Crystal structure, magnetic and dielectric properties of

Journal of Alloys and Compounds 762 (2018) 668e677

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom

Crystal structure, magnetic and dielectric properties of (1-x) BiFe0.80Ti0.20O3 e (x)Co0.5Ni0.5Fe2O4 multiferroic composites Rabichandra Pandey, Lagen Kumar Pradhan, Sunil Kumar, Manoranjan Kar* Department of Physics, Indian Institute of Technology Patna, Bihta, Patna 801103, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 March 2018 Received in revised form 16 May 2018 Accepted 18 May 2018 Available online 21 May 2018

(1-x)BiFe0.80Ti0.20O3e(x)Co0.5Ni0.5Fe2O4 multiferroic composites (x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0) was synthesized by the tartaric acid modified sol-gel method. The Rietveld refinements of all XRD patterns have been carried out to obtain the phase percentages of perovskite BiFe0.80Ti0.20O3 and spinel Co0.5Ni0.5Fe2O4. A change in the lattice parameters of the BFTO and CNFO phase in the composites have been observed, which could be due to strain at the interface of BFTO and CNFO. A change in the peak positions in the Raman spectra of BFTO and CNFO in the composite has been observed and it is well correlated with the XRD results. A significant increase in the magnetization has been observed due to incorporation of Co0.5Ni0.5Fe2O4 phase in the composite. Vegard's law was used to calculate the theoretical value of magnetization for all the composites and compared with the experimental value. Excess value of magnetization observed in the experimental data of composites as compared to the theoretically calculated value and, explained on the basis of interfacial strain effect. An enhancement in the dielectric properties has been observed in the composite system. A correlation between the structure, magnetic and dielectric properties has been reported by assuming strain at the interface of BFTO and CNFO phase. © 2018 Elsevier B.V. All rights reserved.

Keywords: Composite Lattice strain Multiferroic Sol-gel Crystal structure

1. Introduction In recent years, multiferroic materials have attracted most of the researchers in the field of solid state physics due to its wide range of technological applications such as; electric field controlled magnetic memories, transducer, ultrafast optoelectronic devices, spintronics and magnetic field sensors etc. [1,2]. Particularly, coupling between magnetic and electric orders i.e. control of magnetization by varying the electric field or vice versa, defined as magnetoelectric coupling (ME) effect, allows designing new multiferroic devices [3]. However, there are a very few natural multiferroic materials exist which exhibit two or three ferroic orders (i.e. ferroelectric, ferromagnetic, ferroelastic etc.) in a single crystal phase. Moreover, most of them exhibit their ferroic transition below room temperature. Bismuth ferrite (BiFeO3) is a special kind of multiferroic material which exhibits antiferro-paramagnetic (TN z 643 K) and ferro-paraelectric transitions (TC z 1103 K) above room temperature [4,5]. But, high leakage current, weak magnetoelectric responses, low resistivity, high dielectric loss, low

* Corresponding author. E-mail address: [email protected] (M. Kar). https://doi.org/10.1016/j.jallcom.2018.05.198 0925-8388/© 2018 Elsevier B.V. All rights reserved.

remnant polarization etc. limits the potential application of multiferroic BiFeO3 [1,6]. Substitution of different elements (such as rare earth elements i.e. La, Y, Ce, Sm etc., transition metal elements i.e. Ti, Mn, Cr, Nd etc. and alkaline earth elements i.e. Ca, Ba, Sr, Mg etc.) in Bi and/or Fe site of BiFeO3(BFO) have shown an improvement in the electrical and magnetic properties [7e18]. BFO show slow resistivity and high leakage current due to the reduction of Fe3þ to Fe2þ and formation of oxygen vacancies. Particularly, a significant change in the leakage current and electrical resistivity have been observed by the substitution of higher valence ions such as Nb5þ,V5þ,Zr4þ,Ti4þ etc. in place of Fe3þ in BFO [19e22]. However, from technological application point of view, the magnetic parameters as well as ME coefficients are still quite low to be used. An alternative method to fulfill the above requirements can be the suitable mechanical coupling between ferroelectric and piezomagnetic materials and this can be achieved by fabricating composites of BFO with a suitable piezomagnetic material. Cubic spinel ferrite having the general formula MFe2O4, (where M is any divalent ion of metals such as cobalt, nickel, zinc, magnesium, copper etc.) are potential piezomagnetic materials for various technological applications (i.e. ranging from microwave to radio wave frequencies, biomedical, electronic, magnetic recording, transformer core etc.) [19,20]. By considering these facts, several multiferroic

R. Pandey et al. / Journal of Alloys and Compounds 762 (2018) 668e677

composites of BFO with different piezomagnetic materials have been examined(for example; BiFeO3-CoFe2O4 [23,24], BiFeO3-CoxNi1xFe2O4 [25], BiFeO3-CrFe2O4 [26], BiFe0.5Cr0.5O3-NiFe2O4 [27], Bi0.8Dy0.2FeO3eNi0.5Zn0.5Fe2O4 [28],Bi0.85La0.15FeO3-CoFe2O4 [29] etc.) and observed a large enhancement in magnetization as well as enhanced resistivity, dielectric constant and remnant polarization. In particular, Babu et al. have demonstrated that, BCFO-NFO composite exhibits large saturation magnetization and dielectric constant as compared to individual compounds. They have shown improved ME response and the value is twice as large as that of BCFO, and explained it by considering the interfacial strain effect on BCFO [27]. S. C. Mazumdar et al. have reported an enhanced ME voltage response of ~66 mV/cm Oe for the BDFO-NZFO composite with 40% NZFO at 4.7 kOe and attributed to the enhanced mechanical coupling between the two phases [28]. From the literature survey it can be observed that mechanical coupling at the interface of perovskite and spinel phase plays a crucial role for the enhancement of the electrical and magnetic properties as well as the ME coefficient of the composites. There is internal magnetoelectric coupling between the ferroelectricity and antiferromagnetism in BFO as well as exchange interaction between ferrimagnetic CFO and antiferromagnetic BFO will affect the magnetic properties of composite. Lattice mismatch at the interfaces may produce interfacial stress and will affect the various properties. Hence a systematic study of the effect of interfacial strain on the crystal structure, magnetic and electrical properties can explore the possible use of this composite. Among the various elements substitutions in BFO, Ti substitution has drawn considerable attention, because Ti substitution at Fe site in BFO not only reduces the leakage current but also improves the electrical resistivity and magnetic properties in BFO [30]. The conversion of Fe3þ into Fe2þ is compensated by the Ti4þ substitution and it results in reduction of the oxygen vacancies, which subsequently reduces the leakage current. The Ti4þ substitution in BFO fills the oxygen vacancies and/or reduces the formation of Fe2þ obeying the charge compensation mechanism. An enhanced ferroelectricity has been observed by Wang et al. in Ti substituted BFO [31]. Moreover, it has been reported that the magnetization increases with the increase in Ti substitution upto x ¼ 0.20 and then decreases with further increase in Ti substitution [32]. Hence, Ti substitution with x ¼ 0.20 in BFO reveals a great potential for developing a high performance multiferroic materials. Among various piezomagnetic materials, cobalt ferrite (CoFe2O4) is an important member of ferrite family and known as a hard ferrimagnetic material with high coercivity, moderate saturation magnetization, high magnetic anisotropy as well as chemical stability and mechanical hardness [33]. On the other hand, nickel ferrite (NiFe2O4) is a soft ferrimagnetic ferrite material having low coercivity and saturation magnetization but, high electrical resistivity compared to CFO. Hence, NiFe2O4 (NFO) is most suitable material for high frequency technological applications [34]. Further it has been observed that substitution of Ni2þ in place of Co2þ in CFO affects the structural, electrical as well as magnetic properties because Ni2þ ions affect the magnetic properties with its low magnetic moment as compared to Co2þ and reduces the saturation magnetization but simultaneously Ni2þ ions modify the electrical resistivity and dielectric properties of the material [35]. These modification helps the use of Ni substituted CFO for a wide range of applications. Substitution of Ni2þ ion with x ¼ 0.5 in CFO (i.e. Co0.5Ni0.5Fe2O4) have shown the optimized magnetic properties as well as electrical resistivity and dielectric properties [35], and can be considered to be a suitable piezomagnetic material. Hence a composite of BFTO (ferroelectric phase) and CNFO (piezomagnetic phase) has been prepared. A correlation between the structure and magnetic properties has been studied. As per literature survey by

669

using Scopus and other open search engine, no report is available on this particular composite. 2. Experimental section (1-x) BiFe0.8Ti0.2O3 e (x) Co0.5Ni0.5Fe2O4 (i.e. (1-x) BFTO-(x) CNFO) composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 were synthesized by the tartaric acid modified sol-gel method. The detail of the sol-gel synthesis method has been discussed in the earlier report [36]. In the first step, BiFe0.8Ti0.2O3 (BFTO) and Co0.5Ni0.5Fe2O4 (CNFO) compounds were synthesized separately. BFTO compound was synthesized using bismuth nitrate pentahydrate (98% purity, Alfa Aesar), iron nitrate nonahydrate (98% purity, Alfa Aesar), titanium dioxide (99% purity, Merck) and tartaric acid (99.9% purity, Merck) as chemical reagents. Stoichiometric amount of the chemicals were dissolved in de-ionized water (milli Q grade) and concentrated nitric acid to get the aqueous solutions. The final solution was kept on a hot plate at 80 C with a constant stirring of 100 rpm to obtain a viscous and brown colored gel. The gel was kept for overnight in a hot air oven at 100 C to obtain final material. The synthesized BFTO powder was annealed in air environment at 700 C for 3 h. Similar procedure was followed for the synthesis of CNFO. Here, iron nitrate nonahydrate (98% purity, Alfa Aesar), cobalt acetate (99% purity, Alfa Aesar), nickel nitrate (98% purity, Alfa Aesar) and tartaric acid (99.9% purity, Merck) were used as the starting material. After the sol gel process, the final product was annealed at 600 C for 2 h. In the above processes, the tartaric acid was used as a fuel for chemical reaction to takes place and it plays an important role in the formation of the compound. In the second step, BFTO-CNFO composites were prepared by mixing the weight percentage of individually prepared BFTO and CNFO compound in different ratio. The final BFTO, CNFO and BFTO-CNFO composites powders were pressed into cylindrical pellet of 10 mm diameter and 1 mm thickness using a hydraulic press. Finally BFTO compound and BFTO-CNFO composites pellets were sintered at 700 C for 4 h and CNFO pellet was sintered at 600 C for 3 h. For electrical properties measurement, highly conducting silver paste was coated on the flat surfaces of all the pellets and dried in an oven at 150 C for 30 min for proper binding of the silver paste on the surface. The structural analysis of BFTO, CNFO and BFTO-CNFO composites were carried out by the X-ray diffractometer (XRD) (Rigaku TTRX III diffractometer, Japan) with Cu-Ka radiation (l ¼ 1.5418 Å) and a wide range of Bragg's angle 2q (10 2q 110 ) at a scan step of 0.01 and scan rate of 2 /min at room temperature. The Raman spectra at room temperature was measured by the Micro-Raman spectrometer (STR 750 RAMAN Spectrograph, Seki Technotron corporation, Japan) with an excitation source of 514.5 nm wavelength and a 20 microscope. Transmission electron microscopy (TEM JEOL) was used to observe the particle size of the samples. Scanning electron microscopy (SEM) was used to study the microstructural properties of the samples. Vibrating Sample Magnetometer (VSM, Lakeshore Model no. 7410) was used to perform the magnetic measurement with maximum applied field of ±20 kOe at room temperature. The room temperature dielectric constant and tangent losses were measured by the Impedance Analyzer (N4L PSM 1735) with a wide range of frequency (100 Hz 1 MHz). 3. Results and discussions 3.1. Structural analysis XRD patterns of (1-x) BFTO-(x) CNFO composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 at room temperature are shown in Fig. 1. The peak positions of BFTO are indexed to perovskite structure with R3c

670


extracted from the Rietveld refinement analysis and enlisted in Table 1. Some of the R-factors are above 15 (maximum accepted limit for the Rietveld analysis). It could be due to nanocrystalline nature of the sample where signal to noise (background) ratio is very high. Similar R-values have been reported in the literature [38]. Also, lattice parameters of BFTO, CNFO and BFTO-CNFO composites are enlisted in Table 1. There are no significant changes in the lattice parameters of the ferroelectric BFTO and ferromagnetic CNFO have been observed in the composite. However, a small change in the lattice parameter of BFTO and CNFO phase in the composites as compared to their parent phase may be due to interfacial strain exerted on each other by the two phases. The lattice parameters of BFTO in the composites are found to decrease with increase in the CNFO percentage, however lattice parameter of CNFO decreases with the increase in the CNFO percentage. The crystallite size (D) has been determined by using the Scherrer's formula [39]:

D¼

kl bcosq

(1)

In equation (1), constant k depends upon the shape of the crystallite size (¼0.89, assuming the spherical crystallites), b is the full width at half maximum (FWHM) of intensity versus 2q profile, l is the wavelength of the Cu K a radiation (¼1.5418 Å), q is the Bragg's diffraction angle and D is the crystallite size. Here, the value of FWHM has been taken from the Rietveld refinement of corresponding XRD patterns. According to the Rietveld method, the individual contribution to the reflections can be expressed as [39,40],

FWHM2 ¼ U þ D2ST tan2 q þ V tanðqÞ þ W þ

Fig. 1. XRD patterns of the (1-x) BFTO-(x) CNFO multiferroic composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0.

crystal symmetry whereas, CNFO are indexed to spinel structure with Fd3m crystal symmetry. The XRD peak positions in BFTOCNFO composites are match with the individual phases of BFTO and CNFO. The peaks indicated by ‘T’ and ‘N’ in composites are corresponds to BFTO and CNFO phase respectively. No signatures of other phases are traced from the XRD patterns of composites which prevails the possible major chemical reaction between the BFTO and CNFO phases. Also, it is observed an increase of CNFO XRD peak intensity with the increase in CNFO percentage in the composite. The Rietveld refinement of the XRD patterns of BFTO, CNFO and BFTO-CNFO composites were performed to determine the crystallographic parameters with the help of Fullprof software [37]. Although indexing in the XRD patterns of perovskite BFTO has been done with rhombohedra crystal symmetry (R3c space group), the Rietveld refinement is not well converged with only R3c crystal symmetry. Hence different combinations of structural models (R3cþPbam, R3cþP4mm, R3cþPbnm etc.) have been carried out. Finally, R3cþPbnm combination could fit best among all the choices. The CNFO diffraction pattern was refined by Fd3m space group. Hence, all the composites of BFTO-CNFO were refined by using the combination i.e. R3cþPbnmþFd3m space groups. The goodness of fit parameters such as; profile factor (Rp), weighted profile factor (Rwp), expected R factor (Rexp), R-Bragg factor (RBragg) and goodness of Fit (c2 ) determine the quality of refinement. The Rietveld refinement patterns for (1-x)BFTO-(x)CNFO composite with x ¼ 0.3 and x ¼ 0.7 are shown in Fig. 2(a) and (b). All the structural parameters such as; phase percentages, Rp, Rexp, Rwp, RBragg, c2 etc. are

IG cos2 q

(2)

where, U, W and V are the shape parameters and IG is the isotropic size effect and DST is the co-efficient related to strain. The crystallite size of BFTO and CNFO phases are found to be 31 nm and 27 nm respectively. The crystallite sizes of BFTO and CNFO in the composites were calculated individually by considering their corresponding XRD peaks and values are enlisted in Table 1. It is concluded that there is no significant change in the crystallite size in the composites due to the addition of CNFO phases. A small change in lattice parameter in the BFTO phase signifies that, there may be strain exist at the interface between the CNFO and BFTO phases. Hence to support the structural changes in the composites, the Raman analysis has been carried out, which is discussed in the next section.

3.2. Raman analysis Raman spectroscopy is one of the advanced techniques for material characterization. Any change in the crystal structure as well as strain in the material leads to the variation in frequency, bandwidth and intensity of the Raman peaks. It is very much sensitive to the atomic displacement and used as an important tool to investigate the structural phase transitions as well as strain analysis. Hence in order to support the XRD results, strain analysis of the BFTO-CNFO composites has been carried out by the Raman spectroscopy. Fig. 3 shows the measured Raman spectra of BFTO, CNFO and BFTO-CNFO composites at room temperature. Theoretical group analysis predicts that, there are 18 optical phonon modes (4A1þ5A2þ9E) exist for BFO (space group R3c) at room temperature. Out of these, A1 and E modes are both Raman and IR active and, A2 modes are Raman and IR inactive [41]. On the otherhand, there are 39 optical phonon modes has been predicted for spinel phase with space group Fd3m (O7h ). Out of these 39 modes, five


671

Fig. 2. The Rietveld refinement of XRD patterns of (1-x) BiFe0.80Ti0.20O3- (x) Co0.5Ni0.5Fe2O4 for x ¼ 0.3 and 0.7. In the pattern the circle points represent the experimental data and solid line represents the theoretically analyzed data. The vertical lines show the allowed peaks and bottom line represents the difference between the experimental and theoretical data points.

Table 1 Structural parameters of (1-x) BFTO-(x) CNFO multiferroic composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 annealed at 700 C. Errors are given in the bracket. Sample (1-x) BiFe0.8Ti0.2O3 exCo0.5Ni0.5Fe2O4

Phase percentage (%) R3c, Pbnm, Fd3m

Goodness of Fit (GOF) c2

R-Factor Rp, Rwp, Rexp, RBragg

x ¼ 0.0

62.78 37.22 e

3.71

x ¼ 0.3

44.92 27.76 27.32

x ¼ 0.5

Lattice parameters( A)

Average Crystallite size in nm BFTO CNFO

R3c a ¼ b, c in Å

Pbnm a, b, c in Å

Fd3m (a ¼ b ¼ c) in Å

16.5, 21.3, 11.05, 4.88

a ¼ b ¼ 5.584(2) c ¼ 13.877(3)

e

35.4(3)

--

4.01

19.3, 23.5, 11.73, 7.37

a ¼ b ¼ 5.528(5) c ¼ 13.870(3)

8.335(2)

34.5(4)

28.2(3)

32.27 21.56 46.17

3.82

21.5, 26.1, 13.34, 6.93

a ¼ b ¼ 5.469(2) c ¼ 13.867(3)

8.279(4)

33.8(3)

29.1(4)

x ¼ 0.7

19.78 14.29 65.93

3.40

17.8, 24.5, 13.28, 7.49

a ¼ b ¼ 5.546(4) c ¼ 13.860(2)

8.312(3)

34.0(2)

28.7(5)

x ¼ 1.0

–– 100

2.21

14.3, 19.9, 13.39, 6.05

e

a ¼ 5.736(3), b ¼ 5.994(2), c ¼ 7.899(5) a ¼ 5.696(4), b ¼ 5.717(2), c ¼ 8.209(5) a ¼ 5.728(2), b ¼ 5.881(3), c ¼ 8.101(5) a ¼ 5.775(4), b ¼ 5.814(2), c ¼ 7.987(6) e

8.375(5)

–

29.9(3)

modes (3T2gþA1gþEg) are Raman active [42]. Kothari et al. have reported the 13 phonon modes (4A1þ 9E) for polycrystalline BFO, among which the peaks at around 135, 167, 218 and 430 cm1 are belongs to A1 mode and, peaks at 71, 98, 255, 283,

321, 351, 476, 526 and 598 cm1 are belongs to E mode [41]. Similarly, kumar et al. have observed 13 phonon modes for BFO [43]. However, in the present compound (BFTO), 8 Raman peaks at 146.15, 170.48, 216.65, 265.71, 331.36, 365.18, 501.58 and

672


Fig. 3. Raman scattering spectra of BFTO with deconvoluted phonon modes.

756.79 cm1 have been identified(shown in Fig. 3). Detail values of the Raman peaks observed by Kothari et al., Pawan et al. and present work are presented in Table 2 for comparison. From Fig. 3 it can be observed that, with Ti 20% substitution in BFO, the peaks are getting merged and broaden as well as shifted towards higher frequencies as compared to the reported peak frequencies for BFO [41,43]. This is due to decrease in the average atomic mass of Fe site because, the atomic mass of Ti (47.87u) is smaller (85.7%) than that of Fe(55.85u) atom. It is well known that the frequency of the Raman modes are inversely proportional to the reduced mass. Hence, a shift in the phonon mode towards higher frequency has been observed as compared to the reported peak frequencies for BFO. Similarly, from Fig. 4 it can be seen that, five Raman active modes at 188.63, 326.78, 478.82, 636.28 and 699.68 cm1 are for CNFO. These results are consistent with the reported peak frequencies of CNFO [44,45] enlisted in Table 2. The Raman spectra of BFTO-CNFO composites contains signal of both BFTO and CNFO phases (Fig. 5). The main peaks of BFTO and CNFO are denoted as ‘*’ and ‘#’ symbol in the composites respectively. It can be observed from Fig. 5 that, peak positions of BFTO (A1-2 and A1-3 modes) and CNFO (A1g mode) in the composites are shifted towards lower frequencies as compared to the peak positions of parent BFTO and

Fig. 4. Raman scattering spectra of CNFO with deconvoluted phonon modes.

CNFO spectra. Fig. 6(a) and (b) shows the enlarged view of the Raman peaks near 170 and 216 cm1 for BFTO phase and 700 cm1 for CNFO phase for all composites. From Fig. 6(a) it can be observed that, peak frequencies of BFTO near 170 cm1 (A1-2 mode) are shifted by 2.35, 3.67, 3.14 cm1 and near 216 cm1 (A1-3 mode) are shifted by 1.82, 3.28 and 2.18 cm1 towards lower frequency for 0.7BFTO-0.3CNFO, 0.5BFTO-0.5CNFO and 0.3BFTO-0.7CNFO composites respectively (see Table 3). Similarly, peak frequency of A1g mode of CNFO near 700 cm1 are shifted by 1.01, 2.86 and 1.42 cm1 towards lower frequency for 0.7BFTO-0.3CNFO, 0.5BFTO0.5CNFO and 0.3BFTO-0.7CNFO composites respectively. Shifting of the peak position towards higher or lower frequency is a signature of compressive and tensile strain present in the composite [46]. Similar observations and explanations have been reported by other groups [47]. Hence these shifting of peak positions of BFTO-CNFO composites suggest the presence of strain in the composite at the interface of BFTO and CNFO. The shifting of peak positions to lower frequency of BFTO and CNFO spectra in composites is due to the interfacial lattice mismatch of spinel CNFO and perovskite BFTO. Further it can be observed that the shifting is maximum for 0.5BFTO-0.5CNFO composite in all peaks and it can be suggested that, the strain is maximum at this composition. The above

Table 2 List of peak positions of Raman spectra for BFTO and CNFO phase. Raman modes of BiFe0.80Ti0.20O3

Raman modes of Co0.5Ni0.5Fe2O4

Kothari et al. [39] BFO(cm1)

Pawan et al. [40] BFO(cm1)

Present Work BiFe0.80Ti0.20O3 (cm1)

Kumar et al. [41] CNFO (cm1)

Rajnish et al. [42] CFO (cm1)

Present Work Co0.5Ni0.5Fe2O4 (cm1)

71 98 135 167 218 255 283 321 351 430 476 526 598

132.1 143.5 178.4 220.7 264.2 278.9 298.1 342.0 371.6 474.0 525.1 554.1 602.3

146.15 170.48 …… 216.65 265.71 …… …… 331.36 365.18 501.58 …… ……. 756.79

201 313 474 590 690 …… …… …… …… …… …… ……

173 293 492 616 686 …… …… …… …… …… …… ……

188.63 326.78 478.82 636.28 699.68 …… …… …… …… …… …… ……


673

3.3. Microstructural analysis Fig. 7 depicts the TEM micrograph of 0.5BFTO-0.5CNFO composite powder annealed at 700 C. The particles are well dispersed and the average particle sizes are found to be~35 nm. The surface morphology study of composites is carried out by SEM measurement. Fig. 8 shows the SEM micrograph of 0.5BFTO-0.5CNFO composite. From Fig. 8 it can be observed that the samples are dense and particles are spherical having nano grain size distribution which represents the crystalline nature of the compound. ImageJ software has been used to calculate the grain size of the samples and values are found to be ~38 nm. The preparations of the composites are a simple physical mixture of BFTO and CNFO compound in different ratio. Hence there is no significant change in the sizes of the particle and retains their original size. However a small variation in the size has been found, which may be due to the interfacial stain effect (i.e. due to lattice mismatch between BFTO and CNFO phases). The particle size obtained from TEM, SEM and XRD techniques are in good agreement with each other. 3.4. Magnetic properties

Fig. 5. Raman spectra of the (1-x) BFTO-(x) CNFO multiferroic composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0.

discussion reveals that, the Raman spectra results are in well agreement with the XRD patterns analysis.

Magnetic hysteresis loops (M~H loops) of BFTOeCNFO composites at an applied magnetic field of ±20 kOe have been shown in Fig. 9. CNFO shows characteristics of ferrimagnetic M-H hysteresis loop with a saturation magnetization (Ms) of 0.65emu/g at ±20 kOe magnetic field, coercivity (Hc) of 850 Oe and remnant magnetization of15.86emu/g. On the other hand, BFTO exhibits narrow magnetic hysteresis loop with anti-ferromagnetic in nature. It has maximum magnetization of 0.167 emu/g at ±20 kOe applied magnetic field, coercivity (Hc) 2219 Oe and remnant magnetization 0.0287emu/g. M-H hysteresis curves of (1-x) BFTO-(x) CNFO composites exhibit typical ferrimagnetic loop with high saturated magnetization. All experimentally observed values such as; saturation magnetization, coercive field, remnant magnetization are presented in Table 4. The coercivity values in all the composites are

Fig. 6. Enlarged view of the Raman spectra of the (1-x) BFTO-(x) CNFO multiferroic composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 at (a) nearly 170 cm1 and 216 cm1 for BFTO and (b) nearly 700 cm1 for CNFO.

674


Table 3 Shifting of main peaks (A1-2, A1-3 modes) of BFTO and (A1g modes) of CNFO phases in all composites. Raman modes

BFTO

0.7BFTO-0.3CNFO

Shifting in cm1

0.5BFTO-0.5CNFO

Shifting in cm1

0.3BFTO-0.7CNFO

Shifting in cm1

CNFO

A1-2 A1-3 A1g

170.48 216.65 ……

168.13 215.83 698.67

2.35 1.82 1.01

166.81 213.37 696.82

3.67 3.28 2.86

167.34 214.47 698.26

3.14 2.18 1.42

…… …… 699.68

Fig. 9. Magnetic hysteresis loops of the BFTO-CNFO multiferroic composites with different CNFO contents. Inset depicts the M-H loop for BFTO compound.

Fig. 7. TEM micrograph of 0.5BFTO-0.5CNFO composite annealed at 700 C.

composites are shown in Fig. 10. The saturation magnetization of the composite increases with the increase in CNFO percentage in (1-x) BFTO-(x) CNFO composite. Hence the enhanced magnetization in the composite is dominated by CNFO phase. The maximum magnetization in the composites should be equal to the sum of the contribution from individual magnetic moment of BFTO and CNFO phases. Hence, the Vegard's law approximation was employed to calculate the saturation magnetization. According to the Vegard's law [48,49]:

MmaxðcalÞ ¼ MmaxðBFTOÞ ð1 XÞ þ MmaxðCNFOÞ ðXÞ

Fig. 8. SEM micrograph of 0.5BFTO-0.5CNFO composite annealed at 700 C.

higher than that of pure CNFO (HC ¼ 850 Oe). It indicates that the magnetic domains of CNFO phases are constrained by the BFTO phases and unable to switch freely in BFTO-CNFO composites. Hence the coercive fields in the composites are higher than pure CNFO phase. However, it is still lower than the coercive field of BFTO phase (HC ¼ 1917 Oe). It is due to the easy magnetization of composites due to the presence of CNFO magnetic domain as compared to pure BFTO phase. Variation of saturation magnetization, remnant magnetization and coercive field of BFTO-CNFO

(3)

where, X ¼ wt% obtained from the Rietveld refinement (Table 1), MmaxðcalÞ ¼ calculated maximum magnetization of the composites at 20 kOe applied field, MmaxðBFTOÞ and MmaxðCNFOÞ are the observed maximum magnetization of BFTO and CNFO phase at 20 kOe applied field. The theoretically calculated values of saturation magnetization for all composites are enlisted in Table 4. It is observed that, the theoretically calculated values of saturation magnetization are lower compared to that of experimental values for all composites. Similarly the theoretically calculated values of coercivity ðHC Þ and remnant magnetization (Mr ) are lower than that of experimental value. Hence, it can be suggested that, the magnetization in the composite is not only from BFTO and CNFO but, there is some other factor responsible for the excess magnetization in the composites. (i) The excess magnetization may be appears due to presence of dense grain boundaries with a certain texture and structure of amorphous intercrystalline layer [50e52]. From the micrographs (Figs. 7 and 8), it can be seen that, samples are nanograined and contain well developed grain and intergrain boundaries. Hence, the presence of grain boundaries is one of the possible factors for the extra magnetization in the composites. (ii) Secondly, Even though BFTO have weak magnetic moment with


675

Table 4 Magnetic parameters of (1-x) BFTO-(x) CNFO multiferroic composites for x ¼ 0.0, 0.3, 0.5, 0.7and 1.0 annealed at 700 C (MS ¼ Saturation magnetization and HC ¼ Coercive field). Sample (1-x)BiFe0.8Ti0.2O3 exCo0.5Ni0.5Fe2O4

Maximum magnetization at 20 kOe(Ms) emu/g

Remnant Magnetization (Mr) emu/g

Coercivity (Hc) in Oe

Mmax(cal) emu/g

Squareness Ratio (R) ¼ Mr/Ms

X ¼ 0.0 X ¼ 0.3 X ¼ 0.5 X ¼ 0.7 X ¼ 1.0

0.1677 11.9835 18.8635 23.5122 30.6513

0.0287 6.5721 7.8345 12.6235 15.8647

2219 1024 1013 932 850

e 8.4951 14.2419 20.2655 e

0.4692 0.5484 0.4153 0.5368 0.5176

Fig. 10. Variation of saturation magnetization, remnant magnetization and coercive field of (1-x)BFTO-(x) CNFO multiferroic composites for x ¼ 0.00, 0.30, 0.50, 0.70 and 1.0.

canted antiferromagnetic structure, the spin alignments are possibly modified due to strain at the interfaces which is evidenced from the XRD and Raman analysis. The role of interface effect to the enhancement of magnetization can also supported by determining squareness ratio (R). The squareness ratio(R) of BFTO, CFNO and all the composites of BFTOCNFO are calculated by dividing experimentally observed Mr by Ms and listed in Table 4. The importance of squareness ratio is to determine the type of the intergrain exchange interaction. The nonzero value of R, signifies that there is intergrain magnetostatic interaction at the interface of the BFTO and CNFO. Arrott plot (i.e.M2Vs H/M plot) is another method to determine the saturation magnetization and it is basically based on Weiss molecular field theory [53,54]. Hence, to investigate the ferromagnetic nature of the BFTO-CNFO composites, Arrott plot of the virgin curves of all the composites has been carried out and plots are depicted in Fig. 11. The high field data of all the composites have been analyzed by linear function (as shown in the inset of Fig. 11) and extrapolated to the H/M ¼ 0 to obtain the saturation magnetization. The value of saturation magnetizations obtained from the Arrott plots are 0.15, 11.18, 17.88, 22.58 and 30.16 emu/g for x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 respectively for (1-x) BFTO-(x) CNFO composite. These values are nearly equal to the experimentally observed value of saturation magnetization and suggest that, there is strain mediated enhanced magnetization due to lattice mismatch at the interfaces of BFTO and CNFO phases.

Fig. 11. Arrott plot of (1-x) BFTO-(x) CNFO multiferroic composites for x ¼ 0.00, 0.30, 0.50, 0.70 and 1.0. Inset shows the linear fit to the experimental data at high magnetic field for x ¼ 0 sample.

CNFO composite system with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 are depicted in Fig. 12. All samples exhibit dielectric dispersion in the studied frequency range. The dielectric constant decreases with increase in the frequency for all samples. In the low frequency region (100Hz-10 kHz), dielectric constant decreases sharply which can be explained by the Maxwell eWagner interfacial polarization

3.5. Dielectric analysis The variation of dielectric constant (ε0 ) with frequency in the range of 100 Hz to 1 MHz at room temperature for (1-x)BFTO-(x)

Fig. 12. Variation of dielectric constant (ε0 ) with frequency of (1-x) BFTO-(x) CNFO multiferroic composites for x ¼ 0.0, 0.3, 0.5, 0.7and 1.0 at room temperature. Inset shows the curves for dielectric loss (tand) versus frequency.

676


Table 5 Dielectric constant (ε0 ) and Dielectric loss (tand) measured at three different frequency of (1-x) BFTO-(x) CNFO multiferroic composites for x ¼ 0.0, 0.3, 0.5, 0.7and 1.0 at room temperature. Sample Name (1-x) BFTO-(x) CNFO

ε0 at 100 Hz

ε0 at 10 kHz

ε0 at 100 kHz

tand at 100 Hz

tand at 10 kHz

tand at 100 kHz

x ¼ 0.0 x ¼ 0.3 x ¼ 0.5 x ¼ 0.7 x ¼ 1.0

95.03 338.65 644.59 447.33 226.27

17.71 45.96 111.24 51.63 19.91

9.75 12.73 36.01 16.49 7.25

0.07 0.15 0.19 0.18 0.11

0.014 0.032 0.047 0.038 0.025

0.0084 0.0093 0.0106 0.0097 0.0091

[55,56]. Polarization near metal electrode interfaces (i.e. space charge polarization) and grain boundaries are active at low frequency regions along with the electronic, ionic and dipolar polarizations. Hence at low frequency region, dielectric constant is large. However in the high frequency region (10 kHz-1 MHz), the dipoles at the metal electrode interfaces as well as grain boundaries are unable to follow the oscillating electric field. Hence the contribution from space charge polarization is very less. The only active contribution to the dielectric value is from the ionic, dipolar and electronic polarizations. Therefore, the dielectric constant is very low at high frequency region. From Fig. 12, it can be observed that, the dielectric constants of all the composites are higher than the dielectric constant of BFTO and CNFO. It may be due to addition of CNFO into the BFTO matrix which increases the number of intergrain boundaries. Because the ferroelectric BFTO grains are enclosed by ferrite CNFO grains and vice versa, the interfacial polarization will increase at the grain boundary. Hence the interfacial polarization developed at the interfaces of BFTO and CNFO grains in addition to the polarization developed in the BFTO and CNFO grains increases the dielectric constant in the composites. The dielectric constant of 0.5BFTO-0.5CNFO composite is found to be maximum as compared to other composites in all frequency range. This may be due to number of intergrain boundaries are maximum for 0.5BFTO-0.5CNFO composite which leads to the maximum contribution of interface polarization. Inset of Fig. 12 shows the variation of dielectric loss (tan d) with frequency in the range of 100 Hz to 1 MHz at room temperature for (1-x)BFTO-(x)CNFO composite system with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0. The dielectric loss follows the similar trend to the dielectric constant curves. The values of dielectric constant and dielectric loss at 100 Hz, 10 kHz and 100 kHz are enlisted in Table 5. It is worth to note that, the maximum dielectric constant is observed for 0.5BFTO-0.5CNFO which is consistent with the Raman spectra i.e. the maximum Raman peak shift is observed for this composition, which is assumed due to the interfacial strain. The electric dipole polarization is directly depends upon the nature of bond and strain in the lattice. Hence a correlation between strain at the BFTO-CNFO interface in the composite and dielectric constant has been observed. 4. Conclusions (1-x)BFTOe(x) CNFO multiferroic composites with x ¼ 0.0, 0.3, 0.5, 0.7 and 1.0 were synthesized by using the tartaric acid modified sol-gel method. The X-ray diffraction patterns indicate the formation of the composites with perovskite BFTO and the spinel CNFO phase. With increase in the percentage of spinel CNFO phase in the composites, a decrease in the lattice parameters in the perovskite phases has been observed. Strain effect at the interface of BFTO perovskite and CNFO spinel phase may be responsible for this change in lattice parameters and, it has been consistent with the Raman analysis (i.e. shifting of Raman peaks). The Vegard's law could employ to determine the theoretical values of the magnetization and it has been compared with experimentally observed

magnetization. A significant increase in the magnetization has been observed in the BFTO-CNFO composites due to CNFO phase in the composites. The constructive spin alignment due to strain effect at the interface may be responsible for the excess magnetization in the BFTO-CNFO composites. Maximum dielectric constant (644 at 100 Hz)is observed for 0.5BFTO-0.5CNFO composite. It is interesting to point out that, the maximum shift in Raman peak has been observed for the same composition. Hence, the present study concludes a correlation between lattice strain and magnetic coupling as well as dielectric properties of (1-x) BFTOe(x) CNFO multiferroic composite. Acknowledgements The authors are thankful to the Aryabhatta Knowledge University for extending the VSM facilities. References [1] G. Catalan, J.F. Scott, G. Catalan, J.F. Scott, Physics and applications of bismuth ferrite, Adv. Mater. 21 (2009) 2463e2485. [2] W. Eerenstein, M.D. Mathur, J.F. Scott, Multiferroic and magnetoelectric materials, Nature (London) 442 (2006) 759. € hlich, A.V. Goltsev, R.V. Pisarev, Observation [3] M. Fiebig, Th Lottermoser, D. Fro of coupled magnetic and electric domains, Nature London 419 (2002) 818. [4] J. Ma, J. Hu, Z. Li, C. Wen Nan, Recent progress in multiferroic magnetoelectric composites: from bulk to thin films, Adv. Mater. 23 (2011) 1062. [5] Y.J. Chen, et al., Strong magnetic enhancement in self-assembled multiferroicferrimagnetic nanostructures, Nanoscale 5 (2013) 4449. [6] N.A. Hill, Why are there so few magnetic ferroelectrics? J. Phys. Chem. B 104 (2000) 6694e6709. [7] B. Yotburut, P. Thongbai, T. Yamwong, S. Maensiria, Electrical and nonlinear current-voltage characteristics of La-doped BiFeO3 ceramics, Ceram. Int. 43 (2017) 5616e5627. [8] H. Fki, M. Koub, L. Sicard, W. Cheikhrouhou-Koub, A. Cheikhrouhou, S. Ammar-Merah, Influence of Y doping on structural, vibrational, optical and magnetic properties of BiFeO3 ceramics prepared by mechanical activation, Ceram. Int. 43 (2017) 4139e4150. [9] J. Mao, L. Cao, Z. Xiong, X. Wang, J. Zhou, Y. Li, W. Wu, Experimental and first principles investigation of Bi1xCexFeO3: structure, electronic and optical properties, J. Alloys Compd. 721 (2017) 638e645. [10] B. Yotburut, P. Thongbai, T. Yamwong, S. Maensiri, Synthesis and characterization of multiferroic Sm-doped BiFeO3nanopowders and their bulk dielectric properties, J. Magn. Magn. Mater. 437 (2017) 51e61. [11] L. Yao, X. Wu, S. Yang, Y. Zhang, Structural and optical properties of Ca doped BiFeO3 thin films prepared by a sol-gel method, Ceram. Int. 43 (2017) S470eS473. [12] S. Chandel, P. Thakur, M. Tomar, V. Gupta, A. Thakur, Investigation of structural, optical, dielectric and magnetic studies of Mn substituted BiFeO3 multiferroics, Ceram. Int. 43 (2017) 13750e13758. [13] M.S. Bernardo, D.G. Calatayud, T. Jardiel, D. Makovec, M. Peiteado, A.C. Caballero, Titanium doping of BiFeO3 ceramics and identification of minor phases by Raman spectroscopy, J. Raman Spectrosc. 48 (2017) 884e890. [14] O. Singh, A. Agarwal, A. Das, S. Sanghi, A. Jindal, Evolution of structural and magnetic phases in Nd doped BiFeO3multiferroics with sintering time, J. Magn. Magn. Mater. 442 (2017) 200e207. [15] S.S. Arafat, S. Ibrahim, Study of the structural and electrical properties of CrDoped BiFeO3 ceramic, Mater. Sci. Appl. 8 (2017) 716e725. [16] Yi Ting, Chi-Shun Tu, Pin-Yi Chen, Cheng-Sao Chen, J. Anthoniappen, V.H. Schmidt, Jenn-Min Lee, Ting-Shan Chan, Wei-Yu Chen, Rui-Wen Song, Magnetization, phonon, and X-ray edge absorption in barium-doped BiFeO3 ceramics, J. Mater. Sci. 52 (2017) 581e594. [17] A.F. Hegab, I.S.A. Farag, A.M. EL-Shabiny, A.M. Nassaar, A.A. Ramadan, A.M. Moustafa, Role of Sr doping on transport and magnetic properties of bismuth ferrite in correlation with defect formation, J. Mater. Sci. Mater.

R. Pandey et al. / Journal of Alloys and Compounds 762 (2018) 668e677 Electron. 28 (2017) 14460e14470. [18] S. Bi, X. Su, J. Li, G. Hou, Z. Liu, C. Zhong, Z. Hou, High-temperature dielectric properties and microwave absorption abilities of Bi1xMgxFeO3 nanoparticles, Ceram. Int. 43 (2017) 11815e11819. [19] W.S. Chang, C.S. Tu, P.Y. Chen, C.S. Chen, C.Y. Lin, K.C. Feng, Y.L. Hsieh, Y.H. Huang, Effects of Fe3d-O2p and Bi6sp-O2p orbital hybridizations in Nd doped BiFeO3 ceramics, J. Alloys Compd. 710 (2017) 670e679. [20] P. Godara, A. Agarwal, N. Ahlawat, S. Sanghi, K. Kaswan, Effect of doping of vanadium ions on crystal structure, dielectric and magnetic properties of Bi0.8Ba0.2FeO3 multiferroic, J. Magn. Magn. Mater. 406 (2016) 76e82. [21] S. Mukherjee, R. Gupta, A. Garg, V. Bansal, S. Bhargava, Influence of Zr doping on the structure and ferroelectric properties of thin films, J. Appl. Phys. 107 (2010), 123535. [22] G.N. Sharma, S. Dutta, A. Pandey, S.K. Singh, R. Chatterjee, Microstructure and improved electrical properties of Ti-substituted BiFeO3 thin films, Mater. Res. Bull. 95 (2017) 223e228. [23] A. Das, S. De, S. Bandyopadhyay, S. Chatterjee, D. Das, Magnetic, dielectric and magnetoelectric properties of BiFeO3-CoFe2O4 nanocomposites, J. Alloys Compd. 697 (2017) 353e360. [24] M. Rafique, A. Herklotz, K. Dorr, S. Manzoor, Giant room temperature magnetoelectric response in strain controlled nanocomposites, Appl. Phys. Lett. 110 (2017), 202902. [25] S. Ojha, W.C. Nunes, N.M. Aimon, C.A. Ross, Magnetostatic interactions in selfassembled CoxNi1xFe2O4/BiFeO3 multiferroic nanocomposites, ACS Nano 10 (2016) 7657e7664. [26] A. Kumar, K.L. Yadav, H. Singh, R. Pandu, P.R. Reddy, Structural, magnetic and dielectric properties of xCrFe2O4e(1-x)BiFeO3 multiferroic nanocomposites, Physica B 405 (2010) 2362e2366. [27] S. Narendra Babu, Jen-Hwa Hsu, Y.S. Chen, J.G. Lin, Dielectric, magnetic and magnetoelectric properties of multiferroic BiFe0.5Cr0.5O3eNiFe2O4 composites, J. Appl. Phys. 107 (2010), 09D919. [28] S.C. Mazumdar, M.N.I. Khan, Md. Fakhrul Islam, A.K.M. Akther Hossain, Tuning of magnetoelectric coupling in (1-y) Bi0.8Dy0.2FeO3e(y)Ni0.5Zn0.5Fe2O4 multiferroic composites, J. Magn. Magn. Mater. 401 (2016) 443e454. [29] R. Pandey, L.K. Pradhan, M. Kar, Structural, magnetic, and electrical properties of (1-x)Bi0.85La0.15FeO3-(x) CoFe2O4 multiferroic composites, J. Phys. Chem. Solid. 115 (2018) 42e48. [30] M. Kumara, K.L. Yadav, Magnetic field induced phase transition in multiferroic BiFe1xTixO3 ceramics prepared by rapid liquid phase sintering, Appl. Phys. Lett. 91 (2007), 112911. [31] Y. Wang, Ce-Wen Nan, Enhanced ferroelectricity in Ti-doped multiferroic thin films, Appl. Phys. Lett. 89 (2006), 052903. [32] M. Kumar, K.L. Yadav, The effect of Ti substitution on magnetoelectric coupling at room temperature in the BiFe1xTixO3 system, J. Phys. Condens. Matter 18 (2006) L503eL508. [33] A. Franco, F.C. Silva, High temperature magnetic properties of cobalt ferrite nanoparticles, Appl. Phys. Lett. 96 (2010), 172505. [34] S.E. Shirsath, B.G. Toksha, K.M. Jadhav, Structural and magnetic properties of In3þ substituted NiFe2O4, Mater. Chem. Phys. 117 (2009) 163. [35] M. Hashim, Alimuddin, S. Kumar, S.E. Shirsath, R.K. Kotnala, J. Shah, R. Kumar, Synthesis and characterizations of Ni2þ substituted cobalt ferrite nanoparticles, Mater. Chem. Phys. 139 (2013) 364e374. [36] R. Pandey, C. Panda, P. Kumar, M. Kar, Phase diagram of Sm and Mn co-doped Bismuth Ferrite based on crystal structure and magnetic properties, J. Sol. Gel Sci. Technol. 85 (2018) 166e177. [37] J. Rodriguez-Carvajal, Fullprof. 2k, version 4.6 ceMar 2002, Physica B 192

677

(1993) 55. [38] N. Adhlakha, K.L. Yadav, R. Singh, Effect of BaTiO3 addition on structural, multiferroic and magneto-dielectric properties of 0.3CoFe2O40.7BiFeO3 ceramics, Smart Mater. Struct. 23 (2014), 105024 (16pp). [39] B.D. Cullity, Elements of X-ray Diffraction, second ed., Addison-Wesley Series, USA, 1987. [40] R.Y. Young, The Rietveld Method, third ed., Oxford University Press, New York, 1996. [41] D. Kothari, V.R. Reddy, V.G. Sathe, A. Gupta, A. Banerjee, A.M. Awasthi, Raman scattering study of polycrystalline magneto-electric BiFeO3, J. Magn. Magn. Mater. 320 (2008) 548e552. [42] P. Chandramohan, M.P. Srinivasan, S. Velmurugan, S.V. Narasimhan, Cation distribution and particle size effect on Raman spectrum of CoFe2O4, J. Solid State Chem. 184 (2011) 89. [43] P. Kumar, N. Shankhwar, A. Srinivasan, M. Kar, Oxygen octahedra distortion induced structural and magnetic phase transitions in Bi1xCaxFe1xMnxO3 ceramics, J. Appl. Phys. 117 (2015), 194103. [44] R. Kumar, M. Kar, Correlation between lattice strain and magnetic behavior in non-magnetic Ca substituted nano-crystalline cobalt ferrite, Ceram. Int. 42 (2016) 6640e6647. [45] A. Kumar, M.A. Dar, P. Sharma, D. Varshney, Structural and Raman scattering study of Ni-doped CoFe2O4, in: AIP Conference Proceedings, vol. 1591, 2014, p. 1148. [46] N.M. Aimon, D.H. Kim, H. Kyoon, C.A. Ross, Deposition of epitaxial BiFeO3/ CoFe2O4nanocomposites on (001) SrTiO3 by combinatorial pulsed laser deposition, Appl. Phys. Lett. 100 (2012), 092901. [47] M. Tyagi, M. Kumari, R. Chatterjee, P. Sharma, Raman scattering spectra, magnetic and ferroelectric properties of BiFeO3eCoFe2O4nanocomposite thin films structure, Physica B 448 (2014) 128e131. [48] A.R. Denton, N.W. Ashcroft, Vegard's law, Phys. Rev. A 43 (1991) 3161. [49] S. Pillai, D. Bhuwal, A. Banerjee, V. Shelke, Bulk interface engineering for enhanced magnetization in multiferroic BiFeO3 compounds, Appl. Phys. Lett. 102 (2013), 072907. [50] B.B. Straumal, A.A. Mazilkin, S.G. Protasova, A.A. Myatiev, P.B. Straumal, G. Schütz, P.A. van Aken, E. Goering, B. Baretzky, Magnetization study of nanograined pure and Mn-doped ZnO films: formation of a ferromagnetic grain-boundary foam, Phys. Rev. B 79 (2009), 205206. [51] B.B. Straumal, S.G. Protasova, A.A. Mazilkin, E. Goering, G. Schütz, P.B. Straumal, B. Baretzky1, Ferromagnetic behaviour of ZnO: the role of grain boundaries, Beilstein J. Nanotechnol. 7 (2016) 1936e1947. [52] B. Straumal, A. Mazilkin, S. Protasova, A. Myatiev, P. Straumal, E. Goering, B. Baretzky, Influence of texture on the ferromagnetic properties of nanograined ZnO films, Phys. Status Solidi B 248 (7) (2011) 1581e1586. [53] H. Ohta, C. Michioka, A. Matsuo, K. Kindo, K. Yoshimura, Magnetic study of SmCoAsO showing a ferromagnetic-antiferromagnetic transition, Phys. Rev. B 82 (2010), 054421. [54] R. Singh, S.K. Srivastava, A.K. Nigam, V.V. Khovaylo, L.K. Varga, R. Chatterjee, Use of Arrott plots to identify Neel temperature (TN) in metamagnetic Ni48Co6Mn26Al20 polycrystalline ribbons, J. Appl. Phys. 114 (2013), 243911. [55] M. Samet, V. Levchenko, G. Boiteux, G. Seytre, A. Kallel, A. Serghei, Electrode polarization vs. Maxwell-Wagner-Sillars interfacial polarization in dielectric spectra of materials: characteristic frequencies and scaling laws, J. Chem. Phys. 142 (2015) 194703. [56] W.A. Yager, The distribution of relaxation times in typical dielectrics, Physics 7 (1936) 434.