- Email: [email protected]
Low temperature preparation of CaCu3Ti4O12 ceramics with high permittivity and low dielectric loss
Ceramics International xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Low temperature preparation of CaCu3Ti4O12 ceramics with high permittivity and low dielectric loss Yingyi Lia, Wang Lib, Guoping Dua, Nan Chena, a b
⁎
School of Materials Science and Engineering, Nanchang University, Nanchang 330031, China The Center of Collaboration and Innovation, Jiangxi University of Technology, Nanchang 330098, China
A R T I C L E I N F O
A BS T RAC T
Keywords: CaCu3Ti4O12 Low temperature sintering Dielectric property Ceramic
Low temperature preparation of CaCu3Ti4O12 ceramics with large permittivity is of practical interest for cofired multilayer ceramic capacitors. Although CaCu3Ti4O12 ceramics have been prepared at low temperatures as previously reported, they have rather low permittivity. This work demonstrates that CaCu3Ti4O12 ceramics can not only be prepared at low temperatures, but they also have large permittivity. Herein, CaCu3Ti4O12 ceramics were prepared by the solid state reaction method using B2O3 as the doping substance. It has been shown that B2O3 dopant can considerably lower the calcination and sintering temperatures to 870 °C and 920 °C, respectively. The relative permittivity of the low temperature prepared CaCu3Ti4−xBxO12 ceramics is about 5 times larger than the previously reported results in the literature. Furthermore, the dielectric loss of the CaCu3Ti4−xBxO12 ceramics is found to be as low as 0.03. This work provides a beneficial base for the future commercial applications of CaCu3Ti4O12 ceramics with large permittivity for the cofired multilayer ceramic capacitors.
1. Introduction Since the first report of colossal permittivity in the perovskite-type CaCu3Ti4O12 (CCTO) ceramics in 2000 [1], many research groups have intensively investigated this oxide [2–6]. Compared with the commercially used BaTiO3 ceramics in the electronic industry, the permittivity of CCTO ceramics has a better stability with temperature over a wide temperature range [1]. Such a low temperature dependence of the permittivity for CCTO ceramics will provide it with an advantage over the conventional BaTiO3 ceramics in practical applications. After over a decade of extensive studies on CCTO ceramics, the commercial application aspects of this material start to attract attention from researchers. One of the most important applications for colossal dielectric ceramics is the cofired multilayer ceramic capacitor (MLCC) application. In order to fabricate cofired MLCCs, a low temperature in the vicinity of 900 °C is usually required to cofire the dielectric materials with the Ag paste. CCTO ceramics are generally prepared by the conventional solid state reaction method using the raw materials CaCO3, CuO, and TiO2 [2–6]. This method is preferred for large scale production due to its simplicity and low cost. In order to prepare CCTO ceramics with the desirable dielectric properties, a relatively higher sintering temperature around 1080 °C is typically required [7–13]. However, this sintering temperature is too high for
⁎
the cofired MLCC applications. In order to lower the sintering temperatures of CCTO ceramics, researchers have employed compounds with low melting point as sintering aids [14–19]. CCTO powder is first synthesized by calcining the stoichiometric mixture of CaCO3, CuO, and TiO2. Sintering aids are mixed into the CCTO powder, and the mixture is pressed into pellets which are then sintered to obtain the CCTO ceramics. Goswami and Sen [14] used P2O5 as the sintering aid to lower the sintering temperature of CCTO ceramics to 1000 °C, which, however, is still higher than the melting point Tm of Ag (Tm =960 °C). Obviously, this is not suitable for the MLCC applications. Prakash and Varma [15] used B2O3 and BaO–B2O3–SiO2 glasses as the sintering aids and lowered the sintering temperature to 950 °C. However, they only [15] obtained a quite low relative permittivity at about 350 at 1 MHz, and the dielectric loss was high. Wang et al. [16] used SrO–B2O3–SiO2 glass as the sintering aid and lowered the sintering temperature from 1100 °C to 1050 °C, which also is too high for the MLCC applications. Löhnert et al. [17,18] used Bi2O3–B2O3–SiO2–ZnO glass as the sintering aid and lowered the sintering temperature of CCTO ceramics to 900 °C, and they fabricated MLCCs. However, the relative permittivity was also rather low, which is about 450 at 1 MHz [17,18]. Kulawik et al. [19] cofired CCTO MLCCs with Ag paste at 940 °C without using sintering aids. However, they obtained a quite low
Corresponding author. E-mail address: [email protected] (N. Chen).
http://dx.doi.org/10.1016/j.ceramint.2017.04.069 Received 14 March 2017; Received in revised form 6 April 2017; Accepted 11 April 2017 0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: Li, Y., Ceramics International (2017), http://dx.doi.org/10.1016/j.ceramint.2017.04.069
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
relative permittivity, which about 260 at 1 MHz [19]. In this work, instead of employing sintering aids to lower the sintering temperature of CCTO ceramics, we doped B2O3 into CCTO according to the formula CaCu3Ti4−xBxO12 (CCTBO). To synthesize the CCTBO, TiO2 is partially substituted by B2O3 during the calcination stage, rather than adding B2O3 into CCTO powders during the sintering stage as in Ref. [15]. B2O3, with a low melting point (Tm =450 °C), is usually used in glass preparation for forming glass network and lowering the melting temperature of raw oxide materials. It is demonstrated that B2O3 dopant can considerably lower both the calcination temperature and the sintering temperature of CCTBO. In addition, the low temperature prepared CCTBO ceramics exhibit much better dielectric performance than the previously reported results [14–19]. This work provides a beneficial base for fabricating CCTO-based MLCCs with excellent dielectric performance. 2. Experimental The raw materials for the CCTBO ceramics included CaCO3 (99.99%), CuO (99.99%), TiO2 (99.95%), and B2O3 (99.95%). The doping concentration x in the CaCu3Ti4−xBxO12 ceramics was chosen to be x=0, 0.02, 0.04, 0.06, 0.08. Stoichiometric mixtures of the raw materials were weighed and thoroughly ground. The ground powders were put in alumina crucibles, and then calcined in air in a box furnace at 870 °C for 10 h. After being ground, the powders were calcined again at 870 °C for 10 h. The calcined powders were thoroughly ground, and pressed into pellets with a diameter about 10 mm and a thickness about 2 mm under a uniaxial pressure of 200 MPa. The pellets were then placed in a box furnace, and sintered in air at 920 °C for 12 h. The samples were then naturally cooled down with the furnace to room temperature. For comparison, a pristine CCTO ceramic was prepared using the conventional processing conditions. In brief, the conventional calcination was done at temperature 1000 °C for 10 h, and it was conducted twice. The conventional sintering process was performed at 1080 °C for 12 h. Density of the CCTBO ceramics was measured using the Archimedes' method. For the measurements of their dielectric properties, the CCTO ceramics were polished and ultrasonically cleaned, and silver paste was printed on the CCTO ceramics. After being dried at 150 °C for 20 min, the CCTO ceramics were briefly heated at 600 °C for 15 min to obtain the silver electrodes. The crystal structures of the CCTBO ceramics were analyzed using the X-ray diffraction (XRD, Bruker D8 Focus, CuKα) technique, and their microstructure characteristics were observed on the freshly fractured surfaces with the scanning electron microscopy (SEM, Hitachi S-3000). The dielectric properties were measured using an Agilent 4284 A Precision LCR meter operated between 20 Hz and 1 MHz. All measurements were conducted at room temperature.
Fig. 1. XRD patterns of the CCTBO samples after the calcination (a) and after the sintering process (b).
calcination can significantly promote atomic diffusion and therefore enhances the solid state reaction process [22]. Fig. 1b shows the XRD patterns of the sintered CCTBO ceramics. For the undoped CCTBO ceramic, the presence of secondary phases, including TiO2, CuO and CaTiO3, is still notable. However, the doped CCTBO ceramics were almost free of secondary phases, although the two CCTBO ceramics with x=0.06 and 0.08 had very tiny traces of TiO2 phase. In agreement with the results for the calcined CCTBO samples, the addition of B2O3 greatly promoted the solid state reaction. The microstructures of the low temperature sintered CCTBO ceramics are shown in Fig. 2a to e. As shown in Fig. 2a, the undoped CCTBO ceramic contained only small grains with a diameter about 1– 2 µm. This indicates that the atomic diffusion and grain growth were not complete for the undoped CCTBO ceramic during the sintering process at 920 °C. The reason should be because that the sintering temperature is too low for the undoped CCTBO ceramic which typically needs to be sintered at about 1080 °C [7–13]. For the doped CCTBO ceramic with x=0.02 (Fig. 2b), it contained both small and large grains, and the number of small grains was much less than the undoped sample (Fig. 2a). The shape of the small grains in the CCTBO (x=0.02) ceramic (Fig. 2b) was spherical-like, and they also had a larger diameter. The large grains in Fig. 2b had a diameter about 5~7 µm.
3. Results and discussion 3.1. Microstructural characteristics Fig. 1a shows the XRD patterns of the calcined CCTBO samples. For the undoped CCTBO sample (x=0), the primary phase is CCTO, but some secondary phases, such as TiO2, CuO and CaTiO3, can be notably observed. This should be because the calcination temperature (870 °C) in this work is much lower than that generally used by other researchers. In the literature [20,21], the calcination temperature for CCTO is typically in the vicinity of 1000 °C. For the B2O3-doped samples (CCTBO, x=0.02, 0.04, 0.06, 0.08), it can be seen in Fig. 1a that the diffraction peaks for the secondary phases are much weaker. This suggests that the amount of secondary phases in the doped CCTBO samples is much less than in the undoped sample. Hence, B2O3 dopant enhances the solid state reaction process of CCTO. This should be because that B2O3 has a low melting point and it becomes liquid phase during calcination. The presence of liquid phase in 2
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
Fig. 2. SEM images of the low temperature sintered CCTBO ceramics with x=0 (a), x=0.02 (b), x=0.04 (c), x=0.06 (d) and x=0.08 (e), and the conventionally prepared CCTO ceramic (f). Scale bar is 10 µm.
microstructures of the conventionally prepared CCTO ceramic. The conventionally prepared CCTO ceramic exhibited well defined grain boundaries (Fig. 2f), and no blurred grain boundaries can be observed. This comparison implicitly supports the above suggestion of the possible role of liquid phase at grain boundaries for the two CCTBO ceramics with x=0.06 and 0.08. It is also noted that the average grain size of the two CCTBO ceramics (Fig. 2d and e) is comparable to the conventionally prepared CCTO ceramic (Fig. 2f). This suggests that the CCTBO ceramics can be readily sintered at 920 °C. Fig. 3 shows the variation of relative density of the low temperature prepared CCTBO ceramics with x. Densification of the CCTBO ceramics increased quickly from 78.1% for x=0 to 86.5% for x=0.02, and to 91.2% for x=0.04. For higher doping amount of B2O3, as shown in Fig. 3, the increase of the relative density of the CCTBO ceramics became saturated, and it increased gradually to 92.5% and 92.9% for x=0.06 and 0.08, respectively. These results further suggest that the densification of the CCTBO ceramics can be strongly promoted by the B2O3 dopants. For comparison, the relative density of the conventionally prepared CCTO ceramic was also included in Fig. 3. Its relative density was found to be 92.3% (Fig. 3), which is slightly lower than the CCTBO ceramics with x=0.06 and 0.08. It is noted that the relative density of the CCTBO ceramic with x=0.04 is close to the conventionally prepared CCTO ceramic. Hence, the sintering temperature of CCTBO ceramics was effectively lowered from about 1080 °C to around 920 °C, and this allows for the CCTBO ceramics to be potentially used for the fabrication of the cofired MLCCs [17–19].
This suggests that a well advanced sintering process occurred for the CCTBO ceramic. For the CCTBO ceramics with x=0.04 (Fig. 2c), the number of small grains became less, and it contained more large grains than the CCTBO with x=0.02 (Fig. 2b). Their diameter also became larger, and the grain boundaries were well defined. In addition, the grains in Fig. 2c started to contact each other more closely. The disappearance of small grains during sintering should probably be a result of the Ostwald ripening [23]. The above results strongly indicate that B2O3 dopants can strongly promote the sintering process of the CCTO ceramics. For the two CCTBO ceramics with x=0.06 and 0.08 (Fig. 2d and e), the CCTBO grains were well developed, and no small grains were observed. Their grain size was about 15~20 µm, and this is more 10 times larger than the undoped CCTBO. As indicated by the black arrows in Fig. 2d and e, some pores were present in the two ceramics. In addition, some of their grain boundaries were blurred as pointed by the white arrows in Fig. 2d and e, especially for CCTBO ceramic with x=0.08. With higher doping amount of B2O3, more liquid phase can be present in the two ceramics during sintering. On one hand, the liquid phase can greatly enhance the atomic diffusion process during sintering [24], and consequently a lower sintering temperature is to be expected [22]. On the other hand, some liquid phase will stay at grain boundaries and “solder” the bulk grains together at the cooling stage, and this may be responsible for the occurrence of blurred grain boundaries for the two ceramics (Fig. 2d and e). To further compare the grain boundaries, a pristine CCTO ceramic was prepared under the conventional conditions as described in Section 2. Fig. 2f shows the 3
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
Fig. 3. Relative density of the low temperature sintered CCTBO ceramics and the conventionally prepared CCTO ceramic.
3.2. Dielectric properties Fig. 4 shows the dielectric properties of the low temperature prepared CCTBO ceramics. For the undoped CCTBO ceramic, its relative permittivity εr’ was very low across all the measurement frequency, especially in the high frequency range. For instance, its relative permittivity was only about 30 at 1 MHz. Nevertheless, its dielectric loss tanδ was generally high, especially at the frequency range lower than 200 kHz. As shown in Fig. 4a, all of the doped CCTBO ceramics exhibited much higher relative permittivity than the undoped sample, and they had weak dependence on frequency, especially for the two samples with x=0.02 and 0.04. The CCTBO ceramic with x=0.06 had the highest relative permittivity. At 1 MHz, the relative permittivity was 1500, 1600, 2230 and 1500 for the samples with x=0.02, 0.04, 0.06 and 0.08, respectively. For lower frequency, their relative permittivity was larger (Fig. 4a). Their dielectric loss was generally lower than the undoped sample in the frequency range lower than 200 kHz. The reason for the lower εr’ with the undoped is because their grains and grain boundaries were not developed. The B2O3 doped CCTBO ceramics had more developed grains and grain boundaries as well as larger grain size. For CCTO ceramics, it has been shown that larger grain size results in higher permittivity [5,25,26] according to the following equation,
⎛a⎞ ′ ⎜ ⎟ εr′ ≈ εgb ⎝d ⎠
Fig. 4. The relative permittivities and dielectric losses of the CCTBO ceramics. Table 1 Comparison of relative permittivities of the low temperature prepared CCTO ceramics from the literature and of this work.
(1)
where εr′ is the relative permittivity of the CCTO ceramics, εgb′ is the relative permittivity of the grain boundary within the CCTO, a is the grain size, and d is the thickness of the grain boundary. This supports the much higher εr’ of the CCTBO ceramics (Fig. 4a). As shown in Fig. 4b, the dielectric loss for two CCTBO ceramics with x=0.02 and 0.04 was generally smaller than the samples with x=0.06 and 0.08. The lowest dielectric loss from the CCTBO ceramic with x=0.04 was found to be as low as tan δ=0.03 at 1 kHz. Table 1 lists the relative permittivities of the low temperature prepared CCTBO ceramics of this work and the results previously reported in the literature by other researchers [15,17–19,27]. In their work, sintering aids were used to lower the sintering temperature of CCTO ceramics [15,17,18]. Herein, B2O3 dopant was added to CCTO during the calcination stage. It can be seen from Table 1 that the low
Sintering aid (SA) or dopant (DOP)
Sintering temperature (°C)
εr’ at 10 kHz
εr’ at 100 kHz
εr’ at 1 MHz
Ref.
B2O3 (SA) BaO-B2O3-SiO2 glass (SA) Bi2O3–B2O3– SiO2–ZnO glass (SA) Without SA or DOP Without SA or DOP B2O3 (DOP)
950 950
450 900
370 600
350 400
[15] [15]
900
800
600
450
[17,18]
940
~
~
260
[19]
900
280
250
200
[27]
920
3650
3470
2230
This work
temperature prepared CCTBO ceramics of this work had about 5 times higher relative permittivity than the previous results [15,17–19], especially in the high frequencies.
4
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
where ω is the angular frequency ω=2πf, Rg and Rgb are the resistances of the semiconducting grains and the insulating grain boundaries, respectively, and Cp is the capacitance of the sample. This probably explains the lower dielectric loss of the doped CCTBO ceramics with x=0.06 and 0.08 than the undoped CCTBO sample. 4. Conclusions In summary, CaCu3Ti4−xBxO12 (x=0, 0.02, 0.04, 0.06, 0.08) ceramics were successfully prepared at a low temperature by the solid state reaction method using B2O3 as the doping substance. The B2O3 doping is found to considerably lower the calcination and sintering temperatures to 870 °C and 920 °C, respectively. The permittivity of the low temperature prepared CaCu3Ti4−xBxO12 ceramics remains to be high. Their relative permittivity is about 3650 at 10 kHz, 3470 at 100 kHz, and 2230 at 1 MHz, respectively. This is about 5 times larger than the previously reported results in the literature for low temperature prepared CCTO. In addition, the dielectric loss of the CaCu3Ti4−xBxO12 ceramics is as low as 0.03 at 1 kHz. This work shall provide a beneficial base for the possible commercial applications of the CaCu3Ti4O12 ceramics with large permittivity for cofired MLCCs.
Fig. 5. The complex impedance spectra of the CCTBO ceramics.
3.3. Impedance spectral analysis References The colossal permittivity of CCTO ceramics is generally explained with the internal barrier layer capacitance (IBLC) model [5,28]. The impedance spectroscopy is usually used to investigate the underlying factors that influence the dielectric properties of ceramics. The semicircular arc of an impedance spectrum intercepts with the Z′ axis. The intercepted Z′ value at high frequency corresponds to the bulk grain resistance, while the diameter of the arc corresponds to the resistance of the grain boundaries [28]. The impedance spectra of the CCTBO ceramics are shown in Fig. 5. The impedance arcs of the two CCTBO ceramics with x=0.02 and 0.04 had much larger diameter than the undoped sample. This indicates that the former had a much greater grain boundary resistance than the latter. Higher grain boundary resistance in CCTO ceramics leads to lower dielectric losses [28,29]. This explains why the two CCTBO ceramics with x=0.02 and 0.04 had smaller dielectric losses than the undoped sample (Fig. 4b). For the two CCTBO ceramics with x=0.06 and 0.08 as shown in Fig. 5, their impedance arcs had smaller diameter than the undoped CCTBO sample. This seems to suggest that the former had smaller grain boundary resistance than the latter. Nevertheless, further analysis based on their microstructures is needed. It is noted that the measured grain boundary resistance is a cumulative result contributing from all of the grain boundaries within each CCTBO sample. If the average grain size is smaller, then the number of grain boundaries in the CCTBO sample is larger. On the contrary, a larger average grain size results in smaller number of grain boundaries in the CCTBO sample. Comparing the microstructures of the undoped CCTBO (Fig. 2a) and the doped CCTBO with x=0.06 and 0.08 (Fig. 2d and e), the average grain size of the latter is more than 10 times larger than the former. Therefore, the doped CCTBO with x=0.06 and 0.08 had 10 times less number of grain boundaries than the undoped CCTBO sample. When averaging the measured grain boundary resistance to each grain boundary, the resistance of each individual grain boundary in the doped CCTBO samples with x=0.06 and 0.08 should be comparable to that for the undoped CCTBO sample. Another factor influencing the dielectric loss of CCTO is the bulk grain resistance. As shown in the inset of Fig. 5, all of the doped CCTBO ceramics exhibited smaller bulk grain resistance than the undoped sample. It has been shown for CCTO ceramics that smaller bulk grain resistance results in lower dielectric loss according to the following equation [30],
tan δ ≈
1 + ωRg Cp ωRgb Cp
[1] M.A. Subramanian, D. Li, N. Duan, B.A. Reisner, A.W. Sleight, High dielectric constant in ACu3Ti4O12 and ACu3Ti3FeO12 phases, J. Solid State Chem. 152 (2000) 323–325. [2] C.C. Homes, T. Vogt, S.M. Shapiro, S. Wakimoto, A.P. Ramirez, Optical response of high-dielectric-constant perovskite-related oxide, Science 293 (2001) 673–676. [3] Y. Lin, Y.B. Chen, T. Garret, S.W. Liu, C.L. Chen, L. Chen, R.P. Bontchev, A. Jacobson, J.C. Jiang, E.I. Meletis, J. Hortwiz, H.D. Wu, Epitaxial growth of dielectric CaCu3Ti4O12 thin films on (001) LaAlO3 by pulsed laser deposition, Appl. Phys. Lett. 81 (2002) 631–633. [4] S.Y. Chung, I.D. Kim, S.J. Kang, Strong nonlinear current–voltage behaviour in perovskite-derivative calcium copper titanate, Nat. Mater. 3 (2004) 774–778. [5] T.B. Adams, D.C. Sinclair, A.R. West, Giant barrier layer capacitance effects in CaCu3Ti4O12 ceramics, Adv. Mater. 14 (2002) 1321–1323. [6] A.P. Ramirez, M.A. Subramanian, M. Gardel, G. Blumberg, D. Li, T. Vogt, S.M. Shapiro, Giant dielectric constant response in a copper-titanate, Solid State Commun. 115 (2000) 217–220. [7] R. Schmidt, M.C. Stennett, N.C. Hyatt, J. Pokorny, J. Prado-Gonjal, M. Li, D.C. Sinclair, Effects of sintering temperature on the internal barrier layer capacitor (IBLC) structure in CaCu3Ti4O12 (CCTO) ceramics, J. Eur. Ceram. Soc. 32 (2012) 3313–3323. [8] J. Jumpatam, B. Putasaeng, T. Yamwong, P. Thongbai, S. Maensiri, A novel route to greatly enhanced dielectric permittivity with reduce loss tangent in CaCu3– xZnxTi4O12/CaTiO3 composites, J. Am. Ceram. Soc. 97 (2014) 2368–2371. [9] L. Singh, B.C. Sin, I.W. Kim, K.D. Mandal, H. Chung, Y. Lee, A novel one-step flame synthesis method for tungsten-doped CCTO, J. Am. Ceram. Soc. 99 (2016) 27–34. [10] H.A. Ardakani, M. Alizadeh, R. Amini, M.R. Ghazanfari, Dielectric properties of CaCu3Ti4O12 improved by chromium/lanthanum co-doping, Ceram. Int. 38 (2012) 4217–4220. [11] J. Boonlakhorn, P. Thongbai, B. Putasaeng, T. Yamwong, S. Maensiri, Very highperformance dielectric properties of Ca1-3x/2YbxCu3Ti4O12 ceramics, J. Alloy. Compd. 612 (2014) 103–109. [12] B. Zhang, Q. Zhao, A. Chang, H. Ye, S. Chen, Y. Wu, New negative temperature coefficient thermistor ceramics in Mn-doped CaCu3−xMnxTi4O12 (0≤x≤1) system, Ceram. Int. 40 (2014) 11221–11227. [13] L. Ni, X.M. Chen, Enhanced giant dielectric response in Mg-substituted CaCu3Ti4O12 ceramics, Solid State Commun. 149 (2009) 379–383. [14] S. Goswami, A. Sen, Low temperature sintering of CCTO using P2O5 as a sintering aid, Ceram. Int. 36 (2010) 1629–1631. [15] B.S. Prakash, K.B.R. Varma, Effect of the addition of B2O3 and BaO–B2O3–SiO2 glasses on the microstructure and dielectric properties of giant dielectric constant material CaCu3Ti4O12, J. Solid State Chem. 180 (2007) 1918–1927. [16] B. Wang, Y. Pu, H. Wu, K. Chen, N. Xu, Effects of SrO–B2O3–SiO2 glass additive on the microstructure and dielectric properties of CaCu3Ti4O12, J. Mater. Sci.: Mater. Electron 23 (2012) 612–617. [17] R. Löhnert, H. Bartsch, R. Schmidt, B. Capraro, J. Töpfer, Microstructure and electric properties of CaCu3Ti4O12 multilayer capacitors, J. Am. Ceram. Soc. 98 (2015) 141–147. [18] R. Löhnert, B. Capraro, S. Barth, H. Bartsch, J. Müller, J. Töpfer, Integration of CaCu3Ti4O12 capacitors into LTCC multilayer modules, J. Eur. Ceram. Soc. 35 (2015) 3043–3049. [19] J. Kulawik, D. Szwagierczak, B. Synkiewicz, Fabrication of CaCu3Ti4O12 multilayer capacitors, their microstructure and dielectric properties, Ferroelectrics 447 (2013) 19–32. [20] C.C. Wang, L.W. Zhang, Surface-layer effect in CaCu3Ti4O12, Appl. Phys. Lett. 88
(2) 5
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
[26] S.D. Almeida-Didry, C. Autret, A. Lucas, C. Honstettre, F. Pacreau, F. Gervais, Leading role of grain boundaries in colossal permittivity of doped and undoped CCTO, J. Eur. Ceram. Soc. 34 (2014) 3649–3654. [27] A.F.L. Almeida, P.B.A. Fechine, J.C. Góesb, M.A. Valente, M.A.R. Miranda, A.S.B. Sombra, Dielectric properties of BaTiO3 (BTO)–CaCu3Ti4O12 (CCTO) composite screen-printed thick films for high dielectric constant devices in the medium frequency (MF) range, Mater. Sci. Eng. B 111 (2004) 113–123. [28] D.C. Sinclair, T.B. Adams, F.D. Morrison, A.R. West, CaCu3Ti4O12: one-step internal barrier layer capacitor, Appl. Phys. Lett. 80 (2002) 2153–2155. [29] E.A. Patterson, S. Kwon, C.C. Huang, D.P. Cann, Effects of ZrO2 additions on the dielectric properties of CaCu3Ti4O12, Appl. Phys. Lett. 87 (2005) 182911. [30] Y. Yan, L. Jin, L. Feng, G. Cao, Decrease of dielectric loss in giant dielectric constant CaCu3Ti4O12 ceramics by adding CaTiO3, Mater. Sci. Eng. B 130 (2006) 146–150.
(2006) 042906. [21] S.-W. Choi, S.-H. Hong, Effect of Al doping on the electric and dielectric properties of CaCu3Ti4O12, J. Am. Ceram. Soc. 90 (2007) 4009–4011. [22] M.-H. Weng, C.-L. Huang, Single phase Ba2Ti9O20 microwave dielectric ceramics prepared by low temperature liquid phase sintering, Jpn. J. Appl. Phys. 39 (2000) 3528–3529. [23] K.-M. Kim, S.-J. Kim, J.-H. Lee, D.-Y. Kim, Microstructural evolution and dielectric properties of SiO2-doped CaCu3Ti4O12 ceramics, J. Eur. Ceram. Soc. 27 (2007) 3991–3995. [24] J.E. Marion, C.H. Hsueh, A.G. Evans, Liquid-phase sintering of ceramics, J. Am. Ceram. Soc. 70 (1987) 708–713. [25] L. Marchin, S. Guillemet-Fritsch, B. Durand, A.A. Levchenko, A. Navrotsky, T. Lebey, Grain growth-controlled giant permittivity in soft chemistry CaCu3Ti4O12 ceramics, J. Am. Ceram. Soc. 91 (2008) 485–489.
6
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Low temperature preparation of CaCu3Ti4O12 ceramics with high permittivity and low dielectric loss Yingyi Lia, Wang Lib, Guoping Dua, Nan Chena, a b
⁎
School of Materials Science and Engineering, Nanchang University, Nanchang 330031, China The Center of Collaboration and Innovation, Jiangxi University of Technology, Nanchang 330098, China
A R T I C L E I N F O
A BS T RAC T
Keywords: CaCu3Ti4O12 Low temperature sintering Dielectric property Ceramic
Low temperature preparation of CaCu3Ti4O12 ceramics with large permittivity is of practical interest for cofired multilayer ceramic capacitors. Although CaCu3Ti4O12 ceramics have been prepared at low temperatures as previously reported, they have rather low permittivity. This work demonstrates that CaCu3Ti4O12 ceramics can not only be prepared at low temperatures, but they also have large permittivity. Herein, CaCu3Ti4O12 ceramics were prepared by the solid state reaction method using B2O3 as the doping substance. It has been shown that B2O3 dopant can considerably lower the calcination and sintering temperatures to 870 °C and 920 °C, respectively. The relative permittivity of the low temperature prepared CaCu3Ti4−xBxO12 ceramics is about 5 times larger than the previously reported results in the literature. Furthermore, the dielectric loss of the CaCu3Ti4−xBxO12 ceramics is found to be as low as 0.03. This work provides a beneficial base for the future commercial applications of CaCu3Ti4O12 ceramics with large permittivity for the cofired multilayer ceramic capacitors.
1. Introduction Since the first report of colossal permittivity in the perovskite-type CaCu3Ti4O12 (CCTO) ceramics in 2000 [1], many research groups have intensively investigated this oxide [2–6]. Compared with the commercially used BaTiO3 ceramics in the electronic industry, the permittivity of CCTO ceramics has a better stability with temperature over a wide temperature range [1]. Such a low temperature dependence of the permittivity for CCTO ceramics will provide it with an advantage over the conventional BaTiO3 ceramics in practical applications. After over a decade of extensive studies on CCTO ceramics, the commercial application aspects of this material start to attract attention from researchers. One of the most important applications for colossal dielectric ceramics is the cofired multilayer ceramic capacitor (MLCC) application. In order to fabricate cofired MLCCs, a low temperature in the vicinity of 900 °C is usually required to cofire the dielectric materials with the Ag paste. CCTO ceramics are generally prepared by the conventional solid state reaction method using the raw materials CaCO3, CuO, and TiO2 [2–6]. This method is preferred for large scale production due to its simplicity and low cost. In order to prepare CCTO ceramics with the desirable dielectric properties, a relatively higher sintering temperature around 1080 °C is typically required [7–13]. However, this sintering temperature is too high for
⁎
the cofired MLCC applications. In order to lower the sintering temperatures of CCTO ceramics, researchers have employed compounds with low melting point as sintering aids [14–19]. CCTO powder is first synthesized by calcining the stoichiometric mixture of CaCO3, CuO, and TiO2. Sintering aids are mixed into the CCTO powder, and the mixture is pressed into pellets which are then sintered to obtain the CCTO ceramics. Goswami and Sen [14] used P2O5 as the sintering aid to lower the sintering temperature of CCTO ceramics to 1000 °C, which, however, is still higher than the melting point Tm of Ag (Tm =960 °C). Obviously, this is not suitable for the MLCC applications. Prakash and Varma [15] used B2O3 and BaO–B2O3–SiO2 glasses as the sintering aids and lowered the sintering temperature to 950 °C. However, they only [15] obtained a quite low relative permittivity at about 350 at 1 MHz, and the dielectric loss was high. Wang et al. [16] used SrO–B2O3–SiO2 glass as the sintering aid and lowered the sintering temperature from 1100 °C to 1050 °C, which also is too high for the MLCC applications. Löhnert et al. [17,18] used Bi2O3–B2O3–SiO2–ZnO glass as the sintering aid and lowered the sintering temperature of CCTO ceramics to 900 °C, and they fabricated MLCCs. However, the relative permittivity was also rather low, which is about 450 at 1 MHz [17,18]. Kulawik et al. [19] cofired CCTO MLCCs with Ag paste at 940 °C without using sintering aids. However, they obtained a quite low
Corresponding author. E-mail address: [email protected] (N. Chen).
http://dx.doi.org/10.1016/j.ceramint.2017.04.069 Received 14 March 2017; Received in revised form 6 April 2017; Accepted 11 April 2017 0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: Li, Y., Ceramics International (2017), http://dx.doi.org/10.1016/j.ceramint.2017.04.069
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
relative permittivity, which about 260 at 1 MHz [19]. In this work, instead of employing sintering aids to lower the sintering temperature of CCTO ceramics, we doped B2O3 into CCTO according to the formula CaCu3Ti4−xBxO12 (CCTBO). To synthesize the CCTBO, TiO2 is partially substituted by B2O3 during the calcination stage, rather than adding B2O3 into CCTO powders during the sintering stage as in Ref. [15]. B2O3, with a low melting point (Tm =450 °C), is usually used in glass preparation for forming glass network and lowering the melting temperature of raw oxide materials. It is demonstrated that B2O3 dopant can considerably lower both the calcination temperature and the sintering temperature of CCTBO. In addition, the low temperature prepared CCTBO ceramics exhibit much better dielectric performance than the previously reported results [14–19]. This work provides a beneficial base for fabricating CCTO-based MLCCs with excellent dielectric performance. 2. Experimental The raw materials for the CCTBO ceramics included CaCO3 (99.99%), CuO (99.99%), TiO2 (99.95%), and B2O3 (99.95%). The doping concentration x in the CaCu3Ti4−xBxO12 ceramics was chosen to be x=0, 0.02, 0.04, 0.06, 0.08. Stoichiometric mixtures of the raw materials were weighed and thoroughly ground. The ground powders were put in alumina crucibles, and then calcined in air in a box furnace at 870 °C for 10 h. After being ground, the powders were calcined again at 870 °C for 10 h. The calcined powders were thoroughly ground, and pressed into pellets with a diameter about 10 mm and a thickness about 2 mm under a uniaxial pressure of 200 MPa. The pellets were then placed in a box furnace, and sintered in air at 920 °C for 12 h. The samples were then naturally cooled down with the furnace to room temperature. For comparison, a pristine CCTO ceramic was prepared using the conventional processing conditions. In brief, the conventional calcination was done at temperature 1000 °C for 10 h, and it was conducted twice. The conventional sintering process was performed at 1080 °C for 12 h. Density of the CCTBO ceramics was measured using the Archimedes' method. For the measurements of their dielectric properties, the CCTO ceramics were polished and ultrasonically cleaned, and silver paste was printed on the CCTO ceramics. After being dried at 150 °C for 20 min, the CCTO ceramics were briefly heated at 600 °C for 15 min to obtain the silver electrodes. The crystal structures of the CCTBO ceramics were analyzed using the X-ray diffraction (XRD, Bruker D8 Focus, CuKα) technique, and their microstructure characteristics were observed on the freshly fractured surfaces with the scanning electron microscopy (SEM, Hitachi S-3000). The dielectric properties were measured using an Agilent 4284 A Precision LCR meter operated between 20 Hz and 1 MHz. All measurements were conducted at room temperature.
Fig. 1. XRD patterns of the CCTBO samples after the calcination (a) and after the sintering process (b).
calcination can significantly promote atomic diffusion and therefore enhances the solid state reaction process [22]. Fig. 1b shows the XRD patterns of the sintered CCTBO ceramics. For the undoped CCTBO ceramic, the presence of secondary phases, including TiO2, CuO and CaTiO3, is still notable. However, the doped CCTBO ceramics were almost free of secondary phases, although the two CCTBO ceramics with x=0.06 and 0.08 had very tiny traces of TiO2 phase. In agreement with the results for the calcined CCTBO samples, the addition of B2O3 greatly promoted the solid state reaction. The microstructures of the low temperature sintered CCTBO ceramics are shown in Fig. 2a to e. As shown in Fig. 2a, the undoped CCTBO ceramic contained only small grains with a diameter about 1– 2 µm. This indicates that the atomic diffusion and grain growth were not complete for the undoped CCTBO ceramic during the sintering process at 920 °C. The reason should be because that the sintering temperature is too low for the undoped CCTBO ceramic which typically needs to be sintered at about 1080 °C [7–13]. For the doped CCTBO ceramic with x=0.02 (Fig. 2b), it contained both small and large grains, and the number of small grains was much less than the undoped sample (Fig. 2a). The shape of the small grains in the CCTBO (x=0.02) ceramic (Fig. 2b) was spherical-like, and they also had a larger diameter. The large grains in Fig. 2b had a diameter about 5~7 µm.
3. Results and discussion 3.1. Microstructural characteristics Fig. 1a shows the XRD patterns of the calcined CCTBO samples. For the undoped CCTBO sample (x=0), the primary phase is CCTO, but some secondary phases, such as TiO2, CuO and CaTiO3, can be notably observed. This should be because the calcination temperature (870 °C) in this work is much lower than that generally used by other researchers. In the literature [20,21], the calcination temperature for CCTO is typically in the vicinity of 1000 °C. For the B2O3-doped samples (CCTBO, x=0.02, 0.04, 0.06, 0.08), it can be seen in Fig. 1a that the diffraction peaks for the secondary phases are much weaker. This suggests that the amount of secondary phases in the doped CCTBO samples is much less than in the undoped sample. Hence, B2O3 dopant enhances the solid state reaction process of CCTO. This should be because that B2O3 has a low melting point and it becomes liquid phase during calcination. The presence of liquid phase in 2
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
Fig. 2. SEM images of the low temperature sintered CCTBO ceramics with x=0 (a), x=0.02 (b), x=0.04 (c), x=0.06 (d) and x=0.08 (e), and the conventionally prepared CCTO ceramic (f). Scale bar is 10 µm.
microstructures of the conventionally prepared CCTO ceramic. The conventionally prepared CCTO ceramic exhibited well defined grain boundaries (Fig. 2f), and no blurred grain boundaries can be observed. This comparison implicitly supports the above suggestion of the possible role of liquid phase at grain boundaries for the two CCTBO ceramics with x=0.06 and 0.08. It is also noted that the average grain size of the two CCTBO ceramics (Fig. 2d and e) is comparable to the conventionally prepared CCTO ceramic (Fig. 2f). This suggests that the CCTBO ceramics can be readily sintered at 920 °C. Fig. 3 shows the variation of relative density of the low temperature prepared CCTBO ceramics with x. Densification of the CCTBO ceramics increased quickly from 78.1% for x=0 to 86.5% for x=0.02, and to 91.2% for x=0.04. For higher doping amount of B2O3, as shown in Fig. 3, the increase of the relative density of the CCTBO ceramics became saturated, and it increased gradually to 92.5% and 92.9% for x=0.06 and 0.08, respectively. These results further suggest that the densification of the CCTBO ceramics can be strongly promoted by the B2O3 dopants. For comparison, the relative density of the conventionally prepared CCTO ceramic was also included in Fig. 3. Its relative density was found to be 92.3% (Fig. 3), which is slightly lower than the CCTBO ceramics with x=0.06 and 0.08. It is noted that the relative density of the CCTBO ceramic with x=0.04 is close to the conventionally prepared CCTO ceramic. Hence, the sintering temperature of CCTBO ceramics was effectively lowered from about 1080 °C to around 920 °C, and this allows for the CCTBO ceramics to be potentially used for the fabrication of the cofired MLCCs [17–19].
This suggests that a well advanced sintering process occurred for the CCTBO ceramic. For the CCTBO ceramics with x=0.04 (Fig. 2c), the number of small grains became less, and it contained more large grains than the CCTBO with x=0.02 (Fig. 2b). Their diameter also became larger, and the grain boundaries were well defined. In addition, the grains in Fig. 2c started to contact each other more closely. The disappearance of small grains during sintering should probably be a result of the Ostwald ripening [23]. The above results strongly indicate that B2O3 dopants can strongly promote the sintering process of the CCTO ceramics. For the two CCTBO ceramics with x=0.06 and 0.08 (Fig. 2d and e), the CCTBO grains were well developed, and no small grains were observed. Their grain size was about 15~20 µm, and this is more 10 times larger than the undoped CCTBO. As indicated by the black arrows in Fig. 2d and e, some pores were present in the two ceramics. In addition, some of their grain boundaries were blurred as pointed by the white arrows in Fig. 2d and e, especially for CCTBO ceramic with x=0.08. With higher doping amount of B2O3, more liquid phase can be present in the two ceramics during sintering. On one hand, the liquid phase can greatly enhance the atomic diffusion process during sintering [24], and consequently a lower sintering temperature is to be expected [22]. On the other hand, some liquid phase will stay at grain boundaries and “solder” the bulk grains together at the cooling stage, and this may be responsible for the occurrence of blurred grain boundaries for the two ceramics (Fig. 2d and e). To further compare the grain boundaries, a pristine CCTO ceramic was prepared under the conventional conditions as described in Section 2. Fig. 2f shows the 3
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
Fig. 3. Relative density of the low temperature sintered CCTBO ceramics and the conventionally prepared CCTO ceramic.
3.2. Dielectric properties Fig. 4 shows the dielectric properties of the low temperature prepared CCTBO ceramics. For the undoped CCTBO ceramic, its relative permittivity εr’ was very low across all the measurement frequency, especially in the high frequency range. For instance, its relative permittivity was only about 30 at 1 MHz. Nevertheless, its dielectric loss tanδ was generally high, especially at the frequency range lower than 200 kHz. As shown in Fig. 4a, all of the doped CCTBO ceramics exhibited much higher relative permittivity than the undoped sample, and they had weak dependence on frequency, especially for the two samples with x=0.02 and 0.04. The CCTBO ceramic with x=0.06 had the highest relative permittivity. At 1 MHz, the relative permittivity was 1500, 1600, 2230 and 1500 for the samples with x=0.02, 0.04, 0.06 and 0.08, respectively. For lower frequency, their relative permittivity was larger (Fig. 4a). Their dielectric loss was generally lower than the undoped sample in the frequency range lower than 200 kHz. The reason for the lower εr’ with the undoped is because their grains and grain boundaries were not developed. The B2O3 doped CCTBO ceramics had more developed grains and grain boundaries as well as larger grain size. For CCTO ceramics, it has been shown that larger grain size results in higher permittivity [5,25,26] according to the following equation,
⎛a⎞ ′ ⎜ ⎟ εr′ ≈ εgb ⎝d ⎠
Fig. 4. The relative permittivities and dielectric losses of the CCTBO ceramics. Table 1 Comparison of relative permittivities of the low temperature prepared CCTO ceramics from the literature and of this work.
(1)
where εr′ is the relative permittivity of the CCTO ceramics, εgb′ is the relative permittivity of the grain boundary within the CCTO, a is the grain size, and d is the thickness of the grain boundary. This supports the much higher εr’ of the CCTBO ceramics (Fig. 4a). As shown in Fig. 4b, the dielectric loss for two CCTBO ceramics with x=0.02 and 0.04 was generally smaller than the samples with x=0.06 and 0.08. The lowest dielectric loss from the CCTBO ceramic with x=0.04 was found to be as low as tan δ=0.03 at 1 kHz. Table 1 lists the relative permittivities of the low temperature prepared CCTBO ceramics of this work and the results previously reported in the literature by other researchers [15,17–19,27]. In their work, sintering aids were used to lower the sintering temperature of CCTO ceramics [15,17,18]. Herein, B2O3 dopant was added to CCTO during the calcination stage. It can be seen from Table 1 that the low
Sintering aid (SA) or dopant (DOP)
Sintering temperature (°C)
εr’ at 10 kHz
εr’ at 100 kHz
εr’ at 1 MHz
Ref.
B2O3 (SA) BaO-B2O3-SiO2 glass (SA) Bi2O3–B2O3– SiO2–ZnO glass (SA) Without SA or DOP Without SA or DOP B2O3 (DOP)
950 950
450 900
370 600
350 400
[15] [15]
900
800
600
450
[17,18]
940
~
~
260
[19]
900
280
250
200
[27]
920
3650
3470
2230
This work
temperature prepared CCTBO ceramics of this work had about 5 times higher relative permittivity than the previous results [15,17–19], especially in the high frequencies.
4
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
where ω is the angular frequency ω=2πf, Rg and Rgb are the resistances of the semiconducting grains and the insulating grain boundaries, respectively, and Cp is the capacitance of the sample. This probably explains the lower dielectric loss of the doped CCTBO ceramics with x=0.06 and 0.08 than the undoped CCTBO sample. 4. Conclusions In summary, CaCu3Ti4−xBxO12 (x=0, 0.02, 0.04, 0.06, 0.08) ceramics were successfully prepared at a low temperature by the solid state reaction method using B2O3 as the doping substance. The B2O3 doping is found to considerably lower the calcination and sintering temperatures to 870 °C and 920 °C, respectively. The permittivity of the low temperature prepared CaCu3Ti4−xBxO12 ceramics remains to be high. Their relative permittivity is about 3650 at 10 kHz, 3470 at 100 kHz, and 2230 at 1 MHz, respectively. This is about 5 times larger than the previously reported results in the literature for low temperature prepared CCTO. In addition, the dielectric loss of the CaCu3Ti4−xBxO12 ceramics is as low as 0.03 at 1 kHz. This work shall provide a beneficial base for the possible commercial applications of the CaCu3Ti4O12 ceramics with large permittivity for cofired MLCCs.
Fig. 5. The complex impedance spectra of the CCTBO ceramics.
3.3. Impedance spectral analysis References The colossal permittivity of CCTO ceramics is generally explained with the internal barrier layer capacitance (IBLC) model [5,28]. The impedance spectroscopy is usually used to investigate the underlying factors that influence the dielectric properties of ceramics. The semicircular arc of an impedance spectrum intercepts with the Z′ axis. The intercepted Z′ value at high frequency corresponds to the bulk grain resistance, while the diameter of the arc corresponds to the resistance of the grain boundaries [28]. The impedance spectra of the CCTBO ceramics are shown in Fig. 5. The impedance arcs of the two CCTBO ceramics with x=0.02 and 0.04 had much larger diameter than the undoped sample. This indicates that the former had a much greater grain boundary resistance than the latter. Higher grain boundary resistance in CCTO ceramics leads to lower dielectric losses [28,29]. This explains why the two CCTBO ceramics with x=0.02 and 0.04 had smaller dielectric losses than the undoped sample (Fig. 4b). For the two CCTBO ceramics with x=0.06 and 0.08 as shown in Fig. 5, their impedance arcs had smaller diameter than the undoped CCTBO sample. This seems to suggest that the former had smaller grain boundary resistance than the latter. Nevertheless, further analysis based on their microstructures is needed. It is noted that the measured grain boundary resistance is a cumulative result contributing from all of the grain boundaries within each CCTBO sample. If the average grain size is smaller, then the number of grain boundaries in the CCTBO sample is larger. On the contrary, a larger average grain size results in smaller number of grain boundaries in the CCTBO sample. Comparing the microstructures of the undoped CCTBO (Fig. 2a) and the doped CCTBO with x=0.06 and 0.08 (Fig. 2d and e), the average grain size of the latter is more than 10 times larger than the former. Therefore, the doped CCTBO with x=0.06 and 0.08 had 10 times less number of grain boundaries than the undoped CCTBO sample. When averaging the measured grain boundary resistance to each grain boundary, the resistance of each individual grain boundary in the doped CCTBO samples with x=0.06 and 0.08 should be comparable to that for the undoped CCTBO sample. Another factor influencing the dielectric loss of CCTO is the bulk grain resistance. As shown in the inset of Fig. 5, all of the doped CCTBO ceramics exhibited smaller bulk grain resistance than the undoped sample. It has been shown for CCTO ceramics that smaller bulk grain resistance results in lower dielectric loss according to the following equation [30],
tan δ ≈
1 + ωRg Cp ωRgb Cp
[1] M.A. Subramanian, D. Li, N. Duan, B.A. Reisner, A.W. Sleight, High dielectric constant in ACu3Ti4O12 and ACu3Ti3FeO12 phases, J. Solid State Chem. 152 (2000) 323–325. [2] C.C. Homes, T. Vogt, S.M. Shapiro, S. Wakimoto, A.P. Ramirez, Optical response of high-dielectric-constant perovskite-related oxide, Science 293 (2001) 673–676. [3] Y. Lin, Y.B. Chen, T. Garret, S.W. Liu, C.L. Chen, L. Chen, R.P. Bontchev, A. Jacobson, J.C. Jiang, E.I. Meletis, J. Hortwiz, H.D. Wu, Epitaxial growth of dielectric CaCu3Ti4O12 thin films on (001) LaAlO3 by pulsed laser deposition, Appl. Phys. Lett. 81 (2002) 631–633. [4] S.Y. Chung, I.D. Kim, S.J. Kang, Strong nonlinear current–voltage behaviour in perovskite-derivative calcium copper titanate, Nat. Mater. 3 (2004) 774–778. [5] T.B. Adams, D.C. Sinclair, A.R. West, Giant barrier layer capacitance effects in CaCu3Ti4O12 ceramics, Adv. Mater. 14 (2002) 1321–1323. [6] A.P. Ramirez, M.A. Subramanian, M. Gardel, G. Blumberg, D. Li, T. Vogt, S.M. Shapiro, Giant dielectric constant response in a copper-titanate, Solid State Commun. 115 (2000) 217–220. [7] R. Schmidt, M.C. Stennett, N.C. Hyatt, J. Pokorny, J. Prado-Gonjal, M. Li, D.C. Sinclair, Effects of sintering temperature on the internal barrier layer capacitor (IBLC) structure in CaCu3Ti4O12 (CCTO) ceramics, J. Eur. Ceram. Soc. 32 (2012) 3313–3323. [8] J. Jumpatam, B. Putasaeng, T. Yamwong, P. Thongbai, S. Maensiri, A novel route to greatly enhanced dielectric permittivity with reduce loss tangent in CaCu3– xZnxTi4O12/CaTiO3 composites, J. Am. Ceram. Soc. 97 (2014) 2368–2371. [9] L. Singh, B.C. Sin, I.W. Kim, K.D. Mandal, H. Chung, Y. Lee, A novel one-step flame synthesis method for tungsten-doped CCTO, J. Am. Ceram. Soc. 99 (2016) 27–34. [10] H.A. Ardakani, M. Alizadeh, R. Amini, M.R. Ghazanfari, Dielectric properties of CaCu3Ti4O12 improved by chromium/lanthanum co-doping, Ceram. Int. 38 (2012) 4217–4220. [11] J. Boonlakhorn, P. Thongbai, B. Putasaeng, T. Yamwong, S. Maensiri, Very highperformance dielectric properties of Ca1-3x/2YbxCu3Ti4O12 ceramics, J. Alloy. Compd. 612 (2014) 103–109. [12] B. Zhang, Q. Zhao, A. Chang, H. Ye, S. Chen, Y. Wu, New negative temperature coefficient thermistor ceramics in Mn-doped CaCu3−xMnxTi4O12 (0≤x≤1) system, Ceram. Int. 40 (2014) 11221–11227. [13] L. Ni, X.M. Chen, Enhanced giant dielectric response in Mg-substituted CaCu3Ti4O12 ceramics, Solid State Commun. 149 (2009) 379–383. [14] S. Goswami, A. Sen, Low temperature sintering of CCTO using P2O5 as a sintering aid, Ceram. Int. 36 (2010) 1629–1631. [15] B.S. Prakash, K.B.R. Varma, Effect of the addition of B2O3 and BaO–B2O3–SiO2 glasses on the microstructure and dielectric properties of giant dielectric constant material CaCu3Ti4O12, J. Solid State Chem. 180 (2007) 1918–1927. [16] B. Wang, Y. Pu, H. Wu, K. Chen, N. Xu, Effects of SrO–B2O3–SiO2 glass additive on the microstructure and dielectric properties of CaCu3Ti4O12, J. Mater. Sci.: Mater. Electron 23 (2012) 612–617. [17] R. Löhnert, H. Bartsch, R. Schmidt, B. Capraro, J. Töpfer, Microstructure and electric properties of CaCu3Ti4O12 multilayer capacitors, J. Am. Ceram. Soc. 98 (2015) 141–147. [18] R. Löhnert, B. Capraro, S. Barth, H. Bartsch, J. Müller, J. Töpfer, Integration of CaCu3Ti4O12 capacitors into LTCC multilayer modules, J. Eur. Ceram. Soc. 35 (2015) 3043–3049. [19] J. Kulawik, D. Szwagierczak, B. Synkiewicz, Fabrication of CaCu3Ti4O12 multilayer capacitors, their microstructure and dielectric properties, Ferroelectrics 447 (2013) 19–32. [20] C.C. Wang, L.W. Zhang, Surface-layer effect in CaCu3Ti4O12, Appl. Phys. Lett. 88
(2) 5
Ceramics International xxx (xxxx) xxx–xxx
Y. Li et al.
[26] S.D. Almeida-Didry, C. Autret, A. Lucas, C. Honstettre, F. Pacreau, F. Gervais, Leading role of grain boundaries in colossal permittivity of doped and undoped CCTO, J. Eur. Ceram. Soc. 34 (2014) 3649–3654. [27] A.F.L. Almeida, P.B.A. Fechine, J.C. Góesb, M.A. Valente, M.A.R. Miranda, A.S.B. Sombra, Dielectric properties of BaTiO3 (BTO)–CaCu3Ti4O12 (CCTO) composite screen-printed thick films for high dielectric constant devices in the medium frequency (MF) range, Mater. Sci. Eng. B 111 (2004) 113–123. [28] D.C. Sinclair, T.B. Adams, F.D. Morrison, A.R. West, CaCu3Ti4O12: one-step internal barrier layer capacitor, Appl. Phys. Lett. 80 (2002) 2153–2155. [29] E.A. Patterson, S. Kwon, C.C. Huang, D.P. Cann, Effects of ZrO2 additions on the dielectric properties of CaCu3Ti4O12, Appl. Phys. Lett. 87 (2005) 182911. [30] Y. Yan, L. Jin, L. Feng, G. Cao, Decrease of dielectric loss in giant dielectric constant CaCu3Ti4O12 ceramics by adding CaTiO3, Mater. Sci. Eng. B 130 (2006) 146–150.
(2006) 042906. [21] S.-W. Choi, S.-H. Hong, Effect of Al doping on the electric and dielectric properties of CaCu3Ti4O12, J. Am. Ceram. Soc. 90 (2007) 4009–4011. [22] M.-H. Weng, C.-L. Huang, Single phase Ba2Ti9O20 microwave dielectric ceramics prepared by low temperature liquid phase sintering, Jpn. J. Appl. Phys. 39 (2000) 3528–3529. [23] K.-M. Kim, S.-J. Kim, J.-H. Lee, D.-Y. Kim, Microstructural evolution and dielectric properties of SiO2-doped CaCu3Ti4O12 ceramics, J. Eur. Ceram. Soc. 27 (2007) 3991–3995. [24] J.E. Marion, C.H. Hsueh, A.G. Evans, Liquid-phase sintering of ceramics, J. Am. Ceram. Soc. 70 (1987) 708–713. [25] L. Marchin, S. Guillemet-Fritsch, B. Durand, A.A. Levchenko, A. Navrotsky, T. Lebey, Grain growth-controlled giant permittivity in soft chemistry CaCu3Ti4O12 ceramics, J. Am. Ceram. Soc. 91 (2008) 485–489.
6