Effects of silica coating on the microstructures and energy storage properties of BaTiO3 ceramics

Materials Research Bulletin 67 (2015) 70–76

Contents lists available at ScienceDirect

Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu

Effects of silica coating on the microstructures and energy storage properties of BaTiO3 ceramics Yiming Zhang, Minghe Cao * , Zhonghua Yao, Zhijian Wang, Zhe Song, Atta Ullah, Hua Hao, Hanxing Liu State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, Wuhan 430070, PR China

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 March 2014 Received in revised form 29 December 2014 Accepted 25 January 2015 Available online 28 January 2015

We investigated microstructures and energy storage properties of SiO2-coated BaTiO3 ceramics that were prepared by the so-called Stöber process. The thickness of coating layer of BaTiO3 powders could be effectively controlled by the silica content. It could be observed that the secondary phases Ba2TiSi2O8 appeared in the coated BaTiO3 ceramics due to the interdiffusion reactions between SiO2 and BaTiO3 components under sintering. BaTiO3 cores in the coated ceramics remained up to the original grain size indicating the coating layer as an inhibitor. The results showed that both breakdown strength and energy density were improved apparently. The homogeneity of silica coating in the ceramics should dominate contribution to breakdown strength, which can reduce the weak-point breakdown under high electric field. The optimized composition for BaTiO3 ceramic coated with 2.0 wt% SiO2 showed the maximum energy storage density of 1.2 J/cm3 with energy storage efficiency of 53.8%, which is about three times higher than that of pure BaTiO3 (0.37 J/cm3). ã 2015 Published by Elsevier Ltd.

Keywords: Ceramics Dielectric properties Electrical properties Energy storage

1. Introduction Recently, BaTiO3-based ceramics have been widely used in the electronic ceramic industry due to their excellent dielectric and ferroelectric properties, especially as the use of dielectrics for energy storage capacitors [1–3]. However, new applications for energy storage have been driving the demand for dielectric materials which exhibit high breakdown strength (Eb) while still retaining high polarization in order to obtain high energy density [4]. Theoretically, energy storage density (g ) of the three kinds of dielectric materials (linear dielectrics, ferroelectrics, and antiferroelectrics) can be evaluated from their P–E loops, as given by the following equation [5]: Z Dmax g¼ EdD (1) 0

where E is the applied electric field and Dmax is the electric displacement (D) at the highest applied field (Emax). For the dielectrics with high relative dielectric constant, D can be replaced by the polarization (P). Accordingly, the above formula can be written as follows [5]:

* Corresponding author. Tel.: +86 2787885813; fax: +86 2787885813. E-mail address: [email protected] (M. Cao). http://dx.doi.org/10.1016/j.materresbull.2015.01.056 0025-5408/ ã 2015 Published by Elsevier Ltd.

g¼

Z

Pmax 0

Z EdP ¼

Emax

PdE

(2)

0

Evidently, based on Eq. (2), the energy storage density of nonlinear dielectrics can be evaluated by integrating the area between the polarization axis and the curves of P–E loops. Obviously, both dielectric breakdown strength and polarization should be the key factors for the contribution of energy density of dielectric materials [6]. Although BaTiO3 ceramics exhibit a very large polarization, the energy storage values have been very limited due to the large energy loss from hysteresis with large remnant polarization and the relatively low dielectric breakdown strength of ceramics when in the form of sintered pressed disks [7,8]. In order to improve the dielectric breakdown strength, many methods have been adopted, such as controlling grain size [3,9,10], coating/mixing with low-melting point glass to remove the porosity [11–16], and forming new solid solutions with other compounds to reduce the polarization loss [17]. Among these, coating is a simple but effective method and can be operated at ambient temperature. The resultant particles by coating consist of a core made of the base particles and a shell made of coating materials, therefore, known as a core–shell structure. By coating for dielectric materials, the properties of the core component, such as reactivity and dielectric stability, may be modified.

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

71

SiO2 has been an effective coating material used in energy storage dielectrics due to high breakdown strength and low dielectric loss. Yu et al. [18,19] have reported the energy storage properties of pure and coated BaTiO3 homogeneous ceramicspolymer nanocomposites, indicating that the coating of SiO2 can obviously increase energy storage efficiency of the composites. In the present paper, we fabricate BaTiO3 nanoparticles in the ethanol/ammonia medium with the successful coating by silica shell of a few nanometers, and investigate the effects of silica coating on the microstructure and electrical properties of coated BaTiO3 ceramics. 2. Experimental The selected method was derived from the so-called Stöber process widely used for the synthesis of silica beads from a few tens to a few hundreds of nanometers [20–23]. Here, a series of compositions can be obtained by different silica contents as follows: BaTiO3 + x wt% SiO2 (x = 0, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0, 6.0, and 8.0). The silica shell thickness can be tuned by the addition of the amount of tetraethoxysilane (TEOS). BaTiO3 nanoparticles were purchased from Shandong Guoteng Functional Ceramics Co., Ltd., China. The average size of the BaTiO3 nanoparticles is 300 nm. The other analytical-reagent starting materials, such as TEOS (28.4%), acetic acid (99.5%), absolute ethanol, and ammonia water (25–28%), are purchased from Sinopharm Chemical Reagent Co., Ltd., China. It was based on the hydrolysis/condensation of tetraethoxysilane (TEOS) catalyzed by ammonia in alcoholic media. Firstly, the surface of BaTiO3 nanoparticles was activated by acetic acid treatment. 20 g of BaTiO3 nanoparticles and 50 ml of absolute ethanol with 3 ml acetic acid were combined in a roundbottomed flask and placed in a water bath (40  C) with magnetic stirring for 40 min, then sonicated for 30 min. After that, TEOS was added and then redispersed in the same condition. After that, ammonia water was slowly dropped by microburet to form so-called silica-coated BaTiO3 suspension. The obtained BT nanoparticles were washed with deionized water and then dried at 100  C for 12 h in the air. The coated BaTiO3 powders can be obtained. For comparison, the BaTiO3 + 2.0 wt% SiO2 composition was prepared by conventional solid-state synthesis (CSSS). Both BaTiO3 and SiO2 nanoparticles were selected as starting materials. Pellets of 12 mm in diameter and about 0.5 mm thickness were uniaxially pressed at 200 MPa using 5% PVA binder. Slow heating at 600  C for 2 h burned out the binder. After de-binding, these pellets were sintered in air at temperature 1250  C for 2 h by heating rate of 2  C/min. The crystalline structures of the sintered samples were examined by X-ray diffraction (XRD, PANalytical X'Pert PRO). The microstructures and thickness of coating can be revealed by scanning electron microscopy (SEM, JSM-5610LV) and transmission electron microscopy (JEM-2100F STEM/EDS) measurements, respectively. For the electrical measurements, the sintered ceramics were smoothed and coated with silver electrodes on both faces. The temperature dependence of dielectric nonlinearity was measured by a precision multi-frequency inductance capacitance resistance analyzer (Agilent E4980A) connected with an automated temperature controller with heating rate of 2  C/min. The dielectric breakdown strength and P–E hysteresis loop were examined at room temperature using a Radiant precision workstation (Radiant RT66A) based on the Sawyer-Tower circuit at 10 Hz. 3. Results and discussion The so-called Stöber process was used to fabricate the coated BaTiO3 powders to form coating layers of a few nanometers. Direct

Fig. 1. TEM pictures of BaTiO3 nanoparticles coated with different SiO2 content (a) x = 0; (b) x = 1.0; (c) x = 1.5; (d) x = 2.0; (e) x = 2.5; (f) x = 3.0; (g) x = 4.0; (h) x = 6.0; (i) x = 8.0. Scale bar is 50 nm for all pictures.

evidence of the coating of SiO2 shell on the BaTiO3 powder is provided by TEM, as shown in Fig. 1. The continuous and homogeneous coating of BaTiO3 powder can be obtained with a series of silica thickness from 0.5 nm for 1.0 wt% SiO2 to 12.0 nm for 8.0 wt% SiO2, as shown in Table 1.

72

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

Table 1 The average thickness of silica coating of BaTiO3 nanoparticles. x% @SiO2 Shell thickness (nm)

1.0 0.5

1.5 2.2

2.0 3.4

2.5 5.0

3.0 6.8

4.0 7.9

6.0 9.7

8.0 12.0

Fig. 2. XRD patterns of coated BaTiO3 ceramic sintered at 1250  C.

Fig. 2 shows the room temperature X-ray diffraction patterns of coated BaTiO3 ceramics sintered at 1250  C as a function of SiO2 content. The splitting of diffraction peaks of (0 0 2)/(2 0 0) is a symbol of tetragonal symmetry. It is quite evident that all compositions exhibit a single tetragonal symmetry and a secondary phase for coated ceramic is starting to appear when sintering. Indeed, it is difficult to control the interdiffusion and interface reactions between SiO2 and BaTiO3 phases under sintering. The secondary phases evidenced in XRD indicate the appearance of interface reactions between these two phases, which forms the Ba–Ti–Si–O compounds. Otherwise the diffraction peaks of secondary phase are consistent with Ba2TiSi2O8 shown in Fig. 2. So we consider that the secondary phase is the Ba2TiSi2O8 compound. Furthermore, the content of secondary phase increases with silica coating due to the enhancement of interface reactions. The SEM micrographs of coated BaTiO3 ceramics sintered at 1250  C without thermal etching, shown in Fig. 3 reveals obvious difference in grain size for the ceramics with different content of silica coating. The abnormal grain growth of the ceramics can be observed for the compositions without coating and 1.0 wt% SiO2 coating. However, the ceramics with high-content coating (>1.0%) show small grain sizes of BaTiO3 cores almost like the size of pristine BaTiO3 nanoparticles, as seen in the TEM picture Fig. 3(j). It can be deduced that SiO2 coating inhabits grain growth of BaTiO3 cores. Theoretically, the nano-sized SiO2 layer can form in the temperature range between 800  C and 950  C [24,25] and should accelerate the diffusion reactions due to the formation of Ba2TiSi2O8 compounds during sintering. The nano-sized coating layers as an inhibitor indicate a diffusion reaction different from the uncoated BaTiO3 ceramics, namely the contact between BaTiO3 cores is possibly broken by coating layer and therefore retain the original grain size regardless of high sintering temperatures. The temperature dependence dielectric of sintered BaTiO3 ceramics measured at 1 kHz,10 kHz, 100 kHz and 1 MHz, respectively is shown in Fig. 4(a–i). As pictures show the coating can effectively suppress the dielectric anomalous peak near ferroelectric phase transition, which can be introduced into the development of temperature-stable dielectrics. In addition the dielectric

temperature stability of sintered BaTiO3 ceramics increases with SiO2 at the cost of dielectric constant from 4065 for pure BaTiO3 ceramics to 444 for 8.0 wt% SiO2 coated BaTiO3 ceramics at room temperature, as shown in the inset picture of Fig. 4(j). The composition for 2.0 wt% SiO2 coated BaTiO3 ceramics remains a high dielectric constant (2603), as compare to other energy storage dielectrics [3,26,27]. Room-temperature P–E hysteresis loops of BaTiO3 ceramics achieved under different electric fields prior to dielectric breakdown strength are shown in Fig. 5. The polarization of coated BaTiO3 ceramics obviously depends on the content of silica coating. The calculated remnant polarization and saturated polarization are listed in Table 2. As we can see high-content coating exhibits low remnant polarization and saturated polarization values which is due to the non-ferroelectric or weak ferroelectric coating layer. However, the breakdown strength of BaTiO3 ceramic sharply increases with coating, as shown in Fig. 6. As evidenced from SEM, the enhancement of dielectric breakdown strength should be ascribed to the formation of high-resistivity coating layer and small grain sizes in contrast to pure BaTiO3 ceramic. According to Eq. (2), both dielectric breakdown strength and polarization should be responsible for energy density of dielectric materials. The energy storage performance parameters of all samples can be evaluated from the hysteresis loops, as shown in Fig. 6. The energy density g was calculated by integrating the area between the polarization axis and the discharge curve in the P–E hysteresis loop. With the increase of silica coating, the dielectric breakdown strength (Eb) improves obviously, while accompanied by the decrease of maximum polarization (Pmax). The maximum energy density (1.2 J/cm3), which is about three times higher than that of pure BaTiO3 (0.37 J/cm3), can be obtained at the optimal composition of 2.0 wt% SiO2 ceramics. This low energy density for high-content coating is a result of the low polarization level in the dielectric material. Dielectric breakdown strength is determined by both intrinsic and extrinsic (process-related) factors, but there is a general consensus that the breakdown in practical components is limited by defects introduced during processing. As comparing, BaTiO3 ceramics with 2.0 wt% SiO2 addition were prepared by conventional solid state synthesis (CSSS). The aim is to investigate the effect of the homogeneity on the breakdown strength and energy density of dielectric ceramics. The well-distributed compounds can reduce defects of the ceramics, such as weak-point breakdown. Fig. 7 shows room-temperature P–E hysteresis loops of BaTiO3 with 2.0 wt% SiO2 ceramics achieved under different electric fields prior to dielectric breakdown strength. It is obviously observed that the ceramics coated by chemical process exhibit higher breakdown strength (201.8 kV/cm) than that of the ceramics by CSSS method (110.6 kV/cm). It can be concluded that the homogeneity of these two methods is different. Theoretically, the nanopowders by chemical coating can achieve better homogeneity of the final powder than that by CSSS when sintering. This disadvantage can result in serious defects (weak-point) under high electric fields. Furthermore, the coated ceramics by chemical process exhibits higher energy density (1.2 J/cm3) than that by CSSS method (0.58 J/ cm3), which is also higher than that of uncoated BaTiO3 ceramics (0.37 J/cm3). The calculated discharge and charge energy densities and energy density efficiencies are listed in Table 3. The energy storage efficiency of the coated BaTiO3 ceramics sharply increases till to the maximum value (62.7%) at x = 2.5 then decreases with higher coating. In view of the above investigation, the enhancement of dielectric breakdown and energy density should be ascribed to the homogeneity distribution of silica composition in the ceramics, which can reduce the weak-point breakdown under high electric field.

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

73

Fig. 3. SEM pictures of BaTiO3 polished ceramics sintered at 1250  C without thermal etching (a) x = 0; (b) x = 1.0; (c) x = 1.5; (d) x = 2.0; (e) x = 2.5; (f) x = 3.0; (g) x = 4.0; (h) x = 6.0; (i) SEM picture of BaTiO3 ceramics with 2.0 wt% SiO2 by conventional solid-state synthesis; (j) TEM picture of sintered BaTiO3 ceramics with 2.0 wt% SiO2 by chemical process.

74

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

Fig. 4. Temperature dielectric dependence of BaTiO3 ceramics measured at 1 kHz, 10 kHz,100 kHz and 1 MHz. The inset is the room-temperature dielectric spectra of BaTiO3 ceramics at 1 kHz.

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

75

Fig. 5. Room-temperature polarization hysteresis (P–E) loops of BaTiO3 ceramics measured at 10 Hz.

Fig. 7. Room-temperature P–E hysteresis loops of BaTiO3 with 2.0 wt% SiO2 ceramics by different synthesis methods.

Table 2 The remnant and saturated polarization of silica-coated BaTiO3 ceramics.

4. Conclusions

x% @SiO2

Pr (mc/cm2)

Ps (mc/cm2)

0 1 1.5 2 2.5 3 4 6 8

11.688 6.718 6.705 4.842 3.786 3.192 3.957 3.546 2.769

32.660 31.757 27.808 24.915 21.935 19.777 18.507 10.946 9.644

The silica layers with a homogeneous thickness from 0.5 nm to 12.0 nm have been successfully coated on the fine-grain BaTiO3 powders by the so-called Stöber process. It has been demonstrated that the secondary phases are formed when sintering due to the diffusion reaction between silica layer and BaTiO3. The proper silica coating can effectively suppress the grain growth of BaTiO3 grains which can be ascribed to the grain boundary of silica between BaTiO3 cores. The enhancement of breakdown strength can be ascribed to the following factors: the homogeneity distribution of coating layer, high dielectric breakdown strength of coating layer, and small grain size of BaTiO3 cores. However, the homogeneity distribution of silica coating in the ceramics should dominate contribution to breakdown strength, which can reduce the weak-point breakdown under high electric field. The optimized composition for 2.0 wt% SiO2 coating showed the maximum energy storage density of 1.2 J/cm3 with energy storage efficiency of 53.8%, which is about three times higher than that of pure BaTiO3 (0.37 J/ cm3). Acknowledgements

Fig. 6. Energy storage density and breakdown strength of BaTiO3 ceramics calculated by P–E curves.

Table 3 Charge energy storage, discharge energy storage, and energy storage efficiency of silica-coated BaTiO3 ceramics. x% @SiO2

Charge (J/cm3)

Discharge (J/cm3)

Energy storage efficiency (h,%)

0 1 1.5 2 2.5 3 4 6 8

0.940 1.621 1.911 2.230 1.594 1.512 1.792 1.820 1.521

0.379 0.748 0.888 1.200 0.999 0.919 0.915 0.683 0.635

40.3 46.1 46.5 53.8 62.7 60.8 51.1 37.5 41.7

The authors would like to thank the support of key programmer of Natural Science Foundation of China (No. 50932004), the International Technology Cooperation Project from the Ministry of Science and Technology of China (2011DFA52680), Natural Science Foundation of China (No. 51372191 and No. 51102189), the Fundamental Research Funds for the Central Universities (No. 2012-IV-006) and National Basic Research Program of China (973 program: 2015CB654601). References [1] L.G.D. Silveira, M.F.S. Alves, L.F. Cótica, R.A.M. Gotardo, W.J. Nascimento, D. Garcia, J.A. Eiras, I.A. Santos, Mater. Res. Bull. 48 (2013) 1772–1777. [2] H. Tennakone, E. Park, K. Tennakone, R.C. Buchanan, Mater. Sci. Eng. B 176 (2011) 393–400. [3] Z. Song, H. Liu, S. Zhang, Z. Wang, Y. Shi, H. Hao, M. Cao, Z. Yao, Z. Yu, J. Eur. Ceram. Soc. 34 (2014) 1209–1217. [4] N.H. Fletcher, B.W. Hilton, A.D. Ricketts, Phys. D: Appl. Phys. 29 (1996) 253–258. [5] I. Burn, D.M. Smyth, J. Mater. Sci. 7 (1972) 339–343. [6] B. Chu, X. Zhou, K. Ren, B. Neese, M. Lin, Q. Wang, F. Bauer, Q.M. Zhang, Science 313 (2006) 334–336. [7] I. Burn, Bull. Am. Ceram. Soc. 50 (1971) 501–505. [8] A. Young, G. Hilmas, R.W. Zhang, S.C. Schwartz, J. Am. Ceram. Soc. 90 (2007) 1504–1510.

76

Y. Zhang et al. / Materials Research Bulletin 67 (2015) 70–76

[9] L. Curecheriu, S.B. Balmus, V. Buscaglia, L. Mitoseriu, J. Am. Ceram. Soc. 95 (2012) 3912–3921. [10] Y. Ye, S.C. Zhang, F. Dogan, E. Schamiloglu, J. Gaudet, P. Castro, M. Roybal, M. Joler, C. Christodoulou, Proc. Pulsed Power Conf. 1 (2003) 719–722. [11] Z. Li, L.A. Fredin, P. Tewari, S.A. DiBenedetto, M.T. Lanagan, M.A. Ratner, T.J. Marks, Chem. Mater. 22 (2010) 5154–5164. [12] Q. Zhang, L. Wang, J. Luo, Q. Tang, J. Du, J. Am. Ceram. Soc. 92 (2009) 1871–1873. [13] J. Li, J. Claude, L. Norena-Franco, S. Seok, Q. Wang, Chem. Mater. 20 (2008) 6304–6306. [14] X. Wang, Y. Zhang, X. Song, Z. Yuan, T. Ma, Q. Zhang, C. Deng, T. Liang, J. Eur. Ceram. Soc. 32 (2012) 559–567. [15] Z. Wu, H. Liu, M. Cao, Z. Shen, Z. Yao, D. Hao, H. Luo, J. Ceram. Soc. Jpn. 116 (2008) 345–349. [16] X. Wang, Y. Zhang, I. Baturin, T. Liang, Mater. Res. Bull. 48 (2013) 3817–3821. [17] Y. Li, H. Liu, Z. Yao, J. Xu, Y. Cui, H. Hao, M. Cao, Z. Yu, Mater. Sci. Forum 654–656 (2010) 2045–2048.

[18] [19] [20] [21] [22] [23] [24] [25] [26] [27]

K. Yu, Y. Niu, Y. Bai, Y. Zhou, H. Wang, Appl. Phys. Lett. 102 (2013) 102903. K. Yu, H. Wang, Y. Zhou, Y. Bai, Y. Niu, J. Appl. Phys. 113 (2013) 034105. W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62–69. X. Wang, Z. Shen, T. Sang, X. Cheng, M. Li, L. Chen, Z. Wang, J. Colloid Interface Sci. 341 (2010) 23–29. T. Gholamia, M. Salavati-Niasaria, M. Bazarganipourc, E. Nooria, Superlattices Microstruct. 61 (2013) 33–41. K. Tadanaga, K. Morita, K. Mori, M. Tatsumisago, J. Sol–Gel Sci. Technol. 68 (2013) 341–345. R. Mayoral, J. Requena, J.S. Moya, C. López, A. Cintas, H. Miguez, F. Meseguer, L. Vázquez, M. Holgado, Á. Blanco, Adv. Mater. 9 (1997) 257–260. H. Míguez, F. Meseguer, C. López, Á. Blanco, J.S. Moya, J. Requena, A. Mifsud, V. Fornés, Adv. Mater. 10 (1998) 480–483. Z. Wang, M. Cao, Z. Yao, G. Li, Z. Song, W. Hu, H. Hao, H. Liu, Z. Yu, Ceram. Int. 40 (2014) 929–933. J. Wang, L. Tang, B. Shen, J. Zhai, Ceram. Int. 40 (2014) 2261–2266.