Journal of Alloys and Compounds 476 (2009) 560–565
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom
Magnetic and dielectric properties of barium titanate-coated barium ferrite Chao Wang a , Xijiang Han a,∗ , Ping Xu a , Xiaohong Wang b , Xueai Li a , Hongtao Zhao a a b
Chemistry Laboratory Center, Department of Chemistry, Harbin Institute of Technology, Harbin 150001, China Beijing Institute of Aeronautical Materials, Beijing 100095, China
a r t i c l e
i n f o
Article history: Received 10 April 2008 Received in revised form 26 August 2008 Accepted 10 September 2008 Available online 8 November 2008 Keywords: Barium ferrite Barium titanate Magnetic properties Dielectric properties
a b s t r a c t Flaky barium ferrite with hexagonal molecular structure was successfully prepared by reverse microemulsion method, and was coated with barium titanate through a coordination–precipitation technique. The prepared composite particles were characterized with X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDAX), vibrating sample magnetometry (VSM) and vector network analyzer. Barium ferrite nanoparticles are proved to be single magnetic domains and the lattice volume is expanded slightly after being covered with spherical barium titanate (average diameter: 43.7 nm). Magnetic parameters of BaTiO3 -coated BaFe12 O19 are lower than those of pure BaFe12 O19 due to nonmagnetic characteristic of BaTiO3 , while the coverage of BaTiO3 thin layer on the surface of BaFe12 O19 particles changes the dielectric properties due to the interfacial effect. Finally, a novel model is established to pre-estimate the permittivity and shows excellent agreement with measured value. © 2008 Elsevier B.V. All rights reserved.
1. Introduction M-type barium ferrite with hexagonal molecular structure (BaFe12 O19 ) is a high performance permanent magnetic material with hard magnetism, due to its fairly large magnetocrystalline anisotropy, high Curie temperature, relatively large magnetization [1–3]. BaFe12 O19 has been extensively studied as advanced recording applications and microwave absorbing materials [4–6]. Further technological applications require barium ferrite particles to be single magnetic domains and narrow particle size distribution [3,7]. Recently BaFe12 O19 were prepared using reverse microemulsion technique , sol–gel method  and coprecipitation method [5,10] and so forth. On the other hand, perovskite type structure barium titanate (BaTiO3 ) is widely investigated concerning on the dielectric properties, phase transitions, grain size and dielectric relaxation. For example, BaTiO3 and carbon black composites exhibited excellent electromagnetic performance , and a signiﬁcant low-frequency dielectric permittivity increase in the BaTiO3 /polymer composites was observed . Tetragonal phase BaTiO3 with an average diameter of 54 nm has been synthesized via a sol–gel route without surfactant . In addition, dielectric relaxation for interfacial layer and depletion layer in hexagonal BaTiO3 single crystal was detected . In order to pre-estimate the electrical permittivity property of a multiphase mixture, Lichtenecker’s mixture formula based on the effective medium theory
∗ Corresponding author. Tel.: +86 451 86413702; fax: +86 451 86418750. E-mail address: [email protected]
(X. Han). 0925-8388/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2008.09.092
was deduced . This theory gives a theoretical conﬁrmation for this mixing formula, which has hitherto been considered empirical, and broadly reported in recent studies [16–17]. It has been reported that combining functional nanocomposites may be an effective method to enhance their properties [3,18]. To our best knowledge, there are few articles concerning on the magnetic and dielectric properties of BaTiO3 -coated BaFe12 O19 composites, which is the factor that motivates us to investigate the synthesis and properties of ferroelectric–ferromagnetic BaTiO3 -coated BaFe12 O19 composites. In this paper, BaTiO3 -coated BaFe12 O19 composites are prepared via the reverse microemulsion technique and chemical coprecipitation method. Chemical coprecipitation at 60 ◦ C in preparing BaTiO3 -coated BaFe12 O19 can avoid calcinations, which may inﬂuence the component and structure of BaFe12 O19 . The crystal structure, morphology, magnetic and dielectric properties of the composites were measured and compared with those of uncoated BaFe12 O19 particles. Based on the Lichtenecker’s mixture formula, a novel model is established to pre-estimate the dielectric property of BaTiO3 -coated BaFe12 O19 , which shows well agreement between the calculated value and the measured value. 2. Experimental 2.1. Preparation of barium ferrite particles BaFe12 O19 (sample A) were prepared with n-hexanol as the cosurfactant, cetyltrimethylammonium chloride (CTAC) as the surfactant, and cyclohexane as the solvent. R = V(aqueous): V(oil) = 1:6. In the typical experiment, microemulsion I consisted of 36 g of CTAC, 90 mL of n-hexanol, 360 mL of cyclohexane, 30 mL of 0.2 mol/L Ba(NO3 )2 , 31.5 mL of 2 mol/L Fe(NO3 )3 according to a Fe/Ba ratio of 10.5, and 12 mL
C. Wang et al. / Journal of Alloys and Compounds 476 (2009) 560–565
of water to reach the ratio of aqueous phase to oil phase. With the same value of R, microemulsion II contained 12 g of CTAC, 54 mL of n-hexanol, 216 mL of cyclohexane, 22.5 mL of 25% NH3 ·H2 O, and 22.5 mL of 1 mol/L (NH4 )2 CO3 serving as the precipitator. Microemulsion II was added dropwise to microemulsion I in a water bath at 40 ◦ C under mechanical stirring for 2 h. After the solution was aged for 12 h, the precursor particles were washed by centrifuging with anhydrous ethanol and water to remove the remaining surfactants and organic residuals. The prepared nanoparticles were dried at 60 ◦ C in a vacuum dryer for 12 h, followed by thorough grinding. Precursors were precalcined at 400 ◦ C for 4 h, calcined at 950 ◦ C for 6 h, and furnace cooling to room temperature, and another thorough grinding. We did not take pH into consideration because the content of NH3 ·H2 O added was designed to be excessive and pH value could not measured exactly for the existence of much organic solvent. 2.2. Preparation of barium titanate particles BaTiO3 (sample B) powders were prepared by the coprecipitation technique at low temperature. The Ba/Ti molar ratio chosen was 1:1. 0.001 mol Ba(OH)2 ·H2 O was dissolved in 100 mL deionized water with mechanical stirring vigorously at 60 ◦ C for 30 min in water bath. And then 100 mL anhydrous ethanol solution of 0.001 mol Ti(OC4 H9 )4 was added dropwise, and the reaction was left for 3 h. The resulting precipitate was washed with deionized water, anhydrous ethanol several times and dried in vacuum at 60 ◦ C for 24 h. 2.3. Preparation of barium titanate-coated barium ferrite composite particles The composites BaTiO3 -coated BaFe12 O19 were prepared with BaFe12 O19 (mass):BaTiO3 (mass) = 1:0.05 (sample C), 1:0.1 (sample D), respectively. Firstly 1 g BaFe12 O19 with 20 ml anhydrous ethanol were sonicated for 30 min in an ultrasonic cleaning bath so as to dispersing nanoscale BaFe12 O19 . Then corresponding Ba(OH)2 ·H2 O and other materials were added and the rest steps were the same as described in Section 2.2. 2.4. Characterization The obtained samples were analyzed for their composition, lattice parameter, microstructure, magnetic property, complex relative permittivity by using multiple methods and apparatus. The phase composition of the samples was determined by X-ray powder diffraction, and the XRD dates were collected using a Rigaku D/MAX-RC X-ray diffractometer, with a Cu K␣ radiation source ( = 1.5405 Å, 45.0 kV, 50.0 mA). The 2 Bragg angles were scanned over a range of 10–80◦ . Scanning electron microscope (SEM) measurements were carried out on a FEI SEM SIRION (Holland) to study the morphology of the samples, and the samples were mounted on aluminum studs using adhesive graphite tape and sputter-coated with gold before analysis. The chemical component of BaTiO3 -coated BaFe12 O19 was identiﬁed by energy dispersive X-ray analysis (EDAX) (EDAX-Inc). The magnetization curves of the samples were characterized by a vibrating sample magnetometer (VSM, Lake Shore 7410). The permittivity was determined using a HP-5783E vector network analyzer in the frequency range of 2–18 GHz by using coaxial reﬂection/transmission technique. For this, the samples were prepared with 70.0 mass% composite powders loading in parafﬁn wax as binder. The powder–wax composites were pressed to form cylindrical toroidal shaped specimens with 3 mm inner diameter, 7 mm outer diameter and 2 mm thickness. The test samples of toroidal shape were tightly inserted into the standard coaxial line, the measured values of reﬂected and transmitted scattering parameters were used to determine ε , ε .
3. Results and discussion 3.1. Crystal structures To identify the crystal structure, XRD analysis was performed on the BaFe12 O19 , BaTiO3 and BaTiO3 -coated BaFe12 O19 composite particles, as shown in Fig. 1. The peaks of sample A in Fig. 1 can be indexed with the standard pattern for M-type hexagonal BaFe12 O19 crystals (JCPDS # 27-1029). The XRD pattern of sample B shows the peaks of BaTiO3 with cubic structure (JCPDS# 31-0174), and average crystallite size of BaTiO3 can be calculated from three strong peaks corresponding to the (1 0 0), (1 1 0), (1 1 1) planes using Scherrer Eq. (1): d=
k ˇ cos
where d is the primary particle size, k is the Scherrer constant (0.89), is the wavelength of Cu K˛ radiation (1.5405 Å), is the diffraction angle and ˇ is the FWHM of the peak. The average size of
Fig. 1. X-ray diffraction patterns for samples A (BaFe12 O19 ); B (BaTiO3 ); C (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.05); D (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.1).
the sample calculated is 43.7 nm. It is worth noting that the main diffraction peaks of samples C, D indicate the presence of BaTiO3 and BaFe12 O19 . The peaks at 2 = 31.26◦ , 50.22◦ are very obviously detectable in the XRD patterns of sample C, because the (1 1 0) and (2 1 0) plane diffraction peaks of BaTiO3 are wondrously intense, indicating the existence of BaTiO3 in BaTiO3 -coated BaFe12 O19 composites. A further evidence to conﬁrm the presence of BaTiO3 is that the intensity of the two peaks in sample D are clearly enhanced compared with sample C due to the increase in the content of BaTiO3 . However, the remaining peaks of BaTiO3 are not observable since the BaTiO3 content is so low in samples C and D. There are no extra peaks, that is, there are no impurities formation at the interface between BaTiO3 and BaFe12 O19 , indicating no solid state reaction between BaFe12 O19 and BaTiO3 . As we all know BaTiO3 exists as a tetragonal phase in bulk at room temperature and undergoes a phase transformation to cubic phase at approximately 120 ◦ C (Curie temperature). However some researchers have observed the transition from the tetragonal to the cubic phase with a decrease of crystallite size. Uchino et al.  reported that BaTiO3 revealed a drastic decrease of the Curie temperature around a particle size of 120 nm and that the crystal symmetry was cubic even at room temperature. While Cho and Hamada  indicated the presence of cubic BaTiO3 with 20 nm in size at room temperature, and it was assumed that a sufﬁciently small crystallite size might lead thermodynamically to the disappearance of the tetragonal structure. The prepared BaTiO3 , with an average size of 43.7 nm and stably exists as cubic structure at room temperature, shows good agreement with the published works. Planes corresponding to (1 1 0), (1 0 7), (1 1 4), (2 1 7), (2 0 1 1) and (2 2 0) of hexagonal BaFe12 O19 crystals are used to calculate the lattice constants a and c according to the following equation : 1 4 = 3 d2
h2 + k2 + l2 a2
where h, k and l are miller indices, d is interplanar distance. a and c for sample A are 5.8878 Å and 23.1757 Å respectively, which complies well with the published literature data [3,10]. Applying the same crystal planes, lattice constants a and c for BaTiO3 -coated BaFe12 O19 (sample D) particles are calculated to be 5.8890 Å and 23.1911 Å, respectively, indicating increase in the length of both
C. Wang et al. / Journal of Alloys and Compounds 476 (2009) 560–565
Fig. 2. SEM photos of (a) sample A (BaFe12 O19 ); (b) sample B (BaTiO3 ); (c) sample C (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.05); (d) sample D (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.1).
a and c axis for the lattice compared with pure BaFe12 O19 . Lattice volume, V, of hexagonal crystal system can be obtained from the following Eq. (3) with the values of a and c . V changes from 695.7762 Å3 to 696.5224 Å3 which means lattice volume of BaFe12 O19 would be expanded by BaTiO3 coating. √ V=
3 2 a c 2
In order to further conﬁrm the existence of BaTiO3 in BaTiO3 coated BaFe12 O19 , EDAX analysis of sample D was performed, as shown in Fig. 3. Titanium can be clearly observed on the surface of the prepared BaTiO3 -coated BaFe12 O19 powders. Though diffraction peaks of BaTiO3 were not very obviously detected by XRD, it can be conﬁrmed by EDAX, SEM, and XRD that BaTiO3 really exists in the prepared BaTiO3 -coated BaFe12 O19 . 3.3. Magnetic properties
3.2. Morphology Fig. 2 shows the morphology of these particles as synthesized. Particles of sample A are hexagonal platelet crystals with polished surface, and the length ranges from 300 nm to 500 nm with thickness in 20 nm (Fig. 2(a)). While particles of sample B are nanospheres, with an average size of about 60 nm (Fig. 2(b)), slightly larger than the value calculated from Scherrer formula (43.7 nm). This is because BaTiO3 is a kind of low electric conductivity material and electron bombardment will lead to the electric discharge effect at the edge of particles, and thus the size of the particles will be increased from visual sense. Fig. 2(c) and (d) are the morphology of sample C and D, from which we can see the polished BaFe12 O19 surface is coated with spherical BaTiO3 , and the number of BaTiO3 nanospheres increases a lot as a consequence of content’s increase. In most cases, the spherical BaTiO3 particles are uniformly coated on BaFe12 O19 , e.g. the district labelled ‘I’ in Fig. 2(c). However part of BaTiO3 particles are agglomerate resulting some BaFe12 O19 particles are not completely covered, where the district labelled ‘II’ in Fig. 2(c). We can see clearly from the region labelled ‘III’ in Fig. 2(d) that some excess BaTiO3 particles are agglomerate and some BaFe12 O19 particles are not coated at all. So there is no necessity to enlarge the content of BaTiO3 .
BaFe12 O19 is an excellent magnetic material, while BaTiO3 owns no magnetism. The magnetic properties including hysteresis loop,
Fig. 3. EDAX analysis of sample D (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.1).
C. Wang et al. / Journal of Alloys and Compounds 476 (2009) 560–565
for samples C, and D . As the observed magnetic properties of nanoparticles are a combination of many anisotropy mechanisms, BaTiO3 -coated BaFe12 O19 nanoparticles will likely affect the contribution of the surface anisotropy, shape anisotropy and interface anisotropy to the net anisotropy K. 3.4. Relative complex permittivity
Fig. 4. Hysteresis loops of samples A (BaFe12 O19 ); C (mass BaFe12 O19 :BaTiO3 = 1:0.05); D (mass ratio of BaFe12 O19 :BaTiO3 = 1:0.1).
saturation magnetization (Ms), remanent magnetization (Mr) and coercivity (Hc) of composite were measured in this research. The internal area of hysteresis loop represents the capability of magnetic energy storage of magnetic materials. The hysteresis loops of samples A, C, D are shown in Fig. 4, which exhibit the characteristics of hard magnetic materials. The Ms of composites A, C, D is 63.999 emu/g, 57.823 emu/g, and 56.059 emu/g, respectively, which is decreased with the increase of BaTiO3 . The variation may be resulted from the following reason. BaFe12 O19 is the main body to supply magnetism while BaTiO3 is a kind of nonmagnetic material, so the increase of BaTiO3 brings on the decrease of Ms. The Ms of pure BaFe12 O19 (63.999 emu/g) is much larger than that of the sample with a size of 5 m (49.85 emu/g)  and a little larger than the value of the sample prepared by microemulsion with the size smaller than 100 nm (61.2 emu/g) . The theoretical saturation magnetization value calculated for single crystals of barium ferrite is 72 emu/g and magnetic property of nanoparticles was always lower than those of bulky materials , however, BaFe12 O19 particles with growth to 300–500 nm in this study own higher Ms than the 5 m ones. It means magnetism of small powders sometimes greater than the largers, this result is the same as reported elsewhere . The values of Mr/Ms for samples A, C, D are 0.5183, 0.5001, 0.5078, very close to 0.5, indicating that BaFe12 O19 powders of single magnetic domain are produced , which explains the single magnetic domain characteristic of BaFe12 O19 is unchanged by BaTiO3 coating. The Hc of samples A, C, D are 4834.1, 4794.8, and 4321.5 Oe, respectively. Compared with the great variation of Ms, the change of Hc is reduced slightly, due to reduction in surface anisotropy upon coating . According to Stoner–Wohlfarth theory , the coercivity Hc of nanoparticles is determined by magnetocrystalline anisotropy constant K and saturation magnetization Ms and universal constant 0 as showed by the following equation: K=
0 Ms Hc 2
With the decrease in Ms and Hc, samples C and D obviously have smaller values of K. The coercivity is considered as a measure of the magnetic ﬁeld strength that is required to achieve changes of magnetization direction of a material. Lowering the anisotropy constant K of a material will lower the activation energy barrier and a lower applied ﬁeld required for spin reversal, that is, a lower coercivity, which is another explanation for the lower coercivity obtained
In relative complex permittivity, the real part εr symbolizes the storage capability of electric energy, and the imaginary part εr demonstrates the loss of dielectric energy . Fig. 5(a) and (b) shows the relative complex permittivity of samples A, B, C, and D with wax in microwave frequency. The complex relative permittivity εr of sample A is low and almost constant at 4.3 through out 2–18 GHz frequency range. And εr increases slightly with the presence of BaTiO3 for sample C, however εr for sample D changes signiﬁcantly. εr for sample B obviously larger compared with A, C, D, and decreases gradually with the increase of frequency behaving frequency dispersion. In the whole frequency range investigated in Fig. 5(b), εr of sample B is much higher than those of samples A, C, D. Since both pure BaFe12 O19 and pure BaTiO3 own approximal resonance peaks at 3.8 GHz, 9.7 GHz, 13.8 GHz respectively, the frequency of resonance peaks of samples C and D are not changed obviously. The signiﬁcant difference between Fig. 5(a) and Fig. 5(b) is that the increase of BaTiO3 results in no increase in εr effectively, for the curves A, C, D nearly overlapping. The tangent of the phase angle ı of the complex relative permittivity is deﬁned in terms of εr and εr presented by Eq. (5), which means the dielectric loss of electromagnetic wave. Fig. 5(c) shows the loss tangent in microwave frequency range. The tendencies of curves both in Fig. 5(b) and Fig. 5(c) are similar, which may be due to the invariance of εr with frequency for each sample, so tan ıe is determined by εr . tan ıe =
BaTiO3 is a kind of dielectric material, the dominant dipolar polarization and the associated relaxation phenomena constitute the loss mechanisms. Composite materials, in which magnetic particles are coated with dielectric nanolayers, introduce additional interfaces and more polarization charges on the surface of the particles, this makes the dielectric relaxation behavior more complex. Additionally, Joule-heating loss will also occur due to ﬁnite conductivity of the composites. Aggregate dielectric losses in the samples can be described as due to the contributions from both the dc/ac conductivity and dipole relaxation . Since the BaTiO3 -coated BaFe12 O19 composites are a heterogeneous system, and interfacial polarization is an important polarization process and associated relaxation also will give rise to loss mechanism. As to tan ıe of sample C is a little higher than D (curve C is on the top of curve D in Fig. 5(c)), this is owing to excess BaTiO3 grains are agglomerate (e.g. region ‘III’ in Fig. 2(d)) and decrease the interfaces between BaFe12 O19 and BaTiO3 . We can draw a conclusion that higher values of εr and tan ıe for the sample C may be due to signiﬁcant contribution of interfacial polarization. It has conﬁrmed that the properties of interface could have a dominant role in determining dielectric performance . In order to calculate the permittivity property of a multiphase mixture, effective medium theory  was used. The Lichtenecker’s mixing rule describes the dielectric property of composites as in following equation: ln εeff = fX ln εX + fY ln εY
where fX and fY are the X phase and Y phase volume fractions respectively, and εX , εY , εeff are the dielectric constants of the X phase, Y phase, and total [16,17]. In fact, there are three phases
C. Wang et al. / Journal of Alloys and Compounds 476 (2009) 560–565
Fig. 5. Real part (a) and imaginary part (b) of permittivity and tan ıe (c) of samples A–D; comparison between calculated value and measured value (d).
including BaTiO3 , BaFe12 O19 , and wax, in the electromagnetic measurement of samples C, and D. In order to utilize both the Lichtenecker’s mixing rule and permittivity of samples C, and D in Fig. 5(a) and (b), here a new model is established: BaFe12 O19 and corresponding wax (mass ratio (%) = 70:30) are denoted as X phase, while BaTiO3 and corresponding wax (mass ratio (%) = 70:30) are denoted as Y phase. Sample D is taken as an example to compare the calculated value from Eq. (6) and measured value in Fig. 5(d). The calculated value is lower than the measured, which may be caused by the interfacial effect between BaTiO3 and BaFe12 O19 particles. The curves of calculated and measured εr are nearly overlapping, and the relative errors of both εr and εr are less than 10%, which means this model shows good agreement with the experimental results and will be promising in pre-estimating permittivity of composites in practice. 4. Conclusion In this work, BaFe12 O19 and BaTiO3 composite particles were successfully prepared by reverse microemulsion technique and chemical coprecipitation method, respectively. BaTiO3 was determined in BaTiO3 -coated BaFe12 O19 by XRD, SEM, and EDAX. Saturation magnetization of BaFe12 O19 is greater compared with some bulky and ultraﬁne ones published before and its lattice volume is expanded a little after coated with BaTiO3 . Ultraﬁne BaTiO3 crystal with an average size of 43.7 nm can stably exist as cubic structure at room temperature. Both BaFe12 O19 and BaTiO3 coated BaFe12 O19 show single magnetic domain characteristics
while magnetism of the latter displays slightly decrease because of the presence of nonmagnetic BaTiO3 . The real part of the complex relative permittivity of BaTiO3 -coated BaFe12 O19 is found to increase a lot compared with pure BaFe12 O19 , however the enhancement of imaginary part is not obviously and interfacial effect plays an important role in the change of imaginary part. In the end, a new model is established to utilize Lichtenecker’s mixing rule to pre-estimate permittivity and shows excellent agreement with measured value. Acknowledgements This research was supported by the National Natural Science Foundation of China (Nos. 20676024 and 20776032). References  Z. Zheng, B. Guo, X. Mei, J. Magn. Magn. Mater. 78 (1989) 73.  X. Tang, K.A. Hu, Mater. Sci. Eng. B 139 (2007) 119.  P. Xu, X.J. Han, X.H. Wang, C. Wang, H.T. Zhao, W.J. Zhang, Mater. Chem. Phys. 108 (2007) 196.  T.J. Lewis, J. Phys. D: Appl. Phys. 38 (2005) 202.  M.M. Rashad, M. Radwan, M.M. Hessien, J. Alloys Compd. 453 (2008) 304.  H.F. Yu, K.C. Huang, J. Magn. Magn. Mater. 260 (2003) 455.  C.R. Vestal, Z.J. Zhang, J. Am. Chem. Soc. 125 (2003) 9828.  P. Xu, X.J. Han, M.J. Wang, J. Phys. Chem. C 111 (2007) 5866.  J.X. Qiu, L.J. Lan, H. Zhang, M.Y. Gu, J. Alloys Compd. 453 (2008) 261.  K.K. Mallick, P. Shepherd, R.J. Green, J. Eur. Ceram. Soc. 27 (2007) 2045.  X.D. Chen, G.Q. Wang, Y.P. Duan, S.H. Liu, J. Alloys Compd. 453 (2008) 433.  Z.M. Dang, H.P. Xu, H.Y. Wang, Appl. Phys. Lett. 90 (2007) 012901.  R. Kavian, A. Saidi, J. Alloys Compd. 468 (2009) 528.
C. Wang et al. / Journal of Alloys and Compounds 476 (2009) 560–565  J.D. Yu, T. Ishikawa, Y. Arai, S. Yoda, M. Itoh, Y. Saita, Appl. Phys. Lett. 87 (2005) 252904.  T. Zakri, J.P. Laurent, M. Vauclin, J. Phys. D: Appl. Phys. 31 (1998) 1589.  K.C. Cheng, C.M. Lin, S.F. Wang, S.T. Lin, C.F. Yang, Mater. Lett. 61 (2007) 757.  A. Pashkin, S. Kamba, M. Berta, J. Petzelt, J. Phys. D: Appl. Phys. 38 (2005) 741.  P. Xu, X.J. Han, J.J. Jiang, X.H. Wang, X.D. Li, A.H. Wen, J. Phys. Chem. C 111 (2007) 12603.
   
K. Uchino, E. Sadanaga, T. Hirose, J. Am. Ceram. Soc. 72 (1989) 1555. W.S. Cho, E. Hamada, J. Alloys Compd. 266 (1998) 118. V. Pillai, P. Kumar, M.J. Hou, D.O. Shah, Adv. Colloid Interface Sci. 55 (1995) 241. G. Benito, M.P. Morales, J. Requena, V. Raposoa, M. Vazqueza, J.S. Moya, J. Magn. Magn. Mater. 234 (2001) 65.  S.G. Kim, W.N. Wang, T. Iwaki, A. Yabuki, K. Okuyama, J. Phys. Chem. C 111 (2007) 10175.