lable at ScienceDirect

Journal of Alloys and Compounds 650 (2015) 363e369

Contents lists avai

Journal of Alloys and Compounds

journal homepage: http: / /www.elsevier .com/locate/ ja lcom

Cation distribution and magnetic property of Ti/Sn-substitutedmanganeseezinc ferrites

Ke Sun a, *, Guohua Wu a, Bo Wang a, Qiuyu Zhong a, Yan Yang b, Zhong Yu a,Chuanjian Wu a, Peiwei Wei a, Xiaona Jiang a, Zhongwen Lan a

a State Key Laboratory of Electronic Thin Films and Integrated Devices, University of Electronic Science and Technology of China, Chengdu 610054, Chinab Department of Communication Engineering, Chengdu Technological University, Chengdu 611730, China

a r t i c l e i n f o

Article history:Received 13 April 2015Received in revised form8 June 2015Accepted 27 July 2015Available online 30 July 2015

Keywords:MnZn ferritesTi/Sn substitutionRietveld refinementCation distributionCore lossesDC-bias superposition

* Corresponding author.E-mail addresses: ksun@uestc.edu.cn, shmily81102

http://dx.doi.org/10.1016/j.jallcom.2015.07.2580925-8388/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

Manganeseezinc ferrites with the composition of Mn0.782�xZn0.128Mx4þFe2þ0.09þ2xFe3þ2�2xO4 (x ¼ 0;

M ¼ Ti, x ¼ 0.004; M ¼ Sn, x ¼ 0.004) have been prepared by a solid-state reaction method. The cationdistribution has been investigated by the Rietveld refinement of X-ray diffraction (XRD) patterns, fracturemicrostructure by scanning electron microscope (SEM), and magnetic property by LCR and BeH analyzer,respectively. The results show that Ti4þ and Sn4þ ions prefer to occupy octahedron sublattices (B sites),which leads to an increase in lattice parameter. Sn4þ ions substituted MnZn ferrite has much denser andmore uniform microstructure with smaller grains than that of Ti4þ ions substituted and unsubstitutedMnZn ferrites. The Curie temperature (Tc) of Ti4þ and Sn4þ substituted MnZn ferrites is higher than thatof unsubstituted sample. In addition, both Ti4þ and Sn4þ substitutions, due to the compensation ofmagnetocrystalline anisotropy constant, can move the secondary peak of mi ~ T curve to low temperature.However, the opposite variation trend has been observed in a PL ~ T curve. Also, the DC-bias super-position characteristics have been discussed.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

Various light-weight equipments generally use high-performance switching power supplies and DCeDC converters.Due to its high initial permeability (mi), high saturation induction(Bs), and low core losses (PL), manganeseezinc (MnZn) ferrite,which is considered to be the heart of power supplies and con-verters, mainly acts as transformer and inductor cores. In applica-tions of MnZn ferrite, such as transformers, the operatingtemperature is usually controlled to work between 80 �C and100 �C, by changing the composition of the MnZnFeO. In MnZnferrites with iron stoichiometry or iron deficiency the magneto-crystalline anisotropy constant (K1) is negative. Theoretically, onecan make the MnZn ferrites work at the expected temperature byintroducing the ferrites or ions with positive K1 into MnZn ferritesto compensate the total magnetocrystalline anisotropy. There aregenerally two ways to control the temperature dependence ofmagnetic property. One is introducing Co2þ ions into the MnZn

8@126.com (K. Sun).

ferrite to compensate the anisotropy because of its positivecontribution [1]. The other is producing Fe2þ ions in MnZn ferrite,because Fe2þ ions also have a positive contribution to compensatethe negative anisotropy of MnZn ferrite [2]. Currently, there aremany reports about introducing Fe2þ ions into MnZn ferrites, suchas iron excess, Ti4þ, Sn4þ, and Nb5þ ions substitutions, etc. [3e6]. Allof these reports investigated the temperature dependence ofmagnetic property, such as initial permeability and core losses.However, in the modern circuit system, more and more electronicdevices must undergo the influence of DC-bias superpositionaccompanied with AC signals [7]. The DC-bias superposition char-acteristics of these devices could significantly influence the effi-ciency of the circuit systems. The ferrite materials can be regardedas the cores of these devices, whose DC-bias superposition char-acteristics are of vital importance. Ferrite materials with high DBser

(DBser ¼ Bs�Br) could favor the attainment of good DC-bias su-perposition characteristics [8e12]. However, there are few reportson the DC-bias superposition of MnZn ferrites. This work willdemonstrate the Ti4þ and Sn4þ ions substitution on the DC-biassuperposition characteristics as well as the cation distributionand the temperature dependence of magnetic property. In thiscurrent work, one carried out the Rietveld refinement to identify

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369364

the cation distribution of MnZn ferrites and verified Ti4þ and Sn4þ

ions occupying octahedron sublattices (B sites). The Curie temper-ature (Tc) of Ti4þ and Sn4þ substituted MnZn ferrites is higher thanthat of unsubstituted sample, which is ascribed to the enhance-ment of superexchange interaction between the tetrahedron (Asite) and octahedron (B site) sublattices. Furthermore, the DC-biassuperposition characteristics and the temperature dependence ofmagnetic property have been discussed.

2. Experiment procedures

2.1. Sample preparation

Manganeseezinc ferrites with the composition ofMn0.782�xZn0.128Mx

4þFe2þ0.09þ2xFe3þ2�2xO4 (x ¼ 0; M ¼ Ti, x ¼0.004; M ¼ Sn, x ¼ 0.004) were prepared by a solid-state reactionmethod. The analytical grade raw materials of Fe2O3, Mn3O4, ZnO,TiO2 and SnO2 powders were weighed in stoichiometric proportionand mixed with deionized water in planetary mill for 1 h. Afterbeing dried, the slurries were calcined at 890 �C in air for 2 h. Then,the resulting powder, added with CaCO3 (0.015 wt%), V2O5 (0.02 wt%), Co2O3 (0.09 wt%), and NiO (0.02 wt%) additives, were milled indeionized water for 2 h. The ball-milling media were zirconia ballswith super-hardness and the rotation velocity was 241 rpm. Afterbeing dried, the powders were granulated with 10% poly-vinylalcohol (PVA). Then, it was pressed into toroidal shapes with thedimensions of outer diameter in 25 mm, inner diameter in 15 mm,and height in 7 mm. In the end, the green flans were sintered at1360 �C for 3 h with 4% oxygen atmosphere and cooled at equi-librium conditions in N2/O2 atmosphere. The atmosphere was

Fig. 1. The Rietveld analysis of X-ray diffraction patterns for Mn0.782�xZn0.128M4þxFe2þ0.0

controlled by Morineau and Paulus [13] equation for equilibriumoxygen partial pressure.

2.2. Sample characterization

The phase structure was checked by X-ray diffraction (XRD) andcross-section microstructure by scanning electron microscopy(SEM). The Rietveld refinement of structural parameters was doneusing TOPAS software. From enlarged SEMmicrographs of samples,average grain sizes (D), by applying the average value of 5 micro-graphs to each sample, were estimated by intercept method. Theinductancewasmeasured using a LCRmeter (TH2828) and then theinitial permeability (mi) was calculated. The core losses (PL) at100 kHz and 200 mT were measured by a BeH analyzer (IWATSU,SY-8232). The saturated hysteresis loops were measured at 1 kHzand 1200 A/m following the standards of IEC62044-1, IEC62044-2and IEC62044-3. The DC-bias superposition was carried out usingbias supply (Agilent, 42841A). The Curie temperature (Tc) wasmeasured by a TA Q-100 thermogravimetric analyzer. The densitywas measured by the Archimedean method.

3. Results and discussion

3.1. Structural and microstructural property

Fig. 1 presents the Rietveld refinement of X-ray diffraction pat-terns for Mn0.782�xZn0.128M4þ

xFe2þ0.09þ2xFe3þ2�2xO4 (x ¼ 0; M ¼ Ti,x ¼ 0.004; M ¼ Sn, x ¼ 0.004) ferrites. The cross line denotesexperimental data, and the solid line demonstrates calculated in-tensities. The bottom line represents the difference between

9þ2xFe3þ2�2xO4 ferrites: (a) x ¼ 0, (b) M ¼ Ti, x ¼ 0.004, and (c) M ¼ Sn, x ¼ 0.004.

Table 1The diffraction intensity ratio, experiment lattice parameter (a0), theoretical lattice parameter (at), profile parameters (Rp, Rexp and c2), and X-ray density (dx) ofMn0.782�xZn0.128M4þ

xFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

Mn0.782�xZn0.128M4þx

Fe2þ0.09þ2xFe3þ2�2xO4

Intensity ratio a0 (Å) at (Å) Rp(%) Rexp(%) c2 dx (g/cm3)

I400/I422 I220/I400

Exp. Cal. Exp. Cal.

x ¼ 0 2.018 1.965 1.748 1.747 8.4132 8.5452 1.44 1.37 1.24 5.13M ¼ Ti, x ¼ 0.004 1.985 1.947 1.710 1.725 8.4437 8.5461 1.25 1.26 1.01 5.11M ¼ Sn, x ¼ 0.004 1.580 1.634 1.898 1.836 8.4952 8.5464 1.32 1.34 1.12 5.01

Table 2Cations distribution, bond length, bond angle, density (d), porosity (P) and grain size (D) of Mn0.782�xZn0.128M4þ

xFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

Mn0.782�xZn0.128M4þx

Fe2þ0.09þ2xFe3þ2�2xO4

A sublattice B sublattice FeA-O-FeB d(g/cm3) P(%) D (mm)

Bond length (Å) Bond angle (�)

x ¼ 0 Mn2þ0.626Zn2þ

0.128 Fe3þ0.246 Mn2þ0.156Fe3þ1.754 Fe2þ0.090 1.8402 þ 2.1247 125.21 4.88 4.96 14.8 ± 3.0

M ¼ Ti, x ¼ 0.004 Mn2þ0.622Zn2þ

0.128 Fe3þ0.250 Mn2þ0.156Fe3þ1.742 Fe2þ0.098Ti4þ0.004 1.8394 þ 2.1240 125.26 4.85 5.15 15.5 ± 3.8

M ¼ Sn, x ¼ 0.004 Mn2þ0.622Zn2þ

0.128 Fe3þ0.250 Mn2þ0.156Fe3þ1.742 Fe2þ0.098Sn4þ

0.004 1.8394 þ 2.1240 125.26 4.89 2.44 14.4 ± 3.1

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369 365

measured and calculated intensities, and allowed Braggs positionsare marked as vertical lines. It can be detected that there is onlysingle phase of spinel structure despite Ti4þ or Sn4þ ions substi-tution. Ti4þ and Sn4þ ions have incorporated into the MnZn ferrites.All the observed peaks are allowed Bragg 2q positions. The intensityratios (I400/I422 and I220/I400) are sensitive to the cation distri-bution [14]. It is well-acknowledged that Fe3þ and Mn2þ ions candistribute over A and B sublattices, while Zn2þ and Fe2þ ions preferto occupy A and B sublattices, respectively [15]. Ti4þ and Sn4þ havea strong tendency to enter into B sublattices [4]. Based on themention above, the intensity ratios (I400/I422 and I220/I400) for allthe samples were calculated for various cations distribution. Thecalculated values match well with the experimental data, which ispresented in Table 1. Correspondingly, the cation distribution islisted in Table 2. One can also observe that Mn2þ and Fe3þ ionsoccupy A and B sublattices, where the ratio of Mn2þ ions in A and Bsublattices is four to one. Similar results were found in Ni-substituted Mn-ferrites [16]. However, Zn2þ and M4þ (Ti4þ andSn4þ) ions prefer to occupy the tetrahedron sublattice (A site) andoctahedron sublattice (B site), respectively. This preferable octa-hedron site distribution has also been found in Ti4þ substitutedMnZn ferrite [4] and Ti4þ substituted CoFe2O4 ferrite [17]. For thesamples with Ti4þ and Sn4þ substitutions, Ti4þ and Sn4þ ions enterinto B sublattice and induce corresponding Fe2þ ions because of thecharge balance [4]. Especially, the Fe3þ amount in A sublattice

Fig. 2. SEM micrographs of Mn0.782�xZn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4 fe

increases for Ti4þ and Sn4þ substituted samples as shown in Table 2.A closer consideration of cation distribution and lattice

parameter (a) obtained from the Rietveld refinement forMn0.782�xZn0.128M4þ

xFe2þ0.09þ2xFe3þ2�2xO4 samples are listed inTable 1. The low values of profile parameters (Rp and Rexp) areindicative of that the refinements of samples are effective andreliable. The lattice parameter (a0) and theoretical lattice parameter(at) [15] of MnZn ferrite without substitution are 8.4132 Å and8.5452 Å, respectively. However, one can observe a slight increasein lattice parameter (a0) for Ti4þ and Sn4þ substituted samples. Thisphenomenon is attributed to the distortion of crystal lattice [4].

Fig. 2 shows the cross-section microstructure of Mn0.782�x

Zn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites. Correspondingly, the

average grain size (D), density (d), and porosity (P) are presented inTable 2. The unsubstituted (See Fig. 2a) and Sn4þ substituted MnZnferrites (See Fig. 2c) have much uniform and dense microstructurethan that of Ti4þ substituted sample (See Fig. 2b). It is obvious thatSn4þ substituted sample has the smallest grain size (14.4 ± 3.1 mm).The improvement in homogeneity and densification of Sn4þ

substituted samples can be partially attributed to the liquid phaseformation of SnO2 during the sintering process [5,18]. It is knownthat the melting point of SnO2 is 1127 �C, which is lower than thesintering temperature of the sample. So SnO2 can be used as a fluxand form liquid phase in the sintering process, and make the grainrefined. The grains grow continuously and more completely in the

rrites: (a) x ¼ 0, (b) M ¼ Ti, x ¼ 0.004, and (c) M ¼ Sn, x ¼ 0.004.

Fig. 3. Hysteresis loops of Mn0.778Zn0.128Sn4þ0.004Fe2þ0.098Fe3þ1.992O4 ferrites measured at different temperatures: (a) T ¼ 25 �C, (b) T ¼ 100 �C.

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369366

liquid phase sintering environment, the grain sizes decrease, andthe uniformity and compactness of the grain growth can beimproved.

3.2. Temperature dependence of magnetic property

Fig. 3 demonstrates the hysteresis loops of Mn0.778Zn0.128Sn4þ

0.004Fe2þ0.098Fe3þ1.992O4 ferrites measured at different tem-peratures. It is observed that Sn4þ substituted sample shows typicalsoft magnetic characteristics. The higher the temperature is, thelower the saturation induction (Bs) becomes. Correspondingly, theBs ~ T curves of Mn0.782�xZn0.128M4þ

xFe2þ0.09þ2xFe3þ2�2xO4 ferritesare shown in Fig. 4. It is found that the Bs of the samples with Ti4þ

and Sn4þ substitution is smaller than that of unsubstituted sample.This phenomenon is ascribed to the difference of the moleculemagnetic moment induced by the cation distribution. For precisecation distribution as shown in Table 2, one can easily deduce thatthe molecule magnetic moment of unsubstituted MnZn ferrite islarger than that of either Ti4þ substituted or Sn4þ substitutedsample. This is due to the non-magnetic ions (Ti4þ and Sn4þ)occupying B sites, which reduces the net magnetic moment offerrite. Also, one can see that the Bs of Sn4þ substituted sample ismuch larger than that of Ti4þ substituted sample, which resultsfrom the difference of density [19]. Simultaneously, the Bs of all thesamples reduces with the increasing temperature followed theBrillouin's temperature characteristics [20]. The decreasing trend of

Fig. 4. Saturation induction (Bs) versus temperature curves of Mn0.782�xZn0.128

M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

the Bs ~ T curve for all the samples shows a little difference. This isattributed to the various Curie temperature (Tc) and the tempera-ture dependence of Brillouin function [20]. The Tc for unsubstitutedsample is 265 �C, which is lower than that of Ti4þ substituted(270 �C) and of Sn4þ substituted sample (269 �C). However, thedifference in Curie temperature is related to the magnetic ion dis-tribution, the bond length and bond angle of magnetic ions [21]. Asshown in Table 2, Ti4þ and Sn4þ substituted samples increase themagnetic ion-pairs and reduce the bond length of FeA-O-FeB, whichwill produce strong superexchange interaction and hence obtainhigh Curie temperature. It is the difference in Curie temperaturethat induces much sensitive temperature relationship of Bs due tothe transfer of magnetic order [21].

Fig. 5 shows the temperature dependence of initial permeability(mi) of all the samples. At room temperature, the mi of Sn4þ

substituted sample is higher than that of unsubstituted and Ti4þ

substituted samples. Meanwhile, both Ti4þ substitution and Sn4þ

substitution can move the temperature of the secondary maximumpeak in the mi ~ T curves to lower temperature. As presented inTable 2, the tetravalent Ti4þ and Sn4þ ions prefer to occupy theoctahedral sublattice (B site) in the MnZn ferrite. They can generateFe2þ at B site after substitute Fe3þ ions, 2Fe3þ/Ti4þþFe2þ and2Fe3þ/Sn4þþFe2þ [19,22]. The Fe2þ ion has a positive contributionto the magnetocrystalline anisotropy constant (K1). So Ti4þ andSn4þ substitution can reduce the K1 and make it tend to zero at alower temperature [22], whichmoves the position of the secondary

Fig. 5. Initial permeability versus temperature curves of Mn0.782�xZn0.128

M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

Fig. 6. Temperature dependence of PL of Mn0.782�xZn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4

ferrites.Fig. 8. The DBser versus superposition direct current (DC) of Mn0.782�x

Zn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369 367

maximum peak in the mi ~ T curves to lower temperature. However,the difference of temperature of the secondary maximum peak inthe mi ~ T curve is ascribed to the various microstructure such asgrain size, pores, and density [23].

Fig. 6 shows the temperature dependence of core losses (PL)measured at 100 kHz and 200mT. It is observed that the core lossesPL of all the samples reduce at first and then increase with theincreasing temperature. At room temperature, the PL of unsub-stituted and Ti4þ substituted MnZn ferrites is much higher thanthat of Sn4þ substituted sample. In addition, the temperature of theminimum in the PL ~ T curve moves to lower temperature for Ti4þ

and Sn4þ substituted samples.Generally, the core losses (PL) of MnZn ferrite are composed of

three loss factors, hysteresis loss (Ph), eddy current loss (Pe), andresidual loss (Pr) [24,25], namely

PL ¼ Ph þ Pe þ Pr (1)

At the low frequency ranges, Pr can be neglected and PL can beexpressed as

PLzPh þ Pe (2)

Fig. 7 shows the temperature dependence of Ph and Pe of theferrites. The Ph is considered to be caused by the irreversibledomain wall movement, which is inversely proportional to thecubic of the initial permeability mi, namely, Phf1=m3i [19]. Hence the

Fig. 7. Temperature dependence of hysteresis loss (Ph) and eddy curre

valley point temperature of Ph shifted to lower temperature for theTi4þ and Sn4þ substituted samples, which is consistent with thevariation tendency of the secondary maximum peak of the initialpermeability. Furthermore, the eddy current loss (Pe) of all thesamples is relatively low and changes slightly at low temperature,but increases steeply at high temperature. This is because that theiron excess MnZn ferrites are typical N-type semiconductors, inwhich the resistivity decreases with the increasing temperature,and Pe is inversely proportional to the resistivity [26]. Simulta-neously, both Ti4þ and Sn4þ substituted samples show higher eddycurrent loss at high temperature than unsubstituted sample. This isascribed to the rapid decrease in the electrical resistivity. At hightemperature, the thermal energy damages the bonding energy ofTi4þeFe2þ and Sn4þeFe2þ [27]. Hence the electrical resistivity ofsubstituted samples decreases much quicker than that of unsub-stituted sample. It results in an increasing eddy current loss at hightemperature.

3.3. DC-bias superposition characteristics

Fig. 8 shows the variation of DBser of the Mn0.782�x

Zn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites with the increasing DC-

bias superposition (Idc). The DBser increases rapidly at first, andreaches the maximum value when the Idc is close to 1 A, then de-creases slightly when the Idc increases further. It is found that thesample without ions substitution has the largest DBser. When the

nt loss (Pe) of Mn0.782�xZn0.128M4þxFe2þ0.09þ2xFe3þ2�2xO4 ferrites.

Fig. 9. DC-bias superposition characteristics of the incremental permeability (m△).

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369368

Idc is less than 1 A, the DBser of Sn4þ substituted sample is smallerthan that of Ti4þ substituted sample. This should be attributed tothe domain wall pinning originating from the diversity in micro-structure (See Fig. 2 and Table 2). However, the variation curves ofDBser -Idc change slightly when Idc is greater than 1 A. This phe-nomenon is ascribed to the domain rotation mechanism [21].

Fig. 9a shows DC-bias superposition characteristics of the in-cremental permeability (m△) and Fig. 9b presents the percentage ofpermeability variation. Where m1 is the permeability when the Idc iszero, and m2 is the permeability with DC-bias superposition. Withthe increase of Idc, the permeability increases at first, quickly rea-ches the maximum and then drops to a small value. It is becausethat at the low DC-bias field, the magnetization mechanism ismainly caused by the domain wall displacement. Based on themechanism of domain wall movement, the reversible permeabilityof the material goes up with the increase of Idc. When the Idc in-creases to a certain value, the permeability will achieve themaximum. This variation should be attributed to the irreversibledomain wall movement [21]. However, as the Idc increases further,the magnetic core gradually tends to saturate, the permeabilitydecreases continuously, and the permeability finally changes veryslow. In addition, the sample without ions substitution shows goodDC-bias superposition characteristic. This is consistent with thevariation tendency of DBser. That is to say, the sample with highDBser has a better DC-bias superposition characteristic [28].

4. Conclusions

This paper investigates the cation distribution derived from theRietveld refinement and magnetic property of Ti4þ and Sn4þ

substituted MnZn ferrites and the following results have been ob-tained. Ti4þ and Sn4þ ions have incorporated into the MnZn ferritesand occupied B sites, which induces an increase in the latticeparameter. Sn4þ substituted sample has much denser and moreuniform microstructure with smaller grains than that of Ti4þ

substituted and unsubstituted MnZn ferrites. Ti4þ and Sn4þ

substituted MnZn ferrites have higher Curie temperature than thatof unsubstituted sample. Unsubstituted sample shows the highestBs and △Bs-r, and it presents a good DC-bias superposition char-acteristic. Ti4þ and Sn4þ substituted samples can move the tem-perature of the secondary maximum peak in the mi ~ T curves to lowtemperature, which shows the opposite variation trend in PL ~ Tcurve.

Acknowledgments

The authors are grateful for the financial support from the Na-tional Natural Science Foundation of China under Grant 51472045.

References

[1] L. Li, Z. Lan, Z. Yu, K. Sun, Z. Xu, Effects of Co-substitution on wide temperatureranging characteristic of electromagnetic properties in MnZn ferrites, J. AlloysCompd. 476 (2009) 755e759.

[2] L. Li, Z. Lan, Z. Yu, K. Sun, Influence of Fe2O3 stoichiometry on initial perme-ability and temperature dependence of core loss in MnZn ferrites, IEEE Trans.Magn. 44 (2008) 13e16.

[3] H. Ji, Z. Lan, Z. Yu, Z. Xu, Microstructure and temperature dependence ofmagnetic properties of MnZn ferrites for power applications, IEEE Trans.Magn. 46 (2010) 974e978.

[4] H. Ji, Z. Lan, Z. Xu, H. Zhang, G.J. Salamo, Magnetic and M€ossbauer studies ofMn0.679�xZn0.256TixFe2.066O4 spinel ferrites: effect of cation distribution, IEEETrans. Magn. 49 (2013) 4277e4280.

[5] Z. Xu, Z. Yu, K. Sun, L. Li, H. Ji, Z. Lan, Microstructure and magnetic propertiesof Sn-substituted MnZn ferrites, J. Magn. Magn. Mater. 321 (2009)2883e2889.

[6] A.D.P. Rao, B. Ramesh, P.R.M. Rao, S.B. Raju, Thermaoelectric power studies ofSn/Nb substituted MneZn ferrites, J. Mater. Sci. 34 (1999) 621e623.

[7] G. Ott, J. Wrba, R. Lucke, Recent developments of MneZn ferrites for highpermeability applications, J. Magn. Magn. Mater. 254e255 (2003) 535e537.

[8] H. Su, H. Zhang, X. Tang, B. Liu, Z. Zhong, Effects of Co-substitution on DC-bias-superposition characteristic of the NiCuZn ferrites, Phys. B 405 (2010)4006e4009.

[9] T.T. Ahmed, I.Z. Rahman, M. Rahman, Study on the properties of the coppersubstituted NiZn ferrites, J. Mater. Process. Technol. 153 (2004) 797e803.

[10] Kh. Gheisari, S. Javadpour, H. Shokrollahi, B. Hashemi, Magnetic losses of thesoft magnetic composites consisting of iron and NieZn ferrite, J. Magn. Magn.Mater. 320 (2008) 1544e1548.

[11] X. Tang, H. Zhang, H. Su, Z. Zhong, F. Bai, Influence of microstructure on theDC-bias-superposition characteristic of NiZn ferrites, IEEE Trans. Magn. 47(2011) 4332e4335.

[12] S. Yan, L. Dong, Z. Chen, X. Wang, Z. Feng, The effect of the microstructure onthe DC-bias superposition characteristic of NiCuZn ferrite, J. Magn. Magn.Mater. 353 (2014) 47e50.

[13] R. Morineau, M. Paulus, Chart of PO2 versus temperature and oxidation degreefor MneZn ferrites in the composition range: 50＜Fe2O3＜54; 20＜MnO＜35;11＜ZnO＜30 (mole %), IEEE Trans. Magn. 11 (1975) 1312e1314.

[14] A.A. Pandit, A.R. Shitre, D.R. Shengule, K.M. Jadhav, Magnetic and dielectricproperties of Mg1þxMnx, Fe2�2x, O4 ferrite system, J. Mater. Sci. 40 (2) (2005)423e428.

[15] R.L. Dhiman, S.P. Taneja, V.R. Reddy, Structural and M€ossbauer spectral studiesof nanosized aluminum doped manganese zinc ferrites, Adv. Condens. Matter.Phys. (2008) 1e7, http://dx.doi.org/10.1155/2008/839536.

[16] M. Siddique, N.M. Butt, M. Shafi, T. Abbas, M. ul. Islam, Cation distribution inNi-substituted Mn-ferrites by M€ossbauer technique, J. Radioanal. Nucl. Chem.258 (2003) 525e529.

[17] K.S. Rao, A.M. Kumar, M.C. Varma, G.S.V.R.K. Choudary, K.H. Rao, Cation dis-tribution of titanium substituted cobalt ferrites, J. Alloys Compd. 488 (2009)16e19.

[18] K. Sun, Z. Lan, D. Chen, L. Li, Y. Chen, Z. Yu, Effect of titanium and tin additives

K. Sun et al. / Journal of Alloys and Compounds 650 (2015) 363e369 369

on microstructure and high frequency properties of manganeseezinc ferrite,J. Chin. Ceram. Soc. 34 (2006) 818e822.

[19] Y. Huang, S. Li, Z. Lan, Magnetic Materials, first ed., House of Electronics In-dustry, Beijing, 1994, pp. 17e18.

[20] R. Guo, Z. Yu, Y. Yang, L. Chen, K. Sun, X. Jiang, C. Wu, Z. Lan, Relationshipbetween Curie temperature and Brillouin function characteristics of NiCuZnferrites, J. Appl. Phys. 117 (2015), 073905e73911-073905-6.

[21] D. Wan, X. Ma, Magnetic Physics, first ed., House of Electronics Industry,Beijing, 1999, pp. 138e196.

[22] V. Zaspalis, V. Tsakaloudi, E. Papazoglou, M. Kolenbrander, R. Guenther, P. vander Valk, Development of a new MnZn-ferrite soft magnetic material for hightemperature power applications, J. Electroceram. 13 (2004) 585e591.

[23] S. Chen, S. Chang, I. Lin, The influence of grain boundary internal stress onpermeability: temperature curve for MneZn ferrites, J. Magn. Magn. Mater.

209 (2000) 193e196.[24] O. Inoue, N. Matsutani, K. Kugimiya, Low loss MnZn-ferrites frequency

dependence of minimum power loss temperature, IEEE Trans. Magn. 29(1993) 3532e3534.

[25] K. Sun, Z. Lan, Z. Yu, Z. Xu, X. Jiang, Z. Wang, Z. Liu, M. Luo, Temperature andfrequency characteristics of low-loss MnZn ferrite in a wide temperaturerange, J. Appl. Phys. 109 (2011), 106103e106111-106103-3.

[26] K. Sun, Z. Lan, Z. Yu, L. Li, X. Nie, Z. Xu, Analysis of losses in NiO doped MnZnferrites, J. Alloys Compd. 466 (2009) 315e320.

[27] G.C. Jain, B.J. Das, S. Kumari, Effect of doping a MneZn ferrite with GeO2 andSnO2, IEEE Trans. Magn. 16 (1980) 1428e1433.

[28] T. Tsutaoka, T. Kasagi, K. Hatakeyama, Magnetic field effect on the complexpermeability for a MneZn ferrite and its composite materials, J. Eur. Ceram.Soc. 19 (1999) 1531e1535.