- Email: [email protected]
Crystal growth of Ce: PrF3 by micro-pulling-down method
ARTICLE IN PRESS
Journal of Crystal Growth 270 (2004) 427–432 www.elsevier.com/locate/jcrysgro
Crystal growth of Ce: PrF3 by micro-pulling-down method A. Yoshikawaa,, T. Satonagab, K. Kamadaa, H. Satoc, M. Nikld, N. Solovievad, T. Fukudaa,c a
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan b Stella Chemifa Co. Ltd.,3-6-3 Awaji-cho chuo-ku Osaka, Japan c Fukuda Crystal Laboratory Ltd., 6-6-3 Minami-Yoshinari, Aoba-ku, Sendai 989-3204, Japan d Institute of Physics AS CR, Cukrovarnicka 10, 162 53 Prague, Czech Republic Received 28 December 2003; accepted 20 June 2004 Communicated by M. Schieber
Abstract Micro-pulling-down (m-PD) growth apparatus was modified for fluoride crystals. PrF3 was grown with various concentrations of Ce3+ from 0–100%. The crystals were transparent and colorless (CeF3) or greenish and 3 mm in diameter and 15–50 mm in length. Neither visible inclusions nor cracks were observed. Radioluminescence spectra and decay kinetics were measured for the sample set at room temperature. In comparison to the Czochralski or Bridgman method, the m-PD method allows to produce single crystalline material in a faster thus more economic way. Once it is established for the fluoride crystals, it is an efficient tool for exploring the field of new functional fluorides. r 2004 Elsevier B.V. All rights reserved. PACS: 81.10.Fq; 78.20. e; 78.55. m Keywords: A1. Energy transfer; A1. Luminescence; A2. Micro-pulling-down method; A3. Single crystal growth; B1. Ce3+; B1. Fluoride; B3. Scintillators
1. Introduction Over the last decade, systematic effort has been devoted to develop novel functional crystals using oxide [1–4] and fluoride [5,6] compounds. During Corresponding author. Tel.: +81-22-217-5167, fax: +8122-217-5102. E-mail address: [email protected] (A. Yoshikawa).
the studies, even after hundreds of crystal growth attempts, we often found the necessity of further optimization of the melt composition. Based on such experience we have developed the micropulling-down method (m-PD). The m-PD method is a rather fast and economic tool to prepare single crystal systems and to optimize the growth technology. This makes the m-PD method advantageous in comparison to conventional Czochralski or Bridgman growth methods and has been
0022-0248/$ - see front matter r 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2004.06.038
ARTICLE IN PRESS 428
A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
successfully used to investigate oxide single crystalline materials [7–12]. Fluoride materials are under systematic investigations, as they found applications in various fields, such as laser host, optical materials for VUV as well as IR regions, resistant window materials for excimer lasers, etc. Recently, we have started to investigate fluoride crystals with respect to their usage in scintillation detectors [13,14] and we are realizing the necessity of an improved preparation method to facilitate remarkably the screening of fluoride host lattices, dopant concentration and optimization of the growth parameters. The m-PD method introduces a suitable innovative concept for single crystal growth and material screening. CeF3 single crystals were intensively studied for scintillator application in high energy physics in the first half of the 1990’s [15,16]. The lower light yield of this was explained by essentially nonradiative recombination between Ce4+ and an electron, interaction between the closely spaced excited Ce3+ ions and a limited concentration quenching effect. Luminescence of the neighboring Pr3+ rare earth ion was studied mainly due to its usage in up-conversion or photon cascade emission systems [17,18]. However, to our knowledge, detailed study of PrF3 single crystal luminescence has not been published. The question arises, which luminescence and scintillation features can be obtained in Ce-doped PrF3 crystals. In this paper, we present the construction of the m-PD equipment adapted for the fluoride crystal growth. Ce-doped (0%, 0.1%, 0.5%, 1%, 3%, 5%, 10%, 20%, 60%, 80% and 100%) PrF3 single crystal rods were prepared. Their luminescence characteristics were studied and energy transfer processes between Pr3+ and Ce3+ ions were evidenced.
2. Experimental procedure 2.1. Apparatus adaptation for the growth of fluorides Fig. 1 shows the schematic of the m-PD apparatus, which we have adapted for the growth
Fig. 1. Schematic of the m-PD apparatus for the fluoride crystal growth.
of fluoride crystal. The concept is similar as for oxide compounds. However, the growth chamber can be evacuated down to 10 5 Torr by rotary and diffusion pumps. It is equipped with a window made of CaF2 for visual observation of the meniscus region, solid/liquid interface, etc. using a charge coupled device camera with monitor. Carbon or platinum were used for crucible material. The crucibles are surrounded by refractory carbon and inductively heated using a radiofrequency generator. 2.2. Crystal growth procedure Starting materials were prepared from the stoichiometric mixture of 99.99% pure CeF3 and PrF3 powders produced by Stella Chemifa Corporation. They were thoroughly mixed and put into the crucible. The chamber was evacuated up to 10 4 Torr. Then, the crucible was heated up to 600 1C and kept for about 1 h at this temperature in order to remove oxygen traces caused by moisture of raw materials and adsorbates on the chamber surface. During this baking procedure, the chamber is further evacuated down to 10 5 Torr. After the baking, the recipient was filled with high purity Ar (99.999%) until ambient pressure. The crucible was heated up to the melting temperature of about 1430 1C.
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
429
Platinum wire was used for seeding during the initial crystal growth. Further crystal growth attempts were carried out using the seed of PrF3 crystal obtained in the initial experiments. The growth rate was 0.05–0.5 mm/min. 2.3. Phase characterization Parts of grown crystals were crashed and ground into powders. Powder X-ray diffraction analysis was carried out in the 2y range 201–801 using a RINT Ultima (RIGAKU) diffractometer. The X-ray source was CuKa, the accelerating voltage was 40 kV and the current was 40 mA. Quantitative analyses of the crystals for Ce and Pr along the growth direction were performed by electron probe micro-analysis (EPMA) using a JXA-8621MX (JEOL). ZAF correction is used, where Z is the atomic number, A the absorption correction factor and F the fluorescence correction factor. 2.4. Optical measurement Luminescence measurement procedure: Radioluminescence (RL) spectra and photoluminescence (PL) decays at room temperature (RT) were measured with a Spectrofluorometer 199S (Edinburgh Instrument) using an X-ray tube (operated at 35 kV and 16 mA, Mo cathode) and pulsed nanosecond hydrogen-filled flashlamp for the excitation. Single grating excitation and emission monochromators and photon counting detection were employed, for details see Ref. [14]. Absorption spectra were measured with an Shimazu UV 3101 PC absorption spectrometer. Measurements were performed on polished 0.8 mm thick single crystalline samples.
3. Results and discussion 3.1. Crystal growth Single crystals of Pr1 xCexF3 with x=0.0, 0.001, 0.005, 0.01, 0.03, 0.05, 0.10, 0.20, 0.6, 0.8, 1.0 were grown by the m-PD method. Examples are shown in Fig. 2. The grown crystals were transparent and
Fig. 2. Photographs of the grown Ce:PrF3 crystals. (a) 0.5% Ce3+, (b) 20% Ce3+ and (c) 80% Ce3+.
of greenish color (except CeF3: colorless), 3 mm in diameter and 15–50 mm in length. Neither visible inclusions nor cracks were observed. The evolution of the result from powder X-ray diffraction as a function of the Ce3+ concentration is shown in Fig. 3(a). No impurity phase was detected and we concluded that PrF3 and CeF3 form a continuous solid solution. The peaks were shifting towards smaller 2y, with increasing Ce3+ concentration. Fig. 3(b) shows the evolution of the lattice constant. It increases linearly as the Ce3+ concentration increase. Since the ionic radius of Ce3+ is larger than that of Pr3+, this tendency is reasonable and is in good agreement with Vegard’s law. The deviation of the composition in the grown crystal was observed by EPMA and found to be
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
430
50
60
70
80
2θ / degree
Lattice constant / Å
intensity [arb. units]
0 300
400
500
600
0 700
Wavelength[nm] / nm wavelength
300
400
500
600
700
Fig. 4. RL spectra of Ce: PrF3 at room temperature. The solid line stands for 0.1% and the dotted line for 1% Ce3+. In the inset, the absorption spectrum of Ce:PrF3 (0.5%), solid line, and the RL spectrum of Ce: PrF3 (10%), dotted line, are given.
a-axis
7.26 7.24 7.14
c-axis
7.12 7.10 7.08
0 (b)
1
Wavelength / nm
7.30 7.28
2
0.4
006
304 115 411 224 142
223, 214
114 222
004
112
202 211
40
032 221 220
300 113
111 002 002
30
(a)
Intensity / arb. units
Intensity / arb.units
Intensity / arb. unit
20
PrF3 (0%)
3
0.8
absorbance [arb. units]
Ce:PrF3 (60%)
Absorbance / arb. units
4
CeF3 (100%)
20
40
60
80
100
Ce concentration in Ce:PrF3 / %
Fig. 3. (a) The evolution of the powder X-ray diffraction (CuKa) chart as a function of the Ce3+ concentration (0%, 60% and 100%). (b) Evolution of the lattice constant as a function of the Ce3+ concentration.
highly uniform. The effective distribution coefficients keff of Ce is about 1. Homogeneous distribution of Ce3+ in Ce:PrF3 could be observed. The deviation from the calculated Ce3+ concentration within the same crystal did not exceed 2%, which corresponds to the experimental error of the EPMA measurements. 3.2. Luminescence characteristics RL and absorption spectra are given in Fig. 4. In a nearly undoped crystal the dominant band at 395 nm and weaker bands at 335 and 270 nm are observed and are ascribed to radiative transitions of the Pr3+ cation from 1S0 to 3P1, 1D2 and 1G4, respectively. With increasing Ce3+ concentration, the intensity of the emission from these bands
decreases and emission from the band at 290 nm, ascribed to the Ce3+ center, becomes more pronounced. Emission intensity of this band starts to decrease for Ce concentration above 10%, which probably indicate an onset of concentration quenching. Concentration dependence of the luminescence spectra points to the competition in energy capture and energy transfer between the Ce3+ or Pr3+ ions. For completeness, the absorption spectrum of 0.5% Ce-doped PrF3 is given in the inset together with the RL spectrum of Ce:PrF3(10%): At this concentration of Ce dopant the emission of Pr3+ completely disappeared. Observed absorption peaks at 441, 469, 484 and 593 nm are ascribed to the transitions from the 3H4 ground state of Pr3+ to 3P2, 3P1-1I6, 3P0 and 1D2 excited-state levels, respectively. At 260 nm, there is the absorption edge of Ce3+, which is a 4f–5d absorption transition. It is clear from Fig. 4 that neither Ce3+ nor Pr3+ emissions can be reabsorbed by any of the absorption transitions. PL decay kinetics in the 395 nm band was measured under the 215 nm excitation (expected transition 3H4–1S0). Decay curves for undoped, 1% and 3% Ce-doped samples were normalized to unity and are given in Fig. 5. Pronounced shortening of the decay points to the nonradiative energy transfer from Pr3+ to Ce3+ ions is most
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
PrF3 down to 18 ns for the Ce3+ concentration of about 3%.
1
Normalized intensity
431
0.1 und., em=396nm 1%Ce, em=396nm 3%Ce, em=300nm 3%Ce, em=396nm
0.01
0.001 200
400
600
800
1000
1200
1400
time / ns
Fig. 5. Normalized decay curves of the 395 nm emission (exc=215 nm) of Pr3+ in the undoped, 1% and 3% Ce: PrF3 are shown. The emission decay of Ce3+ luminescence at 300 nm under the 250 nm excitation for Ce:PrF3 (3%) is also given.
probably from 1S0 state of Pr3+ to the higher laying 5d states of Ce3+. For Ce-concentrations round 3%, the 395 nm emission decay becomes comparable with that of the Ce3+ 290 nm band (18 ns decay time, also in Fig. 5) so that Pr3+ emission in such samples does not introduce anymore slower components and scintillation response will not be degraded with respect to CeF3. From first comparison of RL intensity it seems that the total light production of CeF3 and PrF3:Ce (3% to 10%) is comparable.
4. Conclusion Ce doped PrF3 single crystal rods without visible defects could be grown by the m-PD method. Modifications of the m-PD apparatus for the fluoride crystal growth became necessary and could be established. The compositional homogeneity of Ce: PrF3 single crystals was evidenced. RL spectra feature the emission bands of Pr3+ at 395, 335 and 270 nm for undoped or weakly Cedoped samples: These emissions arise due to 1 S0-3P1,1D2 and 1G4 transitions in Pr3+ ions, respectively. In heavily doped samples only the Ce3+ emission at 290 nm is observed and the decay time is 18 ns. Energy transfer from Pr3+ to Ce3+ results in shortening the decay time of the Pr3+ luminescence from 580 ns in the undoped
Acknowledgements The authors would like to thank Prof. D. Ehrentraut for the fruitful discussion with him. The authors also would like to thank Mr. Murakami, of the Laboratory for Developmental Research of Advanced Materials in IMR, for his assistance with the EPMA analysis. This work was partially supported by the Industrial Technology Research Grant Program 03A26014a from the New Energy and Industrial Technology Development Organization (NEDO) of Japan. Partial support of the Czech MSMT, KONTAKT Grant 1P2004ME716 is also gratefully acknowledged.
References [1] B.M. Epelbaum, K. Shimamura, K. Inaba, S. Uda, V.V. Kochurikhin, H. Machida, Y. Terada, T. Fukuda, J. Crystal Growth 186 (1998) 607. [2] V.I. Chani, K. Shimamura, S. Endo, T. Fukuda, J. Crystal Growth 171 (1997) 472. [3] A. Yoshikawa, Y. Kagamitani, D.A. Pawlak, H. Sato, H. Machida, T. Fukuda, Mater. Res. Bull. 37 (2002) 1. [4] T. Fukuda, K. Shimamura, V.V. Kochurikhin, V.I. Chani, B.M. Epelbaum, S.L. Baldochi, T. Takeda, A. Yoshikawa, J. Mater. Sci. Mater. Electron. 10 (1999) 571. [5] S.L. Baldochi, K. Shimamura, K. Nakano, Na Mujilatu, T. Fukuda, J. Crystal Growth 200 (1999) 521. [6] H. Sato, H. Machida, M. Nikl, A. Yoshikawa, T. Fukuda, J. Crystal Growth 250 (2003) 83. [7] D.H. Yoon, M. Hashimoto, T. Fukuda, Jpn. J. Appl. Phys. 33 (1994) 3510. [8] B.M. Epelbaum, K. Inaba, S. Uda, T. Fukuda, J. Crystal Growth 178 (1997) 426. [9] B.M. Epelbaum, D. Hofmann, J. Crystal Growth 237–239 (2002) 2098. [10] A. Yoshikawa, T. Akagi, M. Nikl, N. Solovieva, K. Lebbou, C. Dujardin, C. Pedrini, T. Fukuda, Nucl. Instrum. Methods Phys. Res. A 486 (2002) 79. [11] J.W. Shur, T.I. Shin, S.M. Lee, S.W. Baek, D.H. Yoon, Mater. Sci. Eng. B 105 (2003) 16. [12] A. Yoshikawa, H. Itagaki, T. Fukuda, K. Lebbou, A. El Hassouni, A. Brenier, C. Goutaudier, O. Tillement, G. Boulon, J. Crystal Growth 247 (2003) 148. [13] M. Nikl, N. Solovieva, E. Mihokova, M. Dusek, A. Vedda, M. Martini, K. Shimamura, T. Fukuda, Phys. Status Solidi (a) 187 (2001) R1.
ARTICLE IN PRESS 432
A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
[14] M. Nikl, E. Mihokova, Z. Malkova, A. Vedda, M. Martini, K. Shimamura, T. Fukuda, Phys. Rev. B 66 (2002) 184101. [15] E. Auffray, et al., Crystal Clear Collaboration, Nucl. Instrum Methods Phys. Res. A 383 (1996) 367. [16] M. Nikl, Phys. Status Solidi (a) 178 (2000) 595.
[17] A. A. Kaminskii, H. J. Eichler, Phys. Status Solidi (b) (1994) K85. [18] A.P. Vink, P. Dorenbos, J.T.M. de Haas, H. Donker, P.A. Rodnyi, A.G. Avanesov, C.W.E. van Eijk, J. Phys.: Condens. Matter 14 (2002) 8889.
Journal of Crystal Growth 270 (2004) 427–432 www.elsevier.com/locate/jcrysgro
Crystal growth of Ce: PrF3 by micro-pulling-down method A. Yoshikawaa,, T. Satonagab, K. Kamadaa, H. Satoc, M. Nikld, N. Solovievad, T. Fukudaa,c a
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan b Stella Chemifa Co. Ltd.,3-6-3 Awaji-cho chuo-ku Osaka, Japan c Fukuda Crystal Laboratory Ltd., 6-6-3 Minami-Yoshinari, Aoba-ku, Sendai 989-3204, Japan d Institute of Physics AS CR, Cukrovarnicka 10, 162 53 Prague, Czech Republic Received 28 December 2003; accepted 20 June 2004 Communicated by M. Schieber
Abstract Micro-pulling-down (m-PD) growth apparatus was modified for fluoride crystals. PrF3 was grown with various concentrations of Ce3+ from 0–100%. The crystals were transparent and colorless (CeF3) or greenish and 3 mm in diameter and 15–50 mm in length. Neither visible inclusions nor cracks were observed. Radioluminescence spectra and decay kinetics were measured for the sample set at room temperature. In comparison to the Czochralski or Bridgman method, the m-PD method allows to produce single crystalline material in a faster thus more economic way. Once it is established for the fluoride crystals, it is an efficient tool for exploring the field of new functional fluorides. r 2004 Elsevier B.V. All rights reserved. PACS: 81.10.Fq; 78.20. e; 78.55. m Keywords: A1. Energy transfer; A1. Luminescence; A2. Micro-pulling-down method; A3. Single crystal growth; B1. Ce3+; B1. Fluoride; B3. Scintillators
1. Introduction Over the last decade, systematic effort has been devoted to develop novel functional crystals using oxide [1–4] and fluoride [5,6] compounds. During Corresponding author. Tel.: +81-22-217-5167, fax: +8122-217-5102. E-mail address: [email protected] (A. Yoshikawa).
the studies, even after hundreds of crystal growth attempts, we often found the necessity of further optimization of the melt composition. Based on such experience we have developed the micropulling-down method (m-PD). The m-PD method is a rather fast and economic tool to prepare single crystal systems and to optimize the growth technology. This makes the m-PD method advantageous in comparison to conventional Czochralski or Bridgman growth methods and has been
0022-0248/$ - see front matter r 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2004.06.038
ARTICLE IN PRESS 428
A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
successfully used to investigate oxide single crystalline materials [7–12]. Fluoride materials are under systematic investigations, as they found applications in various fields, such as laser host, optical materials for VUV as well as IR regions, resistant window materials for excimer lasers, etc. Recently, we have started to investigate fluoride crystals with respect to their usage in scintillation detectors [13,14] and we are realizing the necessity of an improved preparation method to facilitate remarkably the screening of fluoride host lattices, dopant concentration and optimization of the growth parameters. The m-PD method introduces a suitable innovative concept for single crystal growth and material screening. CeF3 single crystals were intensively studied for scintillator application in high energy physics in the first half of the 1990’s [15,16]. The lower light yield of this was explained by essentially nonradiative recombination between Ce4+ and an electron, interaction between the closely spaced excited Ce3+ ions and a limited concentration quenching effect. Luminescence of the neighboring Pr3+ rare earth ion was studied mainly due to its usage in up-conversion or photon cascade emission systems [17,18]. However, to our knowledge, detailed study of PrF3 single crystal luminescence has not been published. The question arises, which luminescence and scintillation features can be obtained in Ce-doped PrF3 crystals. In this paper, we present the construction of the m-PD equipment adapted for the fluoride crystal growth. Ce-doped (0%, 0.1%, 0.5%, 1%, 3%, 5%, 10%, 20%, 60%, 80% and 100%) PrF3 single crystal rods were prepared. Their luminescence characteristics were studied and energy transfer processes between Pr3+ and Ce3+ ions were evidenced.
2. Experimental procedure 2.1. Apparatus adaptation for the growth of fluorides Fig. 1 shows the schematic of the m-PD apparatus, which we have adapted for the growth
Fig. 1. Schematic of the m-PD apparatus for the fluoride crystal growth.
of fluoride crystal. The concept is similar as for oxide compounds. However, the growth chamber can be evacuated down to 10 5 Torr by rotary and diffusion pumps. It is equipped with a window made of CaF2 for visual observation of the meniscus region, solid/liquid interface, etc. using a charge coupled device camera with monitor. Carbon or platinum were used for crucible material. The crucibles are surrounded by refractory carbon and inductively heated using a radiofrequency generator. 2.2. Crystal growth procedure Starting materials were prepared from the stoichiometric mixture of 99.99% pure CeF3 and PrF3 powders produced by Stella Chemifa Corporation. They were thoroughly mixed and put into the crucible. The chamber was evacuated up to 10 4 Torr. Then, the crucible was heated up to 600 1C and kept for about 1 h at this temperature in order to remove oxygen traces caused by moisture of raw materials and adsorbates on the chamber surface. During this baking procedure, the chamber is further evacuated down to 10 5 Torr. After the baking, the recipient was filled with high purity Ar (99.999%) until ambient pressure. The crucible was heated up to the melting temperature of about 1430 1C.
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
429
Platinum wire was used for seeding during the initial crystal growth. Further crystal growth attempts were carried out using the seed of PrF3 crystal obtained in the initial experiments. The growth rate was 0.05–0.5 mm/min. 2.3. Phase characterization Parts of grown crystals were crashed and ground into powders. Powder X-ray diffraction analysis was carried out in the 2y range 201–801 using a RINT Ultima (RIGAKU) diffractometer. The X-ray source was CuKa, the accelerating voltage was 40 kV and the current was 40 mA. Quantitative analyses of the crystals for Ce and Pr along the growth direction were performed by electron probe micro-analysis (EPMA) using a JXA-8621MX (JEOL). ZAF correction is used, where Z is the atomic number, A the absorption correction factor and F the fluorescence correction factor. 2.4. Optical measurement Luminescence measurement procedure: Radioluminescence (RL) spectra and photoluminescence (PL) decays at room temperature (RT) were measured with a Spectrofluorometer 199S (Edinburgh Instrument) using an X-ray tube (operated at 35 kV and 16 mA, Mo cathode) and pulsed nanosecond hydrogen-filled flashlamp for the excitation. Single grating excitation and emission monochromators and photon counting detection were employed, for details see Ref. [14]. Absorption spectra were measured with an Shimazu UV 3101 PC absorption spectrometer. Measurements were performed on polished 0.8 mm thick single crystalline samples.
3. Results and discussion 3.1. Crystal growth Single crystals of Pr1 xCexF3 with x=0.0, 0.001, 0.005, 0.01, 0.03, 0.05, 0.10, 0.20, 0.6, 0.8, 1.0 were grown by the m-PD method. Examples are shown in Fig. 2. The grown crystals were transparent and
Fig. 2. Photographs of the grown Ce:PrF3 crystals. (a) 0.5% Ce3+, (b) 20% Ce3+ and (c) 80% Ce3+.
of greenish color (except CeF3: colorless), 3 mm in diameter and 15–50 mm in length. Neither visible inclusions nor cracks were observed. The evolution of the result from powder X-ray diffraction as a function of the Ce3+ concentration is shown in Fig. 3(a). No impurity phase was detected and we concluded that PrF3 and CeF3 form a continuous solid solution. The peaks were shifting towards smaller 2y, with increasing Ce3+ concentration. Fig. 3(b) shows the evolution of the lattice constant. It increases linearly as the Ce3+ concentration increase. Since the ionic radius of Ce3+ is larger than that of Pr3+, this tendency is reasonable and is in good agreement with Vegard’s law. The deviation of the composition in the grown crystal was observed by EPMA and found to be
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
430
50
60
70
80
2θ / degree
Lattice constant / Å
intensity [arb. units]
0 300
400
500
600
0 700
Wavelength[nm] / nm wavelength
300
400
500
600
700
Fig. 4. RL spectra of Ce: PrF3 at room temperature. The solid line stands for 0.1% and the dotted line for 1% Ce3+. In the inset, the absorption spectrum of Ce:PrF3 (0.5%), solid line, and the RL spectrum of Ce: PrF3 (10%), dotted line, are given.
a-axis
7.26 7.24 7.14
c-axis
7.12 7.10 7.08
0 (b)
1
Wavelength / nm
7.30 7.28
2
0.4
006
304 115 411 224 142
223, 214
114 222
004
112
202 211
40
032 221 220
300 113
111 002 002
30
(a)
Intensity / arb. units
Intensity / arb.units
Intensity / arb. unit
20
PrF3 (0%)
3
0.8
absorbance [arb. units]
Ce:PrF3 (60%)
Absorbance / arb. units
4
CeF3 (100%)
20
40
60
80
100
Ce concentration in Ce:PrF3 / %
Fig. 3. (a) The evolution of the powder X-ray diffraction (CuKa) chart as a function of the Ce3+ concentration (0%, 60% and 100%). (b) Evolution of the lattice constant as a function of the Ce3+ concentration.
highly uniform. The effective distribution coefficients keff of Ce is about 1. Homogeneous distribution of Ce3+ in Ce:PrF3 could be observed. The deviation from the calculated Ce3+ concentration within the same crystal did not exceed 2%, which corresponds to the experimental error of the EPMA measurements. 3.2. Luminescence characteristics RL and absorption spectra are given in Fig. 4. In a nearly undoped crystal the dominant band at 395 nm and weaker bands at 335 and 270 nm are observed and are ascribed to radiative transitions of the Pr3+ cation from 1S0 to 3P1, 1D2 and 1G4, respectively. With increasing Ce3+ concentration, the intensity of the emission from these bands
decreases and emission from the band at 290 nm, ascribed to the Ce3+ center, becomes more pronounced. Emission intensity of this band starts to decrease for Ce concentration above 10%, which probably indicate an onset of concentration quenching. Concentration dependence of the luminescence spectra points to the competition in energy capture and energy transfer between the Ce3+ or Pr3+ ions. For completeness, the absorption spectrum of 0.5% Ce-doped PrF3 is given in the inset together with the RL spectrum of Ce:PrF3(10%): At this concentration of Ce dopant the emission of Pr3+ completely disappeared. Observed absorption peaks at 441, 469, 484 and 593 nm are ascribed to the transitions from the 3H4 ground state of Pr3+ to 3P2, 3P1-1I6, 3P0 and 1D2 excited-state levels, respectively. At 260 nm, there is the absorption edge of Ce3+, which is a 4f–5d absorption transition. It is clear from Fig. 4 that neither Ce3+ nor Pr3+ emissions can be reabsorbed by any of the absorption transitions. PL decay kinetics in the 395 nm band was measured under the 215 nm excitation (expected transition 3H4–1S0). Decay curves for undoped, 1% and 3% Ce-doped samples were normalized to unity and are given in Fig. 5. Pronounced shortening of the decay points to the nonradiative energy transfer from Pr3+ to Ce3+ ions is most
ARTICLE IN PRESS A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
PrF3 down to 18 ns for the Ce3+ concentration of about 3%.
1
Normalized intensity
431
0.1 und., em=396nm 1%Ce, em=396nm 3%Ce, em=300nm 3%Ce, em=396nm
0.01
0.001 200
400
600
800
1000
1200
1400
time / ns
Fig. 5. Normalized decay curves of the 395 nm emission (exc=215 nm) of Pr3+ in the undoped, 1% and 3% Ce: PrF3 are shown. The emission decay of Ce3+ luminescence at 300 nm under the 250 nm excitation for Ce:PrF3 (3%) is also given.
probably from 1S0 state of Pr3+ to the higher laying 5d states of Ce3+. For Ce-concentrations round 3%, the 395 nm emission decay becomes comparable with that of the Ce3+ 290 nm band (18 ns decay time, also in Fig. 5) so that Pr3+ emission in such samples does not introduce anymore slower components and scintillation response will not be degraded with respect to CeF3. From first comparison of RL intensity it seems that the total light production of CeF3 and PrF3:Ce (3% to 10%) is comparable.
4. Conclusion Ce doped PrF3 single crystal rods without visible defects could be grown by the m-PD method. Modifications of the m-PD apparatus for the fluoride crystal growth became necessary and could be established. The compositional homogeneity of Ce: PrF3 single crystals was evidenced. RL spectra feature the emission bands of Pr3+ at 395, 335 and 270 nm for undoped or weakly Cedoped samples: These emissions arise due to 1 S0-3P1,1D2 and 1G4 transitions in Pr3+ ions, respectively. In heavily doped samples only the Ce3+ emission at 290 nm is observed and the decay time is 18 ns. Energy transfer from Pr3+ to Ce3+ results in shortening the decay time of the Pr3+ luminescence from 580 ns in the undoped
Acknowledgements The authors would like to thank Prof. D. Ehrentraut for the fruitful discussion with him. The authors also would like to thank Mr. Murakami, of the Laboratory for Developmental Research of Advanced Materials in IMR, for his assistance with the EPMA analysis. This work was partially supported by the Industrial Technology Research Grant Program 03A26014a from the New Energy and Industrial Technology Development Organization (NEDO) of Japan. Partial support of the Czech MSMT, KONTAKT Grant 1P2004ME716 is also gratefully acknowledged.
References [1] B.M. Epelbaum, K. Shimamura, K. Inaba, S. Uda, V.V. Kochurikhin, H. Machida, Y. Terada, T. Fukuda, J. Crystal Growth 186 (1998) 607. [2] V.I. Chani, K. Shimamura, S. Endo, T. Fukuda, J. Crystal Growth 171 (1997) 472. [3] A. Yoshikawa, Y. Kagamitani, D.A. Pawlak, H. Sato, H. Machida, T. Fukuda, Mater. Res. Bull. 37 (2002) 1. [4] T. Fukuda, K. Shimamura, V.V. Kochurikhin, V.I. Chani, B.M. Epelbaum, S.L. Baldochi, T. Takeda, A. Yoshikawa, J. Mater. Sci. Mater. Electron. 10 (1999) 571. [5] S.L. Baldochi, K. Shimamura, K. Nakano, Na Mujilatu, T. Fukuda, J. Crystal Growth 200 (1999) 521. [6] H. Sato, H. Machida, M. Nikl, A. Yoshikawa, T. Fukuda, J. Crystal Growth 250 (2003) 83. [7] D.H. Yoon, M. Hashimoto, T. Fukuda, Jpn. J. Appl. Phys. 33 (1994) 3510. [8] B.M. Epelbaum, K. Inaba, S. Uda, T. Fukuda, J. Crystal Growth 178 (1997) 426. [9] B.M. Epelbaum, D. Hofmann, J. Crystal Growth 237–239 (2002) 2098. [10] A. Yoshikawa, T. Akagi, M. Nikl, N. Solovieva, K. Lebbou, C. Dujardin, C. Pedrini, T. Fukuda, Nucl. Instrum. Methods Phys. Res. A 486 (2002) 79. [11] J.W. Shur, T.I. Shin, S.M. Lee, S.W. Baek, D.H. Yoon, Mater. Sci. Eng. B 105 (2003) 16. [12] A. Yoshikawa, H. Itagaki, T. Fukuda, K. Lebbou, A. El Hassouni, A. Brenier, C. Goutaudier, O. Tillement, G. Boulon, J. Crystal Growth 247 (2003) 148. [13] M. Nikl, N. Solovieva, E. Mihokova, M. Dusek, A. Vedda, M. Martini, K. Shimamura, T. Fukuda, Phys. Status Solidi (a) 187 (2001) R1.
ARTICLE IN PRESS 432
A. Yoshikawa et al. / Journal of Crystal Growth 270 (2004) 427–432
[14] M. Nikl, E. Mihokova, Z. Malkova, A. Vedda, M. Martini, K. Shimamura, T. Fukuda, Phys. Rev. B 66 (2002) 184101. [15] E. Auffray, et al., Crystal Clear Collaboration, Nucl. Instrum Methods Phys. Res. A 383 (1996) 367. [16] M. Nikl, Phys. Status Solidi (a) 178 (2000) 595.
[17] A. A. Kaminskii, H. J. Eichler, Phys. Status Solidi (b) (1994) K85. [18] A.P. Vink, P. Dorenbos, J.T.M. de Haas, H. Donker, P.A. Rodnyi, A.G. Avanesov, C.W.E. van Eijk, J. Phys.: Condens. Matter 14 (2002) 8889.