Czochralski growth of Gd2Ti2O7 single crystals

Czochralski growth of Gd2Ti2O7 single crystals

Journal of Crystal Growth 402 (2014) 94–98 Contents lists available at ScienceDirect Journal of Crystal Growth journal homepage:

927KB Sizes 18 Downloads 177 Views

Journal of Crystal Growth 402 (2014) 94–98

Contents lists available at ScienceDirect

Journal of Crystal Growth journal homepage:

Czochralski growth of Gd2Ti2O7 single crystals F.Y. Guo, W.H. Zhang, M. Ruan, J.B. Kang, J.Z. Chen n College of Chemistry and Chemical Engineering, Fuzhou University, Fuzhou 350116, Fujian, PR China

art ic l e i nf o

a b s t r a c t

Article history: Received 4 January 2014 Received in revised form 14 May 2014 Accepted 15 May 2014 Communicated by: R.S. Feigelson Available online 24 May 2014

Gd2Ti2O7 (GTO) single crystals having dimensions of 17  17  20 mm3 were grown by the Czochralski method. These crystals displayed a strong growth habit with {1 1 1} facets. The colors of the as-grown crystals were sensitive to the oxygen concentration both during growth and post-growth annealing. The possible reason for the different colors is discussed and based on transmission, energy dispersive Xray spectroscopy (EDS), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and electron spin resonance (ESR) analyses. & 2014 Elsevier B.V. All rights reserved.

Keywords: A1. Crystal morphology A2. Oxygen vacancy A2. Czochralski method B1. Pyrochlore

1. Introduction

2. Experimental

Rare-earth based titanates (R2Ti2O7, R ¼Y, Sm-Lu) belong to the family of compounds with the general chemical formula A2B2O7. They have the pyrochlore structure with a space group of Fd3m, where R3 þ is coordinated with eight oxygen ions and Ti4 þ with six oxygen ions and two vacant anion sites, R3 þ and Ti4 þ cations lie along 〈110〉 such that each species occupies alternating rows in the parent M4O8 structure [1,2]. In the past two decades, the pyrochlore compounds have attracted much attention due to the geometrical frustrations and interesting low temperature properties [3–11]. In addition, pyrochlore materials have other important properties, such as ionic conductivity [12–14], optical nonlinearity [15] and high radiation tolerance [16,17]. They have many potential applications, including thermal barrier coatings, high-permittivity dielectrics, solid electrolytes in solid-oxide fuel cells, and attractive candidates for safe disposal of actinide-containing nuclear waste [18]. In order to obtain better and bigger R2Ti2O7 crystals, several crystal growth methods have been studied, including the flux method [19], laser-heated pedestal growth (LHPG) and the optical floating-zone technique [20–23]. However, the diameters of these as-grown R2Ti2O7 crystals were less than 10 mm. Recently, our research group has obtained the Dy2Ti2O7 and Ho2Ti2O7 crystals by Czochralski method. Their diameters were up to 20 mm. In this paper, we report on the growth of Gd2Ti2O7 crystals by Czochralski method. The morphology of these as-grown GTO crystals is briefly described. The influences of growth atmosphere and postannealing temperature on the crystals’ color are discussed.

Polycrystalline materials for crystal growth were prepared in air by the solid-state reaction technique. After stoichiometric amounts of Gd2O3 (5 N) and TiO2 (4 N) were accurately weighed, the mixture was sintered at three different temperatures (1250, 1350 and 1450 1C) for 30 h each with intermediate grinding and pressing into tablets. The congruent melting point of GTO is 1820 1C [24]. Crystals were grown by the Czochralski method, in a Ф56 mm  32 mm iridium crucible with radio frequency (RF) induction heating. When the polycrystalline materials melted completely, the furnace was evacuated to near 0.1 Pa pressure to drive away the bubbles produced in the melting process. The furnace was then filled with the different protective gases to 0.05–0.08 MPa pressure for crystal growth. GTO crystals were grown along the [111] or [100] direction at pulling rates between 1.0 mm/h and 1.5 mm/h and rotation rates of 20–30 rpm. With this procedure, differently colored single crystals of GTO were obtained, as shown in Fig. 1. Powder XRD measurements were carried out using a Rigaku D/max-3c diffractometer. The measured diffraction data for yellow-GTO and black-GTO were refined using a Si internal standard. As-grown yellow-GTO and black-GTO single crystals were cut along (111) plane, which were oriented by X-ray diffraction, and then grounded and polished carefully to about 2 mm thickness for spectra, EDS and XPS measurements. The transmission spectra were measured over the wavelength range 400–2000 nm (PerkinElmer Lambda 900). The atom ratios of Gd, Ti and O of yellow-GTO and black-GTO samples were analyzed by means of EDS (HIROX/ SH-4000), in which ten points of each sample were randomly selected to be tested under the same conditions, then the resulted data were averaged and listed in Table 1. The valance band spectra


Corresponding author. Tel./fax: +86 59122866130. E-mail address: [email protected] (J.Z. Chen). 0022-0248/& 2014 Elsevier B.V. All rights reserved.

F.Y. Guo et al. / Journal of Crystal Growth 402 (2014) 94–98


Fig. 1. (a) GTO crystal grown along the [1 1 1] direction under 0.08 MPa N2(99.99%) pressure; (b) GTO crystal grown along the [1 0 0] direction under 0.06 MPa N2(99.9%) pressure; (c) GTO crystal grown along the [1 1 1] direction under 0.06 MPa 98%N2 þ 2%O2 pressure.



yellow-GTO yellow-GTO annealed at 800 °C yellow-GTO annealed at 1200 °C black-GTO

T(%) 40


0 200




1000 1200 1400 1600 1800 2000 2200

Wavelength (nm) Fig. 2. Transmission spectra of yellow and black GTO and yellow-GTO crystals annealed in air at different temperatures. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. XRD patterns of yellow-GTO and black-GTO crystals (Si powder was added as a standard reference material). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 1 The atom ratios of Gd, Ti and O of yellow-GTO and black-GTO samples. Element

Yellow-GTO C atom (at%)

Gadolinium 20.11 Titanium 19.52 Oxygen 60.37

Standard deviation (%)

Black-GTO C atom (at%)

Standard deviation (%)

0.34 0.29 0.39

19.89 19.10 61.01

0.26 0.37 0.33

of two samples were recorded by XPS (Thermo ESCALAB 250XI). Another two samples with sizes 5  5  0.8 mm3 were cut from yellow-GTO and black-GTO for ESR (Bruker ER200-SRC) analysis. All measurements were performed at room temperature.

3. Results and discussion 3.1. Crystal morphology GTO crystals showed relatively strong growth habits during CZ growth. Under suitable transverse temperature gradients, GTO crystal grown along the [111] direction had six fully exposed planes which were identified by X-ray diffraction to be {111} facets. The angles between these facets were approximately 70.5 degrees, which is close to the dihedral angle of a perfect

Fig. 4. The Pyrochlore structure (1/8 unit cell). Larger blue spheres are Gd3 þ ions, small yellow spheres are Ti4 þ ions, large red spheres are O2  ions, black circle is oxygen vacancy. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

octahedron. The as-grown GTO crystals exhibited a characteristic octahedral morphology, which is similar to that of GTO crystal grown by spontaneous nucleation from a Na2B4O7 and NaF flux.


F.Y. Guo et al. / Journal of Crystal Growth 402 (2014) 94–98

The growth morphology could be explained by the periodic bond chain (PBC) theory [25]. GTO crystals grown in the [100] direction had only two {111} facets. The angles between {111} facets and the [100] direction are approximately 54.7 degrees.

3.2. Color change The color of the as-grown GTO crystals is sensitive to the oxygen concentration in growth atmosphere, as shown in Fig. 1. A similar phenomenon also exists in the post-annealing process. The yellow-GTO crystal became dark after annealing in air at 800 1C. It was also found that the crystal color would become deeper when the annealing temperature was higher. Transmission spectrum measurement demonstrated that the absorbance of GTO crystal after annealing significantly increased in the 400–900 nm

wavelength range, while it did not change between 900 and 2000 nm (as shown in Fig. 2). In the annealing experiments, the annealing temperature was found to be the most important factor for color change. Moreover, this change was reversible. When a GTO crystal with black color was re-annealed in high-purity N2 or vacuum (0.1 Pa), its color returned to yellow. In the usual case, some oxide crystals grown by the CZ method in N2 protective atmosphere had an oxygen deficiency giving rise to an absorbance increase in the visible region. After annealing in O2 or air atmosphere, the absorption decreased due to the elimination of oxygen deficiency [26–29]. But for GTO crystals, the opposite absorption effect in the visible region was found. EDS data in Table 1 shows that the oxygen content of black-GTO is slightly larger than that of yellow-GTO. Both yellow and black GTO crystals were analyzed by powder XRD and showed no any impurity phase (Fig. 3). The absence of h ¼4n, k ¼4n, l ¼2n

Fig. 5. XPS spectra of yellow-GTO and black-GTO crystals (a) Gd4d electron, (b)Ti2p electron, (c) O1s electron. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

F.Y. Guo et al. / Journal of Crystal Growth 402 (2014) 94–98

reflections indicates that the cations remain on the 16c and 16d sites [30]. X-ray powder diffraction patterns were indexed, and the unit cell parameters of the two crystals were found to be 1.0195(2) nm for the yellow-GTO and 1.0194(1) nm for the black-GTO, respectively. The increase of oxygen content in GTO crystal did not have any measurable change to the lattice size. It is well known that pyrochlore structure can also be described as an anion-deficient, vacancy-ordered fluorite structure. There are three different kinds of oxygen ions sites in pyrochlore structure: 8b (3/8, 3/8, 3/8), 48f (x, 1/8, 1/8) and 8a (1/8, 1/8, 1/8), O(I) at 8b site; O(Π) at 48f sites; the oxygen vacancy is at the 8a site where is surrounded by 4 Ti4 þ ions [31]. In GTO six out of every seven oxygen ions occupy the 48f sites and each is coordinated by two Gd3 þ and two Ti4 þ ions. The other oxygen ion occupies the 8b site and is surrounded by 4 Gd3 þ ions, as shown as Fig. 4. When the Ti3 þ ions exists in some pyrochlore compounds prepared under stringent reducing conditions, the compounds were black but still retained the pyrochlore structure [32]. However, XPS analysis revealed no difference in the valence of Gd and Ti between black-GTO and yellow-GTO, as shown in Fig. 4a and b. No Ti3 þ species were observed in XPS. The absence of peak broadening of Ti 2p3/2 signals (FWHM was about 1.1 eV) also indicates the presence of Ti4 þ species only [33,34]. As shown in Fig. 5c, the O1s spectra show two peaks, 529.47 eV and 531.01 eV for yellow-GTO, 529.50 eV and 531.11 eV for blackGTO. According to the literature [35], the difference in binding energy of oxygen is the consequence of the difference in the effective negative charge on oxygen. Lower the BE, higher the average electron density on the element. The ionic character of the Gd–O bond is greater than that of Ti–O bond because of the lower electronegativity of Gd3 þ compared with that of Ti4 þ . This means that the electron density on oxygen for Gd–O bond is higher than that of the Ti–O bond. Thus the peak in the O1s spectra at lower BE is assigned to oxygen ions at 8b sites and the peak at higher BE is assigned to O at 48f sites. The area ratio of peak (529.47 eV)/peak (531.01 eV) of yellow-GTO is 1.31, while the area ratio of peak (529.50 eV)/peak (531.11 eV) of black-GTO is 1.38. The calculated results indicate that there is a decrease in the oxygen content at the 48f sites in black-GTO. The reason may be that a small amount of oxygen ions have occupied the 8a sites surrounded by 4 Ti4 þ ions. This could lead to partial oxygen vacancies transferred from the 8a to 48f sites in the pyrochlore structure at high temperature [21]. This transference and oxygen excess could give rise to a

Fig. 6. ESR spectra of of yellow-GTO and black-GTO crystals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


distortion of Ti–O octahedra of pyrochlore structure. However, no change of electron environment around paramagnetic Gd3 þ ions was found in the ESR measurements, as shown in Fig. 6. This result suggests that the quantity of transferred oxygen vacancies was too small to be observed.

4. Conclusions Gd2Ti2O7 (GTO) single crystals were grown successfully by Czochralski method. They displayed a strong growth habit with {111} facets. GTO crystal grown along the [111] direction had an octahedral morphology. The color of the as-grown GTO crystals was sensitive to the oxygen concentration during growth and in the post-annealing atmosphere. The presence of oxygen at high temperatures can darken the color of GTO crystal. EDS, XRD, XPS and ESR analyses suggested that this phenomenon may be associated with a small amount of oxygen ions entering into the oxygen vacancies of pyrochlore structure.

Acknowledgements This work is supported by the National Natural Science Foundation of China (nos. 91022025, 51072036) and the Science and Technology Development Fund of Fuzhou University (2011-XQ-11). References [1] E. Aleshin, R. Roy, J. Am. Ceram. Soc. 45 (1962) 18. [2] B.J. Wuensch, K.W. Eberman, J. Miner. Met. Mater. Soc. 52 (2000) 20. [3] M.J. Harris, S.T. Bramwell, D.F. McMorrow, T. Zeiske, K.W. Godfrey, Phys. Rev. Lett. 79 (13) (1997) 2554. [4] A.P. Ramirez, A. Hayashi, R.J. Cava, R. Siddharthan, B.S. Shastry, Nature 399 (1999) 333. [5] N.P. Raju, M. Dion, M.J.P. Gingras, T.E. Mason, J.E. Greedan, Phys. Rev. B: Condens. Matter 59 (1999) 14489. [6] S.T. Bramwell, M.N. Field, M.J. Harris, I.P. Parkin, J. Phys. Condens. Matter 12 (2000) 4834. [7] M.J.P. Gingras, B.C. den Hertog, M. Faucher, J.S. Gardner, S.R. Dunsiger, L. J. Chang, B.D. Gaulin, N.P. Raju, J.E. Greedan, Phys. Rev. B: Condens. Matter 62 (2000) 6496. [8] J.A. Hodges, P. Bonville, A. Forget, M. Rams, K. Krolas, G. Dhalenne, J. Phys. Condens. Matter 13 (2001) 9301. [9] N. Hamaguchi, T. Matsushita, N. Wada, Y. Yasui, M. Sato, J. Magn. Magn. Mater. 272–276 (2004) e1007. [10] D.E. MacLaughlin, M.S. Rose, J.E. Anderson, L.e.i. Shu, R.H. Heffner, T. Kimura, G. D. Morris, O.O. Bernal, Physica B 374–375 (2006) 143. [11] P. Dasgupta, Y. Jana, D. Ghosh, Solid State Commun. 139 (2006) 424. [12] S.A. Kramer, H.L. Tuller, Solid State Ionics 82 (1995) 15. [13] O. Porat, M.A. Spears, C. Heremans, I. Kosacki, H.L. Tuller, Solid State Ionics 86/ 88 (1996) 285. [14] M. Mori, G.M. Tompsett, N.M. Sammes, E. Suda, Y. Takeda, Solid State Ionics 158 (2003) 79. [15] F.W. Shi, X.J. Meng, G.S. Wang, T. Lin, J.H. Ma, Y.W. Li, J.H. Chu, J. Phys. Condens. Matter 370 (2005) 277. [16] W.J. Weber, R.C. Ewing, Science 289 (5487) (2000) 2051. [17] G.R. Lumpkin, M. Pruneda, S. Rios, K.L. Smith, K. Trachenko, K.R. Whittle, N.J. Zaluzec, J. Solid State Chem. 180 (2007) 1512. [18] M.B. Johnson, D.D. James, A. Bourque, H.A. Dabkowska, B.D. Gaulin, M.A. White, J. Solid State Chem. 182 (2009) 725. [19] G. Garton, B.M. Wanklyn, J. Mater. Sci. 3 (1968) 395. [20] J.K. Yamamoto, A.S. Bhalla, Mater. Lett. 10 (1991) 497. [21] G. Balakrishnan, O.A. Petrenko, M.R. Lees, D.M.K. Paul, J. Phys. Condens. Matter 10 (1998) L723. [22] D. Prabhakaran, A.T. Boothroyd, J. Cryst. Growth 318 (2011) 1053. [23] Q.J. Li, L.M. Xu, C. Fan, F.B. Zhang, Y.Y. Lv, B. Ni, Z.Y. Zhao, X.F. Sun, J. Cryst. Growth 377 (2013) 96. [24] P. Hartman, W.G. Perdok, Acta Crystallogr. 8 (1955) 525. [25] J.L. Waring, S.J. Schneider, J. Res. Nat. Bur. Stand. A 69 (3) (1965) 257. [26] G.M. Kuz’micheva, D.A. Lis, K.A. Subbotin, V.B. Rybakov, E.V. Zharikov, J. Cryst. Growth 275 (2005) e1837. [27] O.V. Voloshina, V.N. Baumer, V.G. Bondar, D.A. Kurtsev, T.E. Gorbacheva, I. M. Zenya, A.V. Zhukov, O.T. Sidletskiy, Nucl. Instrum. Methods Phys. Res., Sect. A 664 (2012) 301. [28] F.Y. Guo, J.J. Ru, H.Z. Li, N.F. Zhuang, B. Zhao, J.Z. Chen, J. Cryst. Growth 310 (2008) 4390.


F.Y. Guo et al. / Journal of Crystal Growth 402 (2014) 94–98

[29] K. Kawamura, H. Nakao, H. Furuya, S. Yasuda, Y.K. Yap, M. Yoshimura, Y. Mori, T. Sasaki, J. Cryst. Growth 237–239 (2002) 629. [30] J.L. Fourquet, H. Duroy, P. Lacorre, J. Solid State Chem. 114 (1995) 575. [31] M.A. Subramanian, G. Aravamudan, G.V.S. Rao, Prog. Solid State Chem. 15 (1983) 55. [32] G.D. Blundred, C.A. Bridges, M.J. Rosseinsky, Angew. Chem. Int. Ed. 43 (2004) 3562.

[33] V.V. Atuchin, T.A. Gavrilova, J.C. Grivel, V.G. Kesler, Surface Science 602 (2008) 3095. [34] N.D. Abazović, L. Mirenghi, I.A. Janković, N. Bibić, D.V. Šojić, BF Abramović, MI Čomor, Nanoscale Res. Lett. 4 (6) (2009) 518. [35] J. Chen, J. Lian, L.M. Wang, R.C. Ewing, R.G. Wang, W. Pan, Phys. Rev. Lett. 88 (2002) 105901.