J Mater Sci: Mater Electron DOI 10.1007/s10854-015-3593-z

Magnetic and dielectric properties of Ni0.5Zn0.5Fe2O4/barium titanate (BaTiO3) ceramic composites prepared by an in situ sol–gel method Pengli Zhu1 • Qi Zheng1 • Rong Sun1

Received: 11 June 2015 / Accepted: 4 August 2015 Ó Springer Science+Business Media New York 2015

Abstract Ferrimagenitc/ferroelectric Ni0.5Zn0.5Fe2O4 (N ZFO)/BaTiO3 (BT) ceramic composites (NBSG) are synthesized via an in situ sol–gel process. The formation of the BaTiO3 layer on the Ni–Zn ferrite particles was performed in two steps, the first formation of sol BaTiO3 precursor layer and then heat treatment at high temperature. The morphology and microstructure of the ceramic composites were studied by the field-emission scanning electron microscope, X-ray diffraction measurements are used to demonstrate the phase change with different molar ratio of NZFO:BT, and their magnetic and dielectric properties of the composites have been investigated by a M–H loop measurement and RF impedance/material analyzer. Results show that the NZFO particles are well surrounded naturally by the perovskite BT layer which could help to avoid the aggregation of NZFO phase, and also in this structure, the interactions between the two constituent phases endow the ceramic composites with both good magnetic and dielectric properties at high frequency.

1 Introduction Recently, the ever-increasing demand for the high density electronic packages technology in electronic industry has greatly accelerated the miniaturization and high integration of the electronic components with smaller size, high performance, multifunctionality, high efficiency and low cost [1–5]. On the way to realize the objects of integration and & Pengli Zhu [email protected] 1

Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China

miniaturization of modern electronic devices, the multiferroic materials exhibiting ferroelectricity and ferromagnetism simultaneously have been attracting more and more attentions, as they could be used both as capacitors and inductors and other multifunctional devices, such as electromagnetic interference filters, sensors and signal processing components [6–10]. In recent years, lots of the researches have been explored and studied the intrinsic or extrinsic factors that influence the electromagnetic properties of these ferroelectric/ferromagnetic composites although the depth of understanding is still far from satisfaction [9]. However, natural multiferroic single-phase materials are very less, as the competing symmetry requirements for each kind of ferroic materials are different, for example, ferroelectricity needs d0 in the outer electron configuration, while d shell electrons of ferromagnetic materials call for being partially filled [11, 12]. Based on this limitation, many works turn to develop ferromagnetic–ferroelectric ceramic composite materials to meet the requirements of the multifunctional components [13–15]. For example, multilayered thin film of PZT (Pb(Zr0.52Ti0.48)O3) and Zn-doped cobalt ferrite (Co0.9 Zn0.1Fe2O4), with different volume fractions of each constitutive phase are fabricated and the thin film exhibited a coexistence of ferroelectric and ferromagnetic ordering [16]. Zhou et al. [17] synthesized various ferrite (Fe3O4, Fe2O3, CoFe2O4)/perovskite oxide (PbTiO3, Pb(Zr, Ti)O3, BaTiO3) core/shell nanostructures and showed that these core/shell particles could enhance the ME (magnetoelectric effect) coupling to a large extent for practical applications. Furthermore, Yu et al. reported BaTiO3–(MnZn)Fe2O4 system which has simultaneous giant capacitance (dielectric constant *105) and a rather large static permeability [18]. Overall, the comprehensive performance of this ceramic composites are greatly affected by changing their

123

J Mater Sci: Mater Electron

composition, topological microstructure and preparation method, and especially the microstructural factors, such as the porosity, density, grain size, number of interfaces and the contact effect of the constituent phases [19, 20]. Among the ferroelectric/ferromagnetic composites, the NZFO/BT ceramic composites is a most studied and typical one which owns outstanding dielectric and magnetic properties. In this ceramic composite, the magnetic properties of the ceramic composite show a nonlinear enhancement as increasing the volume fraction of the magnetic NZFO phase [21]. Nevertheless, due to the semiconductive property of the NZFO phase, near the percolation threshold, the dielectric loss would be greatly increased due to formation of conductive NZFO phase [9]. Based on this concept, in this work, we use the in situ sol– gel method to prepared ferrimagenitc/ferroelectric Ni0.5 Zn0.5Fe2O4 (NZFO)/BaTiO3 (BT) ceramic composites (NBSG) with the NZFO:BT molar ratio of 1:0, 0.2:1, 0.5:1, 0.8:1, 1:1, 1.5:1, 2:1 and 0:1. The effect of chemical composition on the morphology, crystal structure, magnetic and dielectric properties of the composites is systematic investigated. In this kind of ceramic composites, the NZFO particles are coated by the BT, in which the dielectric medium is the continuity phase and could help to avoid the aggregation of the magnetic phase and also endow the composites with good dielectric properties.

2 Experimental Iron nitrate nonahydrate [Fe(NO3)3
9H2O], zinc nitrate hexahydrate [Zn(NO3)2
6H2O] and sodium hydroxide [NaOH] were brought from Sinopharm Chemical Reagent Co. Ltd. nickel nitrate hexahydrate [Ni(NO3)2
6H2O] was purchased from Yong Da Chemical Reagent Co. Ltd. Tetrabutyl titanate [C16H36O4Ti] (from Lingfeng Chemical Reagent Co. Ltd, Shanghai, China), barium acetate [C4 H6BaO4] (from Sinopharm Chemical Reagent Co. Ltd, Shanghai, China), acetic acid [C2H4O2] (from Lingfeng Chemical Reagent Co. Ltd, Shanghai, China) were materials used in the BT synthesis. Polyvinyl alcohol (PVA) acts as the binder in the ceramic sintering process. To prepare NZFO precursor powder, Ni(NO3)2
6H2O, Zn(NO3)2
6H2O and Fe(NO3)3
9H2O of analytical grade in the molar ratio of 0.5:0.5:2 were dissolved in distilled water to form a homogeneous solution, and sodium hydroxide (NaOH) as a precipitant is added. The solution was operated at 60 °C for 4 h. After the reaction, the Ni0.5Zn0.5Fe2O4 precursor was collected via filtration and thoroughly washed to eliminate the remnant sodium ions and nitrate ions and dried at 60 °C for 12 h. The magnetic Ni0.5Zn0.5Fe2O4 precursor powders are obtained and for further use.

123

To prepare the Ni0.5Zn0.5Fe2O4 (NZFO)/BaTiO3 (BT) ceramic composites (NBSG), the surface coating BaTiO3 (BT) layer is synthesized through the in situ sol–gel method. The as-prepare NZFO precursor powders were first mixed with a certain amount of tetrabutyl titanate, acetic acid and ethyl alcohol, and the solution was stirred for 1 h. Then, barium acetate aqueous solution with stoichiometric ratio was added into the mixture slowly. The reaction solution was kept stirring until forming the gel emplaced under 50 °C in water bath. The resulting gel was further heated at 70 °C to attain the precursor of BNSG. By changing the stoichiometric proportion of NZFO/BT, NBSG ceramic precursor powders with NZFO:BT molar ratio of 1:0, 0.2:1, 0.5:1, 0.8:1, 1:1, 1.5:1, 2:1 and 0:1 were obtained, respectively. To prepare the cyclic annular and slice ceramic composites, the BNSG composite precursor powders were mixed with a few drops of PVA aqueous solution, grinded uniformly in the mortar and then added into the mold, followed by pressing at 8 MPa for the cyclic annular ceramic samples and 6 MPa for slice ceramics for 2 and 1 min, respectively. The annular and slice samples were then annealed in air at 1000 °C for 2 h. The morphology and microstructure of the composites were investigated by field-emission scanning electron microscope (FE-SEM, FEI Nova Nano SEM 450). The crystal structures of the BNSG ceramic composites were examined at room temperature through the powder Rigaku X-ray diffraction (XRD) (D/Max-2500, Japan, Cu-Ka, k = 0.15418 nm) in the 2h range from 20° to 90° with a scanning speed of 5° min-1 and under the condition of 40 kV and 100 mA. The magnetic hysteresis loop of the ceramic composites powders were obtained through Vibrating Sample Magnetometer (VSM, HH-20, Nanjing Nanda Instrument Plant, 298 K with maximum magnetic field of 2 T). The cyclic annular and slice ceramic samples were employed by Agilent E4991A precision impedance analyzer (Agilent Technology) to measure the permeability and permittivity at room temperature in a frequency range of 10 MHz–1 GHz.

3 Results and discussion The typical XRD patterns of the NBSG composite ceramic with different NZFO:BT molar ratio are shown in Fig. 1. It can be seen that for pure NZFO with good spinel structure are obtained in the samples synthesized with the NZFO:BT molar ratio of 1:0. Three characteristic peaks located at 30.02°, 35.44°, 62.56°, corresponding to the (022), (113), (044) faces of the standard Ni0.5Zn0.5Fe2O4 sample (PDF: 52-0278), respectively. Similarly, the same is true for the sample with the molar ratio of 0:1 which is denoted as pure

J Mater Sci: Mater Electron

(a)

1:0

(b) 1:0

2:1

2:1

1.5:1

1:1 0.8:1 0.5:1 0.2:1

1:1

Intensity (a.u.)

Intensity (a.u.)

1.5:1

0.5:1 0.2:1

0:1

0:1

PDF# 52-0278 PDF# 76-0744

20

30

40

50

60

70

80

90

2θ (deg)

BaFe2 O4

0.8:1

20

21

22

23

24

25

26

27

28

29

2θ (deg)

Fig. 1 XRD patterns of a BNSG with different NZFO:BT molar ratio and b the amplification from 20° to 29°

BaTiO3, owning typical perovskite structure. The sharp peaks (2h = 31.52°, 38.88°, 45.36° and 56.26°) whose corresponding to the miller indices of (101), (111), (200) and (211), fit the standard BaTiO3 sample (PDF: 76-0744) well and indicate the good crystallinity of the BaTiO3 ceramic which were prepared by the sol–gel method. The XRD patterns of the NBSG ceramic composites could be disassembled to two evident sets well-defined peaks, one of which belongs to the spinel Ni–Zn ferrite, and the other to perovskite BaTiO3. In order to further investigate whether there are any intermediate or interfacial phases, XRD patterns between 20° and 29° is given in Fig. 1b. It can be seen that some angle offset are observed and also a small peaks belong to BaFe2O4 are generated. The reaction and the ionic diffusion at the interfaces of NZFO and BT under the high sintering temperature 1000 °C may be the key factors of these tiny angle deviations from the standard samples. The NBSG precursor powders obtained from the mixture of NZFO precursors and the BT sol–gel precursors, during the following sintering process, tiny amount of Ba2? may enter into the NZFO crystal lattice to form Ba ferrite, and generate some BaFe2O4 in the composites. And this kind of miscellaneous phase could act as the barriers at the interfaces which could make adverse impact on the magnetic and dielectric properties of the ceramics, and also indicate the interaction between the two phases. In this work, the NZFO particles were first prepared through the coprecipitation method and got the precursor of NZFO, and their morphology were shown in Fig. 2a. It can be seen that the size of the NZFO precursor are quite uniform and with diameter in the range of 10–20 nm. The NZFO ceramic could be obtained after calcinating the precursor at high temperature and their grain size is nonuniform (300–800 nm) with smooth surface and high crystallization (Fig. 2b). Figure 2c, d shows the low and high magnification SEM images of the BT ceramic

particles synthesized via the sol–gel method and followed by further calcination. It reveals that the particle size is between 250 and 450 nm and the BT particles are sintered together. Combined with the XRD results, it indicates that the perovskite BaTiO3 ceramic could be successfully synthesized using the sol–gel method together with the following high temperature annealing procedure. The cross-section SEM images of the NBSG ceramic composites with NZFO:BT molar ratio at 02:1, 05:1, 08:1 and 1:1 are shown in Fig. 3. As the molar ratio of NZFO:BT is as low as 0.2:1, in which the BT content is much more higher than that of the NZFO phase, so no obvious NZFO spinel phase structures or the NZFO particles are typically well wrapped by the perovskite phase (Fig. 3a). Moreover, due to the template of the small NZFO precursor particles, the latter BT phase tend to form on the surface of ZNFO precursor particles via the sol–gel process. So it is observed that quite uniform particles with an average diameter of 475 nm were obtained as shown in Fig. 3b. As gradually increase the molar ratio of NZFO:BT to 0.5:1 and 0.8:1, the morphology of the particles become irregular both with larger and small particles (Fig. 3c–f). Further increasing the molar ratio of NZFO:BT to 1:1, it is clearly seen from Fig. 3g and h that the NZFO spinel phase and BT peroviskite phase are separated with each other. So, the above results indicates that the morphology of the NBSG ceramic composites are greatly influenced by the molar ratio of NZFO:BT. Magnetic properties of the NBSG ceramic composites were characterized by the magnetization–magnetic field (M–H) and polarization–electric field (P–E) hysteresis curves at room temperature [22]. Figure 4a shows the hysteresis curve of the pure NZFO, BT, and NBSG ceramic composites with different NZFO:BT molar ratio. All the composites materials exhibit typical magnetic-hysteresis loops as well as remanent magnetization except the BT

123

J Mater Sci: Mater Electron

Fig. 2 SEM image of the precursor of NZFO particles (a) and NZFO ceramic after calcinations (b), and bare BT ceramic prepared via the sol– gel method (c, d)

Fig. 3 SEM photographs of the transverse section of the NBSG ceramic composites with NZFO:BT molar ratio of a, b 0.2:1; c, d 0.5:1; e, f 0.8:1; g, h 1:1

which in non-magnetic materials. Simply, the value of the saturation magnetization (Ms) of the NBSG ceramic composites increase as increasing the NZFO:BT molar ratio. The detail variation of the Ms, remnant magnetization (Mr),

123

and coercivity (Hc) and calculated initial permeability (li) on the basis of the magnetic hysteresis curves of NBSG ceramic composites are listed in Table 1 and the variation trends are plotted in Fig. 4b. It is clear that the Ms and Mr

J Mater Sci: Mater Electron

100

(a)

90

(b)

200

80 75

60

175

60

M(emu/g)

20 0 BT(0:1) 0.2:1 0.5:1 0.8:1 1:1 1.5:1 2:1 NZFO(1:0)

-20 -40 -60 -80 -100

150 45

Hc

Ms

30

125

100

Mr

15

75

0 4

-1.5x10

4

-1.0x10

3

-5.0x10

0.0

3

5.0x10

4

1.0x10

4

1.5x10

Hc(Oe)

Ms (emu/g)

40

0.2:1

H (Oe)

0.5:1

0.8:1

1:1

1.5:1

2:1

1:0

The molar ratio of NZFO:BT

Fig. 4 a Room temperature magnetic hysteresis curve of NBSG ceramic composites with different NZFO:BT molar ratio; b evolution of the Mr, Ms, Hc of NBSG ceramic composites as a function of different molar ratio of NZFO:BT Table 1 The parameters of NBSG ceramic composites with different composition calculated from the M–H curves

Molar ratio of NZFO:BT

li

Ms (emu/g)

Mr (emu/g)

Hc (Oe)

0.2:1 0.5:1

206 514

11.49 29.66

1.70 3.76

198 186

0.8:1

606

37.60

6.12

163

1:1

640

38.05

6.93

151

1.5:1

797

48.07

8.08

146

2:1

884

53.98

7.67

120

1:0

1038.0

90.44

12.37

94

take on the rising trend as increasing the molar ratio of NZFO:BT, while Hc drops down. The increase of Ms and Mr and decrease of Hc is agreed with the increase of NZFO magnetic phase and non-magnetic BT phase which could acts as the obstacle to hinder the movement of the magnetic domain of NZFO. The permeability and permeability loss of the NBSG ceramic composites on applied frequency are shown in Fig. 5. The permeability increases with increasing the ferrite content in the NBSG ceramic composites. The composites of the samples with NZFO:BT molar ratio of 2:1 shows the highest initial permeability, and its value is 5.02. Further, for the ceramic composites with lower NZFO content (0.2:1, 0.5:1, 0.8:1), the permeability of the samples show good frequency stability in the whole test frequency. While ever, for the samples with higher NZFO content, the value of permeability maintains stability in a certain frequency range and then drops down as further increasing the frequency, which also named as the cut-off frequency fr. The cut-off frequency decreases with increasing the NZFO contents in the composites, which is also agreeable normally with the Snoek Rule [23]. The permeability loss of the samples is displayed in Fig. 5b and also increase with increasing the frequency. The larger

NZFO contents in the ceramic composites, the more dependency of the permeability loss on the frequency change. The comparison of the permeability value and permeability loss at 50 MHz of the NBSG ceramic composites with different NZFO:BT ratio is drawn in Fig. 5c. The variation of the permeability properties of the samples depends on the compositions could be clearly seen. In Fig. 5c, for the permeability loss, there’s an obvious turning point at 1:1 of NZFO:BT molar ratio. Overall, the permeability loss increase with increasing the molar ratio of NZFO:BT, the permeability loss major results from the NZFO additive, also some reasons come from the defects and the mutual effects between NZFO and BT. Combined with the SEM images of Fig. 3g and h for NZFO:BT = 1:1 samples, it indicates that the NZFO spinel phase and BT peroviskite phase are separated with each other and shows uniform particles morphology, which all will help to get the relatively lower permeability loss as compared with the near data points (Table 2). Usually, in the ferromagnetic materials, the quality factor Q, defined as the reciprocal of tan d, together with the value of li 9 Q (li using the data collected from the VSM) are the two important factor parameters to evaluate the performance of magnetic materials. The relative data

123

J Mater Sci: Mater Electron

5

0:1 0.2:1 0.5:1 0.8:1 1:1 1.5:1 2:1

(a)

0:1 0.2:1 0.5:1 0.8:1 1:1 1.5:1 2:1

2.0 1.5

3

tan δ

Permeability

4

(b)

2.5

1.0

2 0.5 1 0.0 10M

100M

1G

10M

100M

Frequency (Hz)

5

1G

Frequency (Hz)

(c)

0.10

0.06 0.04

3

0.02 2

Tan δ

Permeability

0.08 4

0.00 -0.02

1

-0.04 0:1

0.2:1

0.5:1

0.8:1

1:1

1.5:1

2:1

The molar ratio of NZFO:BT

Fig. 5 The variation of permeability (a) and permeability loss (b) on frequency for NBSG with different NZFO:BT molar ratio; c the variation of the NBSG ceramic system on the molar ratio of NZFO:BT at room temperature and 50 MHz

Molar ratio of NZFO:BT

0.2:1

0.8:1

1:1

1.5:1

2:1

1:0

Q (250 MHz)

3.33

2.65

1.63

1.97

0.82

0.70

0.74

li 9 Q

685.98

1363.38

987.78

1260.80

653.54

620.126

768.12

4 1400 3

1200

1000

2

800

1

Q

are listed in Table 1 and also plotted in Fig. 6. From Fig. 5b, it’s knows that the magnetic loss increase with increasing the molar ratio of NZFO:BT, which results from the increasing NZFO contents, the defects and also mutual effects between BT and NZFO, so the quality factor Q take on a declined trend as shown in Fig. 6. However, due to the increase of li and decrease of Q as increasing the molar ratio of NZFO:BT, for the value of li 9 Q, it is a little disorder, due and the sample with NZFO:BT molar ratio at 0.5:1 has the highest value of li 9 Q, which might be caused by the relatively homogeneous of the NZFO and BT phase in the NBSG ceramic composites. The dielectric constant and dielectric loss of the NBSG ceramic composites with different molar ratio of NZFO:BT in the frequency range from 10 MHz to 1 GHz are shown in Fig. 7. In Fig. 7a, the samples with NZFO:BT molar ratio at

123

0.5:1

μi x Q

Table 2 The quality factor Q and li 9 Q of the NBSG ceramics with varying the molar ratio of NZFO:BT

600 0 0.2:1

0.5:1

0.8:1

1:1

1.5:1

2:1

1:0

The molar ratio of NZFO:BT Fig. 6 The li 9 Q value and quality factor of NBSG ceramics on the molar ratio of NZFO:BT

J Mater Sci: Mater Electron

0:1 0.2:1 0.5:1 0.8:1 1:1 1.5:1 2:1 1:0

(a)

Permittivity

80 60 40

0:1 0.2:1 0.5:1 0.8:1 1:1 1.5:1 2:1

(b)

4 3 2 1

20 0 10M

5

Tan δ

100

0

100M

1G

Frequency (Hz)

-1 10M

100M

1G

Frequency (Hz)

Fig. 7 The variation of permittivity (a) and permittivity loss (b) on frequency for NBSG with different NZFO:BT molar ratio

0:1 shows the largest permittivity value of 91, which also indicates the sol–gel method prepared BT ceramic has good permittivity properties. Otherwise, as the molar ratio of NZFO:BT is 1:0, the samples indicates the lowest permittivity value only about 7, which corresponding the pure ferrite NZFO phase. Moreover, except the samples with NZFO:BT molar ratio is 0:1 and 0.2:1, the permittivity of other samples display good frequency stabilities.

4 Conclusions We have been successfully prepared NBSG ceramic composites with various molar ratios of NZFO:BT through the in situ sol–gel process together with the following sintering procedure under high temperature. During the calcinations, interphase BaFe2O4 could be generated in the composites, which could help to improve the interaction between the NZFO and BT phases. The permeability of the NBSG ceramic composites take on the growth trend with increasing the molar ratio. Considering the quality factor Q and the product of the li 9 Q, the NBSG ceramic composites with the molar ratio of 0.5:1 owns the highest value of li 9 Q and optimal dielectric behavior simultaneously. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (21101165), Guangdong Innovative Research Team Program (Nos. 2011D052 and KY PT20121228160843692), Shenzhen High Density Electronic Packaging and Device Assembly Key Laboratory (ZDSYS201405091 74237196), Shenzhen basic research plan (GJHS20120702091802836 and JCYJ20140610152828685).

References

2. Y. Shen, Z.X. Yue, M. Li, C.W. Nan, Adv. Funct. Mater. 15, 1100–1103 (2005) 3. J.Q. Huang, P.Y. Du, L.X. Hong, Y.L. Dong, M.C. Hong, Adv. Mater. 19, 437–440 (2007) 4. Y.C. Zhou, H. Wang, L. Wang, K. Yu, Z.D. Lin, L. He, Y.Y. Bai, Mater. Sci. Eng. B 177, 892–896 (2012) 5. Y.C. Zhou, Y.G. Yao, C.Y. Chen, K. Moon, H. Wang, C.P. Wong, Sci. Rep. 4, 4779 (2014) 6. C.W. Nan, M.I. Bichurin, S.X. Dong, D. Viehland, G. Srinivasan, J. Appl. Phys. 103, 1–34 (2008) 7. J. Bennett, A.J. Bell, T.J. Stevenson, R.I. Smith, I. Sterianou, I.M. Reaney, T.P. Comyn, Mater. Lett. 94, 172–175 (2013) 8. H. Schmid, Ferroelectrics 162, 317–338 (1994) 9. B. Xiao, Y. Tang, G.D. Ma, N. Ma, P.Y. Du, Appl. Phys. A 119, 1291–1300 (2015) 10. A. Mandal, C.K. Das, J. Appl. Polym. Sci. 131, 39926 (2014) 11. T.N. Narayanan, B.P. Mandal, A.K. Tyagi, A. Kumarasiri, X.B. Zhan, M.G. Hahm, M.R. Anantharaman, G. Lawes, P.M. Ajayan, Nano Lett. 12, 3025–3030 (2012) 12. R. Ramesh, N.A. Spaldin, Nat. Mater. 6, 21–29 (2007) 13. K.H. Wu, Y.M. Shin, C.C. Yang, W.D. Ho, J.S. Hsu, J. Polym. Sci. Polym. Chem. 41, 2657–2664 (2006) 14. B.W. Li, Y. Shen, Z.X. Yue, C.W. Nan, J. Appl. Phys. 99, 123909 (2006) 15. M. Liu, O. Obi, J. Lou, Y.J. Chen, Z.H. Cai, S. Stoute, M. Espanol, M. Lew, X.D. Situ, K.S. Ziemer, V.G. Harris, N.X. Sun, Adv. Funct. Mater. 19, 1826–1831 (2009) 16. H.C. He, J. Wang, J.P. Zhou, C.W. Nan, Adv. Funct. Mater. 17, 1333–1338 (2007) 17. R.Z. Liu, Y.Z. Zhao, R.X. Huang, Y.J. Zhao, H.P. Zhou, J. Mater. Chem. 20, 10665–10670 (2010) 18. G. Yu, F.M. Bai, H.W. Zhang, Adv. Mater. Sci. Eng. 2012, 109856 (2012) 19. P.S.S.R. Krishnan, Q.M. Ramasse, W.I. Liang, Y.H. Chu, V. Nagarajan, P. Munroe, J. Appl. Phys. 112, 104102 (2012) 20. A. Testino, L. Mitoseriu, V. Buscaglia, M.T. Buscaglia, I. Pallecchi, A.S. Albuquerque, V. Calzona, D. Marre, A.S. Siri, P. Nanni, J. Eur. Ceram. Soc. 26, 3031 (2006) 21. B. Xiao, N. Ma, P.Y. Du, J. Mater. Chem. C 1, 6325 (2013) 22. J.S. Kim, C.L. Cheo, Y.N. Choi, P.W. Jang, J. Appl. Phys. 93, 9263–9270 (2003) 23. H. Zheng, Y.L. Dong, X. Wang, W.J. Weng, G.R. Han, N. Ma, P.Y. Du, Angew. Chem. Int. Ed. 48, 1–5 (2009)

1. X.H. Zhang, L. Zhu, Y.L. Dong, W.J. Weng, G.R. Han, N. Ma, P.Y. Du, J. Mater. Chem. 20, 10856–10861 (2010)

123