Materials Letters 60 (2006) 2814 – 2818 www.elsevier.com/locate/matlet
Effect of ZnO doping on the microwave dielectric properties of LnTiNbO6 (Ln = Sm or Dy) ceramics Sam Solomon a,⁎, James T. Joseph a , H. Padma Kumar a , Jijimon K. Thomas b a
b
Department of Physics, St. John's College, Anchal, Kollam District, Kerala 691 306, India, Pin-691306 Electronic Materials Research Laboratory, Department of Physics, Mar Ivanios College, Nalanchira, Thiruvananthapuram, Kerala 695 015, India, Pin-695015 Received 19 December 2005; accepted 31 January 2006 Available online 20 February 2006
Abstract LnTiNbO6 (Ln = Sm or Dy) ceramics, doped with ZnO, were prepared in the solid-state ceramic route. The cylindrical samples were sintered at temperatures between 1260 and 1385°C. The densities were measured using Archimedes method. Samples were characterized by X-ray diffraction and scanning electron microscopic methods. Microwave dielectric properties of the cylindrical samples were measured using the network analyzer. Doping of ZnO reduced the sintering temperature and increased the dielectric constant (εr). The variation of the resonant frequency with respect to temperature was reduced with the increase in doping concentration. The unloaded quality factor (Qu × f) is also improved for low doping concentrations. These materials can be used as dielectric resonators in microwave circuits. © 2006 Elsevier B.V. All rights reserved. Keywords: Ceramics; Sintering; Doping; Dielectric resonators; Microwave dielectrics
1. Introduction Ceramic materials have high heat resistance and are either insulators or semiconductors with varying magnetic and dielectric properties. In comparison with metallic cavity resonators, ceramic materials are more suitable as dielectric resonators (DRs) in communication systems due to their compactness, thermal stability, low-cost of production, high efficiency and adaptability to microwave integrated circuits [1]. The important characteristics required for DRs are high dielectric constant for miniaturization, low loss for selectivity and low temperature variation of resonant frequency for stability. Sebastian et al. have reported the microwave dielectric properties of RETiNbO6 (RE = Ce, Pr, Nd, Sm, Eu, Gd, Tb, Y and Yb) ceramics [2]. They observed that lower rare earths (atomic numbers 57–63) form ceramics of aeschynite orthorhombic structure having a positive temperature coefficient of resonant frequency (τf) with a high dielectric constant while higher rare earths (atomic numbers 64–71) form ceramics of euxenite orthorhombic structure having a negative τf with a ⁎ Corresponding author. Fax: +91 471 2532445. E-mail address:
[email protected] (S. Solomon). 0167-577X/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2006.01.097
lower dielectric constant. Recently RETiNbO6 (RE = rare earth metal) is reported as a useful ceramic material, which finds application as dielectric resonator [3,4]. Many other dielectric resonator materials also have been reported which find application in the microwave field [5–11]. The effect of sintering time on the microwave dielectric properties and crystal structure of Y2BaZnO5 ceramic was investigated by Akinori Yoshida et al. [12]. Fang et al. have reported Ba3La2Ti2Nb2O15 and Ba2La3Ti3NbO15 as ceramics suitable for dielectric resonators [13]. Both of these materials have A5B4O15 type cation-deficient hexagonal perovskite structure having high dielectric constant and low τf value. The effect of doping on the dielectric properties of cerium oxide as a ceramic dielectric resonator is studied by Santha et al. [14]. Solomon et al. have studied the effect of bismuth oxide addition in the barium rare earth titanate system [15]. The effect of doping with MnCO3 and SnO2 on BaO–TiO2–ZnO system was studied by Wang et al. [16]. Huang and Chiang have studied the effect of CuO addition on the microwave dielectric properties and the microstructure of ZnTa2O6 ceramics [17]. They have showed that zero temperature coefficient of the resonant frequency can be achieved by properly adjusting the concentration of additives.
S. Solomon et al. / Materials Letters 60 (2006) 2814–2818
In this paper we report the effect of ZnO doping on the sintering and microwave dielectric properties of LnTiNbO6 (Ln = Sm,Dy) ceramics. The microstructure is analysed using scanning electron microscopy (SEM) and the structure is confirmed using X-ray diffraction (XRD) patterns. 2. Experimental The samples were prepared by the conventional solid-state ceramic route. Analar samples of metal oxides (CDH, 99.9%) were weighed in stoichiometric ratios and mixed thoroughly in acetone medium in an agate mortar. The samples were dried and calcined at 1200°C for 5 h in electrically heated furnace. A definite mass of calcined powder was mixed with 1wt.% zinc oxide and ground well. To this 5% polyvinyl alcohol was added as a binder and again ground well and dried. The powder was then pressed in the form of a cylindrical pellet at 100MPa pressure using a hydraulic press. In a similar way pellets were made with 2, 4 and 6wt.% doping. Samples were also prepared without doping. The pellets were then sintered as follows. Initially they were heated at a rate of 4 °C/min up to 600 °C and soaked for an hour in order to expel the polyvinyl alcohol. Then, it was heated at a rate of 5 °C/min up to the sintering temperature and soaked for 4 h. The furnace was then cooled slowly to a lower temperature. The sintered samples were polished well and the densities were measured using Archimedes method. Polished samples were thermally etched at a temperature that is 50 °C below the sintering temperature and used for SEM. Powdered samples were used for X-ray diffraction studies using Cu Kα radiation. The dielectric properties of the samples were measured in the microwave frequency range using the network analyzer. The specimen was placed at the center of a cylindrical cavity whose size is 3–4 times greater than it. The microwave was coupled to the specimen through E-field probes and TE01δ mode of resonance whose quality factor is intimately related to the dielectric loss, was identified. The dielectric constant (εr) and the unloaded quality factor (Qu × f) were then calculated using the computer interfaced network analyzer. The coefficient of thermal variation of resonant frequency (τf) was also measured over a range of temperature 20–80 °C.
2815
Table 2 Microwave dielectric properties of ZnO-doped DyTiNbO6 ceramics Percentage Frequency Qu × f Dielectric τf (ppm/K) of ZnO (GHz) (GHz) constant (εr) 0 1 2 4 6
7.7600 5.5696 5.5003 5.5264 5.5624
19 100 19 200 14 368 8898 4070
21.01 21.37 21.84 21.58 20.56
− 42 − 32 − 29 − 24 − 20
Sintering temperature (°C) 1385 1370 1370 1340 1340
respectively. The sintering temperature of the doped sample decreases with the increase in doping concentration. Fig. 1 shows that the dielectric constant and the density of SmTiNbO6 ceramics increase with the increase in the percentage of doping of ZnO initially, reach a maximum value, and then decrease. The Qu × f value also shows a similar variation as shown in Fig. 2. This is due to the formation of secondary phases in the highly doped (>2%) samples as reported earlier [18–21]. Fig. 3 shows that the τf decreases with the increase in the percentage of doping of ZnO and obtaining greater thermal stability. The 1% and 2% ZnO-doped sample with relatively high εr, high Qu × f and low τf is a suitable material to use as a DR. Fig. 4 shows that the dielectric constant and the density of DyTiNbO6 ceramics increase with the increase in the percentage of
Fig. 1. Variation of dielectric constant (εr) and density (d) of SmTiNbO6 with the wt.% of ZnO doped.
3. Results and discussion The microwave dielectric properties of Zinc oxide doped samples of SmTiNbO6 and DyTiNbO6 are given in Tables 1 and 2,
Table 1 Microwave dielectric properties of ZnO-doped SmTiNbO6 ceramics Percentage Frequency Qu × f of ZnO (GHz) (GHz)
Dielectric τf (ppm/K) constant (εr)
Sintering temperature (°C)
0 1 2 4 6
44.05 46.39 47.90 44.57 41.09
1360 1360 1340 1310 1260
4.8900 4.5444 4.3663 4.4916 4.6628
7640.0 10 367.9 14 190.2 8950.0 3749.7
50 49 48 40 31
Fig. 2. Variation of Qu × f of SmTiNbO6 with the wt.% of ZnO doped.
2816
S. Solomon et al. / Materials Letters 60 (2006) 2814–2818
Fig. 3. Variation of the temperature coefficient of resonant frequency (τf) of SmTiNbO6 with the wt.% of ZnO doped.
Fig. 6. Variation of the temperature coefficient of resonant frequency (τf) of DyTiNbO6 with the wt.% of ZnO doped.
Fig. 4. Variation of dielectric constant (εr) and density (d) of DyTiNbO6 with the wt.% of ZnO doped.
Fig. 5. Variation of Qu × f of DyTiNbO6 with the wt.% of ZnO doped.
Fig. 7. XRD patterns of (a) SmTiNbO6 and (b) DyTiNbO6 ceramics.
S. Solomon et al. / Materials Letters 60 (2006) 2814–2818
2817
Fig. 8. SEM photographs of: (a) 1% ZnO-doped SmTiNbO6; (b) 2% ZnO-doped SmTiNbO6; (c) 1% ZnO-doped DyTiNbO6; (d) 2% ZnO-doped DyTiNbO6.
doping of ZnO initially, reach a maximum value and then decreases. The Qu × f value is maximum for 1% doped sample and decreases for higher doping concentrations as shown in Fig. 5. This is also due to the formation of secondary phases in the highly doped (> 1%) samples [18–21]. Fig. 6 shows that the negative value of τf decreases with the increase in the percentage of doping of ZnO, obtaining more thermal stability. The 1% and 2% ZnO doped samples with relatively high εr, high Qu × f and low τf are suitable materials to use as a DR. The XRD patterns that are obtained using Cu Kα radiation of SmTiNbO6 samples are given in Fig. 7(a). It is evident from the XRD patterns that the main structure is aeschynite orthorhombic. In the case of highly doped samples, there is secondary phase formation. There are additional peaks at 27.02, 28.07 and 35.18 for doping concentrations of ZnO greater than 2%. The peaks at 27.02, and 35.18 are that of Zn0.17Nb0.33Ti0.5O2 (JCPDS file No. 39-291) and that at 28.07 is that of Sm2O3 (JCPDS file No. 42-1461). The XRD patterns of DyTiNbO6 are given in Fig. 7(b). The main structure is that of euxenite orthorhombic. There are additional phases at 28.46, 30.54 and 35.16 degrees for highly doped samples. These peaks are due to the secondary phases of Dysprosium–Zinc compound (JCPDS file No 26-597), Dysprosium Titanium Oxide (JCPDS file No 40-974) and ZnTiO3 (JCPDS file No 39-190) respectively. Fig. 8(a)–(d) show the surface morphology of the samples. These photographs reveal that the samples are well sintered.
4. Conclusion Ceramic materials can be doped to get materials with desirable properties for microwave applications. Secondary phases are formed due to the addition of ZnO. Addition of ZnO reduced the sintering temperature and increased the
thermal stability of the materials in this particular system. Small doping concentrations can increase the dielectric constant and the quality factor. SmTiNbO6 and DyTiNbO6 doped with 1% and 2% ZnO are found to be the most suitable materials to use as a dielectric resonator. However, work is in progress using dopants in niobate and tantalate systems to get DRs with desirable properties.
Acknowledgements We would like to express our sincere gratitude to Dr. M. T. Sebastian and Dr. Manoj Raama Varma for their help in microwave measurements. Authors Sam Solomon and H. Padma Kumar are thankful to Kerala State Council for Science, Technology and Environment (KSCSTE) for financial assistance. References [1] R.J. Cava, J. Mater. Chem. 11 (2001) 54. [2] M.T. Sebastian, S. Solomon, R. Ratheesh, J. George, P. Mohanan, J. Am. Ceram. Soc. 84 (7) (2001) 1487. [3] Sam Solomon, PhD Thesis, University of Kerala, India, 1999. [4] M.T. Sebastian, S. Solomon, R. Ratheesh, H. Sreemoolanathan, P. Mohanan, Mater. Res. Bull. 32 (1997) 1279. [5] G. Wolfram, E. Gobel, Mater. Res. Bull. 16 (1981) 1455. [6] H. Tamura, T. Kousike, Y. Takabe, K. Wakino, J. Am. Ceram. Soc. 69 (1984) C59. [7] J.K. Plourde, D.F. Linu, H.M. O'Bryan Jr., J. Thomson Jr., J. Am. Ceram. Soc. 58 (1975) 418. [8] M.T. Sebastian, J. Mater. Sci. 10 (1999) 475. [9] K. Wakino, K. Minai, H. Tamura, J. Am. Ceram. Soc. 67 (1984) 278.
2818
S. Solomon et al. / Materials Letters 60 (2006) 2814–2818
[10] H. Sreemoolanathan, M.T. Sebastian, P. Mohanan, Mater. Res. Bull. 30 (1995) 653. [11] Y. Kawashinu, M. Nishida, Y. Ueda, H. Ouchi, J. Am. Ceram. Soc. 66 (1983) 421. [12] Akinori Yoshida, Hirotaka Ogawa, Akinori Kan, Takashi Kondo, J. Eur. Ceram. Soc. 25 (2005) 2897. [13] Liang Fang, Hui Zhang, X.K. Hong, F.C. Merg, J.F. Yang, T.H. Huang, Mater. Lett. 58 (2004) 3884. [14] N. Santha, M.T. Sebastian, P. Mohanan, N.Mc N. Alford, K. Sarma, R.C. Pullar, S. Kamba, A. Pashkin, P. Samukhina, J. Petzelt, J. Am. Ceram. Soc. 87 (7) (2004) 1233.
[15] S. Solomon, N. Santha, I.N. Jawahar, H. Sreemoolanathan, M.T. Sebastian, P. Mohanan, J. Mater. Sci., Mater. Electron. 11 (2000) 595. [16] Guoqing Wang, Shunhua Wu, Hao Su, Mater. Lett. 59 (2005) 2229. [17] Cheng-Liang Huang, Kuo-Hau Chiang, Mater. Res. Bull. 39 (2004) 1701. [18] D. Kolar, Z. Stadler, S. Gaberscek, D. Suvorov, Ferroelectrics 27 (1980) 269. [19] J.M. Durand, J.P. Boilot, J. Mater. Sci. Lett. 6 (1987) 134. [20] H. Takahashi, Y. Baba, K. Ezaki, V. Okamota, K. Shibata, K. Kuroto, U.S. Patent 5244851 (1993). [21] H. Takami, S. Yano, M. Abe, U.S. Patent 5185304 (1993).