystalline transparent scintillators are characterized by XRD, optical absorption, light decay measurement and gamma-ray spectral response. Their properties are comparable to that of traditional LSO single crystals. Keywords:
1.
lutecium orthosilicate, sol-gel, glass-ceramics, scintillators, Lu2 SiO5
Introduction
The discovery of cerium-doped lutetium oxyorthosilicate LuSiO5 (LSO) has been the most promising development in the field of scintillators in over fifty years [1]. LSO offers high light output, fast decay time and excellent chemical stability compared to other scintillators. Therefore, this inorganic scintillator has many potential applications in medical imaging (such as Positron Emission Spectroscopy), nuclear physics, high energy physics and environmental monitoring. However, the production of commercial LSO single crystals is hindered by the difficulties inherent to the Czochralski method [2–4] and to date, LSO is not available commercially. In this paper, we describe a low temperature and low cost synthesis of bulk transparent polycrys∗ To
whom all correspondence should be addressed.
talline Ce-doped lutetium silicates. The optical properties of the polycrystalline, sol-gel derived samples compare favorably with bulk single crystals of LSO, and may offer an alternative to the fabrication of bulk single crystals. 2.
Experimental
A range of compositions with various Lu : Si mole ratios (from 1 : 100 to 2 : 1) were synthesized. 0.2–0.5 g of 40-mesh lutetium powder (Aldrich Chemicals) were used per batch. The metal powder was refluxed in anhydrous isopropanol (60 ml/g Lu) at ∼82◦ C under nitrogen for 3 days. 10−4 –10−3 moles of HgCl2 /g Lu were added to catalyze the reaction. A separate solution containing TEOS, anhydrous isopropanol and cerium nitrate solution was prepared (10 ml isopropanol/1 ml TEOS. Ce(III)NO3 (calculated
326
Figure 1.
Bescher et al.
TEM of lutetium silicate crystals in transparent matrix.
to 1 wt% of the final oxide) (Aldrich Chemicals) was added and the solution was stirred at room temperature for 2 hours. This solution was subsequently added to the lutetium solution and refluxed for two hours. After centrifugation to separate the amalgam from the liquid, the colorless solution was cast and exposed to ambient atmosphere. A gel formed after a few days, which turned light yellow and transparent. After drying, the gel was fired at 1◦ C/min to up to 1200◦ C, for 1–4 hours.
Optical absorption was measured by using a tunablewavelength monochromator system. Scintillation was measured by coupling the samples to a PMT and irradiating it with Cs-137 or Am-241. Since the system gain was constant, the difference in peak position (# channels) was directly proportional to the difference in light output. The light decay constant was measured by placing the samples on a very fast PMT and irradiating it with Na-22. In addition to the LSO samples, a plastic
New Lutetium Silicate Scintillators
Figure 2.
Differential thermal analysis showing crystallization of lutetium silicate crystals at 1087◦ C.
scintillator, BC404, was measured to test the speed of the PMT and to make sure that the system limitations do not affect the measurements. 3.
327
Results and Discussion
Samples with Lu/Si ratio between 1 : 6.5 and 1 : 4.5 retained their transparency after firing. Samples with higher Lu content became opaque after firing, probably due to excessive crystal growth. TEM analysis of the transparent samples fired at 1200◦ C revealed the pres˚ crystals in an amorphous matrix ence of small (500 A) (Fig. 1). The XRD showed the presence of well-defined crystals. Samples fired below 1200◦ C were amorphous by XRD. DTA analysis (Fig. 2) of a wet gel shows no other feature than a well-defined exotherm at 1087◦ C. It is still unclear whether the matrix is made up of pure silica or if it contains other components. We seems to have evidence of a Lu2 SiO7 compound. It is important to note that this crystallization temperature is well below temperatures required by the Czochralski method (ca. 2000◦ C) and also well below temperatures and times required for the solid-state synthesis of LuSiO5 from SiO2 and Lu2 O3 powders (14 days at 1400◦ C). The transmission spectra in the visible of a Czochralski-grown crystal and the sol gel samples were
compared. The transparency of both types of samples was higher than 95% in the 400–480 nm range. The scintillating properties of the material are shown in Fig. 3. The light output of the bulk sol gel sample was 65.5% that of the Czochralski crystal when irradiated with Cs-137 (662 keV). With Am-241 radiation (60 keV), the light output of the bulk sol gel sample increased to 69.5% that of the output of the Czochralski crystal. The light decay of the Czochralski crystal and the sol gel sample was fit to single exponential functions. The recent report on the decay constant for the Czochralski LSO crystal is in the range of 28–46 ns, with an average of 37 ns. The Czochralski crystal had a decay time of 41.1 ns, and the sol gel sample a decay time of 34 ns. The decay constant for both samples is within expected experimental parameters. The emission and excitation spectra of the sol-gel sample are shown in Fig. 4. The doublet structure corresponding to the Ce+3 ion transition from the 5d level to the two 4f ground states is mostly washed out at room temperature but can still be seen in these spectra. The measured excitation spectrum peaks at 346 nm for 380 nm emission wavelength. The general shape and location these spectra agree with what is reported for Czochralski LSO, with a slight wavelength shift.
328
Bescher et al.
Figure 3. Scintillation properties. (a) Comparison of the Am-241 energy spectra collected from a sol-gel crystal (1.5 × 2 × 0.85 mm3 ) and a Czochralski crystal (2.0 × 1.7 × 2.5 mm3 ) with energy scales individually calibrated. (b) PMT pulses from the plastic scintillator, the standard LSO (Czochralski) and a sol-gel sample (∼1.5 × 5 × 0.75 mm3 ). The average decay time of the sol-gel samples is 34.6 ns.
4.
Conclusions
Ce-doped lutetium silicate crystals were grown in a silica matrix by the sol gel technique. The process involves reaction of Lu metal with alcohol, followed by reflux with TEOS and slow hydrolysis of the purified solution. After drying and heating, lutetium silicate crystals are obtained in an amorphous matrix. The material is transparent if the crystals are sufficiently small. The sol gel derived scintillator exhibits properties similar to the conventionally grown LSO. Sol-gel derived,
Figure 4. (a) Wavelength distribution of scintillation light when excited at 356 nm light at room temperature. (b) Excitation spectrum of a sol-gel sample (2.0 × 1.7 × 2.5 mm3 ). This corresponds to an emission wavelength of 380 nm, which is the location of the maximum seen in spectrum (a).
polycrystalline bulk LSO may be an excellent alternative to single crystal LSO obtained by the Czochralski technique. References 1. C.L. Melcher and J.S. Schweitzer, IEEE Trans. Nucl. Sci. 39, 502 (1992). 2. C.D. Brandle, A.J. Valentino, and G.W. Berkstresser, J. Cryst. Growth 79, 308 (1989). 3. C.L. Melcher, R.A. Manente, C.A. Peterson, and J.S. Schweitzer, J. Cryst. Growth 128, 1001 (1993). 4. C.L. Melcher, US Patent 4,958,080, 1990.
