Influence of strontium on manganese-doped barium titanate ceramics

Influence of strontium on manganese-doped barium titanate ceramics

January 2000 Materials Letters 42 Ž2000. 21–24 www.elsevier.comrlocatermatlet Influence of strontium on manganese-doped barium titanate ceramics H.T...

441KB Sizes 12 Downloads 164 Views

January 2000

Materials Letters 42 Ž2000. 21–24 www.elsevier.comrlocatermatlet

Influence of strontium on manganese-doped barium titanate ceramics H.T. Langhammer a b


b , T. Muller , K.-H. Felgner a , H.-P. Abicht ¨


Fachbereich Physik, Martin-Luther-UniÕersitat ¨ Halle-Wittenberg, D-06099, Halle, Germany Fachbereich Chemie, Martin-Luther-UniÕersitat ¨ Halle-Wittenberg, D-06099, Halle, Germany Received 30 April 1999; accepted 15 June 1999

Abstract The influence of Sr on the crystal structure, microstructure, and thermal expansion of Ba 0.98 Sr0.02Ti 1yx Mn xO 3 ceramics Ž0 F x F 0.02. was investigated. Compared to Sr-‘‘free’’ samples the transition region between tetragonal and hexagonal phase is shifted to significant higher Mn concentrations. Mn contents higher than 2.0 mol% strongly hinder grain growth at 14008C. The Goldschmidt tolerance factor is used to discuss the stabilization of the tetragonal phase by the Sr impurity. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Barium titanate ŽBaTiO 3 .; Mn-doped; Sn-doped; Tetragonal–hexagonal phase transition

1. Introduction Manganese as a doping element in BaTiO 3 ceramics is used for several purposes. The main applications concern capacitors and devices exhibiting an extraordinarily high positive temperature coefficient of resistance ŽPTCR. where Mn influences both the electrical properties and the microstructure due to it being an effective acceptor dopant. Much work has been done to elucidate the role of Mn regarding these effects Žsee e.g., Refs. w1–5x.. Considerably less was published about the influence of Mn on the crystal structure Žsee e.g., Ref. w6x., and the microstructure Žsee Ref. w7x.. Continuing our studies on the influence of manganese on these properties of BaTiO 3 ceramics, this paper is dedicated to the ) Corresponding author. Tel.: q49-345-552542; fax: q49-3455527595; E-mail: [email protected]

effect of the Sr Ba impurity in Mn-doped BaTiO 3 with respect to the properties mentioned above. Furthermore, this paper, together with the publication w8x, serve as a correction of our former publication w7x related to Mn-doped BaTiO 3 ceramics. In that paper, the experimental results of the microstructure, of X-ray diffraction ŽXRD. and dilatometric investigations at a doping level between 0.5 and 1.5 mol% Mn applied to BaTiO 3 with an unfortunately unnoticed Sr impurity of approximately 2 mol%. The Sr content of all other samples amounted to at least one order less. The reported drastic change both of the microstructure and of the crystal structure between 1.5 and 1.8 mol% Mn does not occur in practice but is more broadened for Sr-‘‘free’’ samples as well as for the samples with significant Sr impurity. Strontium is one of the main impurities in barium carbonate because of the great chemical similarity

00167-577Xr00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 5 7 7 X Ž 9 9 . 0 0 1 5 3 - 6


H.T. Langhammer et al.r Materials Letters 42 (2000) 21–24

between these elements. Often, the Sr impurity is harmless or, on the contrary, quite advantageous because of the inhibition of the grain growth, especially, to make dielectrics. But the significant influence of Sr on the stability of the cubic Žperovskite. modification of BaTiO 3 cannot be neglected if phase transitions in barium titanate are the aim of the investigations. In this paper, we report on systematic investigations of the microstructure, XRD, the solubility of Mn and the linear thermal expansion of Mn-doped Ba 0.98 Sr0.02TiO 3 ceramics. It is the parallel work to similar investigations of Sr-‘‘free’’ Mn-doped BaTiO 3 w8x.

Measurements of the linear thermal expansion coefficient a were carried out on a SETARAM TMA 92 dilatometer with heating and cooling rates of 1–5 Krmin to minimize hysteresis effects caused by temperature measurement. To improve the measuring precision, rectangular samples with a geometry of 20 = 6 = 3 mm3 were prepared.

3. Results The results of the average grain size, of the XRD-determined phase composition Žroom tempera-

2. Experimental procedure The ceramic powder with a nominal composition of Ba 0.98 Sr0.02Ti 1yx Mn xO 3 Ž0 F x F 0.02. was prepared by the classical mixed-oxide powder technique. After mixing Žagate balls, water. and calcination Ž11008C, 2 h. of appropriate amounts of BaCO 3 ŽMerck, Darmstadt, No. 1713, containingf 2 mol% Sr., TiO 2 ŽMerck, No. 808. and MnCO 3 ŽRiedel de Haen, p.a.., the powder was milled again and densified to disks with a diameter of 12 mm and a height of nearly 3 mm. The samples were sintered in air at Ts s 14008C for 1 h. To avoid interfering contamination during sintering, they were contained in ZrO 2 covered Al 2 O 3 dishes. The microstructure of the polished and chemically etched specimens was examined by optical microscopy and by scanning electron microscopy ŽSEM.. To determine the distribution of Mn in the grains and intergranular regions wavelength dispersive X-ray electron probe microanalysis ŽWDXEPMA. was performed ŽCAMECA, model CAMEBAX.. For the XRD investigations ŽFreiberger Prazisionsmechanik, model URD 63., the sintered ¨ samples were crushed again and mixed with silicon powder for the purpose of calibration. To estimate the occurring high temperature phases, the samples were heated up to temperatures between 1480 and 15808C and then quenched. The phase composition was determined quantitatively by analyzing of the intensity ratios Ž111. tetragonalrŽ103. hexagonal and Ž200. tetragonalrŽ103. hexagonal ŽSiemens D5000 diffractometer..

Fig. 1. Average grain size, XRD-determined content of tetragonal phase and dilatometric data Žjump of D l r l near T Ž a min . and phase transition temperature Ttc Ž T Ž a min .. in dependence on the Mn content x of Ba 0.98 Sr0.02Ti 1yx Mn x O 3 ceramics. The percentage values at the grain size curves denote roughly estimated area portions of the different grain fractions Žfor x G 0.015.. It should be noted that the presented grain sizes are roughly estimated values, especially in the case of plate-like grains which are obtained by random cuts in the two-dimensional polished plane.


H.T. Langhammer et al.r Materials Letters 42 (2000) 21–24


Fig. 2. Optical micrographs of the microstructure of Ba 0.98 Sr0.02Ti 1yx Mn x O 3 ceramics with x s 0 Ža., x s 0.015 Žb., and x s 0.02 Žc., respectively. All graphs exhibit the same magnification.

ture. and of the dilatometric investigations of Srdoped samples are shown in Fig. 1. Compared with Sr-‘‘free’’ samples Žsee Ref. w8x. the transition region tetragonal–hexagonal is shifted to higher Mn concentrations. Up to 1.6 mol% Mn the ceramic is completely tetragonal. For a higher Mn concentration, the tetragonal portion decreases until 80% at 2.0 mol%. Since at Mn concentrations higher than 2.0 mol% a sintering temperature of 14008C Ž1 h. is not sufficient for a complete grain growth, only results up to 2.0 mol% Mn are presented. The existence of a minimum Mn level for producing entire hexagonal phase is not known. Similar to the results of Sr-‘‘free’’ samples, the occurrence of the hexagonal phase corresponds to the developing of a bimodal microstructure consisting of globular and plate-like grains w8x. But such large exaggerated plate-like grains like those of the samples without Sr are not detected. Optical micrographs of the microstructure of samples with 0, 1.5, and 2.0 mol% nominal Mn content are shown in Fig. 2. It was proved by EPMA that nearly all manganese ions are incorporated into the lattice of the grown grains up to a nominal concentration of 2 mol%. To confirm the change of the crystallographic structure, dilatometric measurements were performed in the range around the phase transition temperature tetragonal-cubic, Ttc , which was taken as the temperature of the minimum linear expansion coefficient. Corresponding to the growing portion of hexagonal phase, the anomaly of the thermal expansion becomes smaller. The temperature Ttc decreases with

increasing Mn content which also has been found by dielectric investigations reported elsewhere w1–4x. The stabilization of the cubic Žat room temperature tetragonal. phase by the incorporation of strontium is illustrated in Fig. 3. The phase transition cubic-hexagonal in undoped barium titanate with a

Fig. 3. XRD patterns of undoped Ža. and Sr-doped Žb. BaTiO 3 ceramics at selected temperatures. The curves for the different temperatures are vertically shifted for clarity.


H.T. Langhammer et al.r Materials Letters 42 (2000) 21–24

little surplus of Ti is detectable at T G 15408C Žsee also Ref. w9x.. Samples with 2.0 mol% Sr show no phase transition up to 15808C. 4. Discussion The discussion of the Goldschmidt tolerance factor of perovskites ABO 3 rA q rO ts '2 Ž r B q rO . Ž rA , r B and rO are the effective ionic radii of the A site, the B site and the oxide ion, respectively. as a criterion of the stability of the cubic Žperovskite. phase is used by Ren et al. w10x who showed that the substitution of the Ba2q-ion by the smaller Sr 2q-ion reduces t from 1.06 ŽBaTiO 3 . to 1.00 ŽSrTiO 3 . and the cubic phase gains stability Žthe values of the effective ionic radii were taken from Shannon w11x.. It is well-known that in pure strontium titanate there exists no hexagonal high temperature phase w12x. In barium titanate without Mn-doping, a content of 2 mol% strontium totally suppresses the formation of hexagonal phase up to 15808C ŽFig. 3b.. For comparison with pure barium titanate see Fig. 3a. The shift of the phase transition cubic-hexagonal to higher temperatures than 14608C is due to a Ti-enrichment by water milling during the powder preparation w13x and in agreement with w9x. Our systematic investigations show that in the same way the effect of strontium on Mn-doped barium titanate can be understood. Air-sintered Mndoped barium titanate exhibits oxidation numbers of Mn of q3 and q4 with a mean value of nearly 3.3 w10x. Hence, the observed stabilization of the hexagonal phase by Mn can be explained by the ionic radius of Mn4q Ženlargement of the Goldschmidt factor t because the radius of Mn4q is smaller than the radius of Ti 4q . and by the electronic state of Mn3q Žd 4-configuration. and the resulting strong Jahn– Teller-distortion Žsee also Ref. w8x.. Because of the smaller ionic radius of Sr 2q compared with Ba2q ions, the substitution of Ba2q ions by Sr 2q countereffects the enlargement of t by Mn4q ions. This is clearly shown by the shift of the transition region tetragonal–hexagonal to distinctly higher values of Mn concentration compared with Sr-‘‘free’’ samples Žsee Fig. 1 and Ref. w8x..

On the other hand, the occurrence of hexagonal phase also in Mn-doped Ba 0.98 Sr0.02TiO 3 shows that the fundamental effect of the Jahn–Teller-distortion caused by the Mn3q ions is not suppressed by strontium. 5. Summary Ž1. The hexagonal phase stabilizing effect of Mn in BaTiO 3 is partly suppressed by the Sr Ba impurity. Ž2. Sr-doped BaTiO 3 shows hexagonal phase only at Mn concentrations ) 1.5 mol%, whereas 0.5 mol% Mn is sufficient to stabilize traces of hexagonal phase in Sr-‘‘free’’ samples. Ž3. At Mn concentrations higher than 2.0 mol%, a sintering temperature of 14008C Ž1 h. is not sufficient for a complete grain growth in Sr-doped BaTiO 3 . Acknowledgements The authors would like to thank Dr. Pollandt ŽFB Chemie. and Dr. Eisenschmidt ŽFB Physik. from the University of Halle for their support in the XRD investigations. Financial support from the Kultusministerium des Landes Sachsen-Anhalt and the Deutsche Forschungsgemeinschaft is gratefully acknowledged. References w1x F. Batllo, E. Duverger, J.-C. Jules, J.-C. Niepce, B. Jannot, M. Maglione, Ferroelectrics 109 Ž1990. 113. w2x I. Burn, J. Mater. Sci. 14 Ž1979. 2453. w3x H.-J. Hagemann, H. Ihrig, Phys. Rev. B 20 Ž1979. 3871. w4x S.B. Desu, E.C. Subbarao, Ferroelectrics 37 Ž1981. 665. w5x Y.-C. Chen, G.-M. Lo, C.-R. Shih, L. Wu, M.-H. Chen, K.-C. Huang, Jpn. J. Appl. Phys. 33 Ž1994. 1412. w6x R.M. Glaister, H.F. Kay, Proc. Phys. Soc. 76 Ž1960. 763. w7x H.T. Langhammer, T. Muller, A. Polity, K.-H. Felgner, H.-P. ¨ Abicht, Mater. Lett. 26 Ž1996. 205. w8x H.T. Langhammer, T. Muller, K.-H. Felgner, H.-P. Abicht, ¨ to be published in J. Ceram. Am. Soc., 1999. w9x K.W. Kirby, B.A. Wechsler, J. Am. Ceram. Soc. 74 Ž1991. 1841. w10x F. Ren, S. Ishida, S. Mineta, J. Ceram. Soc. Jpn. 102 Ž1994. 106. w11x R.D. Shannon, Acta Crystallogr. A 32 Ž1976. 751. w12x M.J. Weber, R.R. Allen, J. Chem. Phys. 38 Ž1963. 726. w13x H.-P. Abicht, D. Voltzke, A. Roder, R. Schneider, J. ¨ ¨ Woltersdorf, J. Mater. Chem. 7 Ž1997. 487.