ARTICLE IN PRESS

0304-8853/$

doi:10.1016

�Correspsity of Tec

Tel.: +431

E-mail a

reiko.sato@

Journal of Magnetism and Magnetic Materials 310 (2007) e361–e363

www.elsevier.com/locate/jmmm

Effect of structure on magnetoelectric properties ofCoFe2O4–BaTiO3 multiferroic composites

Giap V. Duonga,b, R. Groessingera, R. Sato Turtellia,�

aInstitute of Solid State Physics, Vienna University of Technology, Wiedner Hauptstrasse 8-10, 1040, Vienna, AustriabFaculty of Chemical Engineering, Hanoi University of Technology, No. 1 Dai Co Viet, Hai Ba Trung, Hanoi, Vietnam

Available online 7 November 2006

Abstract

The 50%CoFe2O4–50%BaTiO3 (in mass) composites with four different building structures, namely: CoFe2O4–BaTiO3 core–shell

structure with CoFe2O4 in core, BaTiO3–CoFe2O4 core–shell structure with BaTiO3 in core, CoFe2O4–BaTiO3 mixed structure, and

BaTiO3–CoFe2O4–BaTiO3 layer structure, have been synthesized and studied. The core–shell structures give higher magnetoelectric

(ME) coefficients compared to the other structures. When using CoFe2O4 as core, the ME coefficient is highest, reaching

3.4mV cm�1Oe�1 for the sample pressed at 6 ton/cm2 and sintered at 1250 1C for 12 h.

r 2006 Elsevier B.V. All rights reserved.

PACS: 75.80.+q; 77.65.�j; 77.84.Lf

Keywords: Magnetoelectric effect; Multiferroic composite; Effect of structure; Barium titanate; Cobalt ferrite

The origin of the magnetoelectric (ME) effect in MEcomposites is the coupling between the magnetostrictiveand piezoelectric phases [1]. So the micro-structure of thecomposites which affects the interactions between the twophases as well as some physical properties such as electricalresistance, dielectric constant may play an important rolein ME composites. In this work, 50%CoFe2O4–50%Ba-TiO3 (in mass) composites with four different buildingstructures, namely: CoFe2O4–BaTiO3 core–shell structurewith CoFe2O4 in core, BaTiO3–CoFe2O4 core–shell struc-ture with BaTiO3 in core, CoFe2O4–BaTiO3 mixedstructure, and BaTiO3–CoFe2O4–BaTiO3 layer structure,have been synthesized and studied.

The core–shell structure samples were prepared by wetchemical method as described elsewhere [2]. In general, thecore was prepared by co-precipitation (CoFe2O4, averagegrain size of about 10 nm) or sol–gel (BaTiO3, averagegrain size of about 42 nm) technique, then introduced to

- see front matter r 2006 Elsevier B.V. All rights reserved.

/j.jmmm.2006.10.338

onding author. Institute of Solid State Physics, Vienna Univer-

hnology, Wiedner Hauptstrasse 8-10, 1040, Vienna, Austria.

58801 13152; fax: +431 58801 13899.

ddresses: giap@ifp.tuwien.ac.at (G.V. Duong),

ifp.tuwien.ac.at (R. Sato Turtelli).

the homogeneous solution containing chelating agent andelements that form the shell. This solution evaporated andgelated on the surface of the core during heating andstirring. The obtained products were pre-sintered at 450 1Cfor 3 h, pressed into pellets under a pressure of 3–7.5ton/cm2, sintered at 1000–1250 1C for 1–20 h, then cooleddown naturally to room temperature (RT).The mixed structure was prepared by simply mixing the

two initial powders: CoFe2O4 (average rain size of 10 nm)and BaTiO3 (average grain size of 40 nm), then pressed intopellets under a pressure similar to those for the core–shellstructure samples. The layer structure was prepared bycasting the powder into the matrix, slightly pressed beforecasting the other layers, and then the whole powders in thematrix were pressed under a pressure of 6 ton/cm2. Theheat treatment of the mixed and layer structure is similar tothat of the core–shell structure samples. After heattreatment, all samples were poled under an electric fieldof 7500V/cm and painted by silver paste for electricalcontacts.X-ray diffraction characterization showed that all

composites consisted of two single phases only: CoFe2O4

and BaTiO3. Magnetic studies showed that the magneticproperties of the CoFe2O4 component are similar to those

ARTICLE IN PRESSG.V. Duong et al. / Journal of Magnetism and Magnetic Materials 310 (2007) e361–e363e362

of the bulk sample: saturation magnetization (Ms) of72 emu/g and coercivity (Hc) of 460Oe at RT. Forreference, a CoFe2O4 bulk sample prepared by citrate gelmethod at RT has Ms of 78 emu/g, Hc of 825Oe andmagnetostriction l of �130 and 70 ppm for parallel andperpendicular measurements, respectively. Very similarvalues are obtained for the other samples.

The ME effect was measured using a lock-in technique asdescribed in Ref. [3]. Fig. 1 shows the ME coefficient asfunction of the DC bias field (aE–HDC) of the sample withCoFe2O4 in core using an AC field of 10Oe, 270Hz. Thissample is formed by pressing the powder under a pressureof 6 ton/cm2, and then sintered at 1250 1C for 12 h. It is

Fig. 1. The aE–HDC curves of the CoFe2O4–BaTiO3 composite with

CoFe2O4 in core at RT (sample preparation: pressed at 6 ton/cm2 and

sintered at 1250 1C for 12 h).

Fig. 2. Effect of structure on aE at RT. Sample index: (1) CoFe2O4 as core

and BaTiO3 as shell; (2) BaTiO3 as core and CoFe2O4 as shell; (3) mixed

structure; (4) BaTiO3–CoFe2O4–BaTiO3 layer structure. Preparation of

samples: pressed at 6 ton/cm2 and sintered at 1250 1C for 16 h.

clear that the curves experience a maximum and showremanence as well as hysteresis behavior. The maximumME coefficient is 3.4mV cm�1Oe�1 for longitudinalmeasurement and 2.0mV cm�1Oe�1 for transverse case.The effect of structure on ME coefficient is shown in

Fig. 2. It was found that the CoFe2O4–BaTiO3 core–shellstructure with CoFe2O4 in core has the highest MEcoefficient: 1.62mV cm�1Oe�1 for sample annealed at1200 1C for 16 h, which is of about 4.2, 8.1 and 11 timeshigher than that of the BaTiO3–CoFe2O4 core–shellstructure with BaTiO3 in core, CoFe2O4–BaTiO3 mixedstructure, BaTiO3–CoFe2O4–BaTiO3 layer structure, pre-pared under the same conditions, respectively. The higherME coefficient of the core–shell structure may beattributed to the better coupling between the two phasesdue to its larger interface area. Additionally, the highermeasured ME coefficients when CoFe2O4 was used as corecompared to the case of BaTiO3 in core may be understoodalso as the discharging effect in the former was lesser. The

Fig. 3. Effect of synthesis temperature (a) and duration (b) on the ME

coefficient of core–shell structure composites.

ARTICLE IN PRESSG.V. Duong et al. / Journal of Magnetism and Magnetic Materials 310 (2007) e361–e363 e363

reason is: the electrical resistance of BaTiO3 is higher thanthat of CoFe2O4 which is confirmed by electrical resistancemeasurements on shape normalized samples at RT:330MO for sample with CoFe2O4 in core, 50MO forsample with BaTiO3 in core, 80MO for mixed structure,42000MO for layer structure.

It is also worth to remind that, beside the physicalproperties such as electric resistance, the microstructure ofthe sample also affects seriously on the ME coefficient.When changing the sintering temperature and durationfrom 1250 1C and 12 h to 1200 1C and 16 h, the longitudinalaE of the sample with CoFe2O4 in core decreases from 3.4to 1.62mV cm�1Oe�1 as can be seen in Figs. 1 and 2. Thereason is the changes in sample microstructure which isnow under investigation. For different structures, theoptimum preparation conditions are different, e.g., forthe core–shell with CoFe2O4 in core, pressed under 6 ton/cm2 and sintered at 1250 1C for 4 h suited best (maxaE ¼ 3.88mV cm�1Oe�1), but when BaTiO3 used as core,the optimum is to press at 3 ton/cm2 and sintered at1150 1C for 8 h (max aE ¼ 0.63mV cm�1Oe�1) as shown inFig. 3. These values are in the range of those reported for

particulate composites: 0.43mV cm�1Oe�1 in 15%Ni0.8-Cu0.2Fe2O4+85%Ba0.9Pb0.1Ti0.9Zr0.1O3 (in mole) [4],3.0–5.58mVcm�1Oe�1 in BaO–TiO–FeO–CoO [5] and0.19mV cm�1Oe�1 in 50%CoFe2O4–50%BaTiO3 (inmass) mixed composite [6].This work is supported by the FWF Proj. Nr.

P16500–N02, Proj. Nr. P15737 and the Austrian ExchangeService (OAD).

References

[1] J. Van Suchetelene, Philips Res. Rep. 27 (1972) 28.

[2] G.V. Duong, R. Groessinger, R. Sato Turtelli, Magnetoelectric

properties of CoFe2O4–BaTiO3 core–shell structure composite, IEEE

Trans. Magn. 42 (2006) 3611.

[3] G.V. Duong, R. Groessinger, M. Schoenhart, D. Bueno-Basques, The

Lock-in technique for Studying Magnetoelectric Effect, J. Magn.

Magn. Mater., in print.

[4] C.M. Kanamadi, L.B. Pujari, B.K. Chougule, J. Magn. Magn. Mater.

295 (2005) 139.

[5] S. Mazuder, G.S. Bhattacharyya, Ceram. Int. 30 (2004) 389.

[6] R.P. Mahajan, K.K. Patankar, M.B. Kothale, S.C. Chaudhari, V.L.

Mathe, S.A. Patil, Pramana-J. Phys. 58 (2002) 1115.