PDF Room temperature multiferroic and magnetodielectric properties in ... - NJU

I. Introduction

Multiferroic materials that simultaneously exhibit ferroelec- tric, ferromagnetic, and/or ferroelastic orderings have attracted much attention due to their interesting fundamental physics and their potentials for new type of device applications [1–6].

Among the reported multiferroic materials, BiFeO₃ (BFO) is the most well-known room temperature magnetoelectric material because of its high Curie temperature (~1103 K) and Neel temperature (~643 K) [5]. The ferroelectricity is orig- inated from the Bi³⁺ ions with its lone electron pairs (6s²) displaced away from the centrosymmetric position relative to the adjacent oxygen ions. The magnetism results from the par- tially filled d orbital of Fe³⁺ ion (d⁵) [2]. Currently, large spon- taneous polarization of ~100 μC cm⁻² has been reported in films and single crystals [2, 6]. However, its ferromagnetism

in macroscopic size is weak due to its G-type canted antifer- romagnetic (AFM) along with a long-range cycloidal spatially modulated spiral spin structure [7]. Thus, the coupling effect between the ferroelectric and ferromagnetic orders is very weak, which limits its applications in the field of multiferroic devices.

It is worth noting that the excellent ferroelectric properties are rarely reported in BFO bulk ceramics owing to the large leakage current from many factors, e.g. second phases, charge defects and nonstoichiometry [8, 9]. In addition, the weak magnetism in BFO is also needed to be improved to satisfy the requirements for applications in multiferroic devices. To improve the ferroelectric and ferromagnetic properties simul- taneously, many studies have endeavored to synthesize high quality BFO ceramics [9–21]. Substitution of A- and/or B-sites by other elements in BFO has been extensively adopted with

Journal of Physics D: Applied Physics

Room temperature multiferroic and

magnetodielectric properties in Sm and Sc co-doped BiFeO ₃ ceramics

C A Wang¹, H Z Pang¹, A H Zhang¹, M H Qin¹, X B Lu¹, X S Gao¹, M Zeng¹ and J-M Liu^1,2

1 Institute for Advanced Materials and Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials, South China Normal University, Guangzhou 510006, People’s Republic of China

2 National Laboratory of Solid State Microstructures and Collaborative Innovation Center for Advanced Microstructures, Nanjing University, Nanjing 210093, People’s Republic of China

E-mail: [email protected] and [email protected] Received 4 May 2015, revised 29 July 2015

Accepted for publication 4 August 2015 Published 3 September 2015

Abstract

In this work, Sm and Sc co-doped Bi1−xSmxFe1−yScyO3 (x = 0.00–0.20; y = 0.03) ceramics are fabricated by a rapid liquid phase sintering method, in order to develop single-phase multiferroics with large magnetization and polarization. X-ray diffraction and Raman

spectroscopic studies reveal that the ceramics are single-phase with a structural transition from rhombohedral to orthorhombic structures near x = 0.15. Electric and magnetic measurement results indicate that the transition significantly enhances the multiferroic properties, which stems from the Sm/Sc doping induced collapse of space-modulated spin structure and internal structural distortion. At an optimized composition of Bi0.85Sm0.15Fe0.97Sc0.03O3

(x = 0.15), a remanent polarization of 16.5 μC cm⁻², a magnetization 0.2020 emu g⁻¹, and a magnetodielectric effect of 0.46% can be obtained. These results clearly demonstrate a potential application for Sm/Sc doped BiFeO3 ceramics in the field of multiferroic devices.

Keywords: BiFeO3, ceramics, multiferroics, ferroelectric polarization, magnetization (Some figures may appear in colour only in the online journal)

C A Wang et al

Printed in the UK 395302

JPAPBE

© 2015 IOP Publishing Ltd 2015

48

J. Phys. D: Appl. Phys.

JPD

0022-3727

10.1088/0022-3727/48/39/395302

Papers

39

Journal of Physics D: Applied Physics IOP

doi:10.1088/0022-3727/48/39/395302 J. Phys. D: Appl. Phys. 48 (2015) 395302 (6pp)

the aim to suppress the spatial-modulated AFM spin ordering and stabilize ferroelectric distortion [9, 12–21]. For instance, Zhang et al [14] and Zou et al [16] reported the enhanced ferromagnetism in BFO ceramics through the substitution of Bi³⁺ by La³⁺, and Fe³⁺ by Ti⁴⁺ ions, respectively. However, the reports on doped BFO ceramics that exhibit large polariza- tion and magnetization simultaneously are still rather limited.

Among various co-doped BFO systems reported, substi- tution of non-magnetically active Sc ion into Fe-site of pure BFO compounds is able to improve sample density and ferro- electric properties due to the stable electronic configuration of Sc³⁺ ions. In addition, Sc³⁺ ion could help break the spatially antiferromagnetic spin-modulated structure, hence improving the ferromagnetic properties [11, 13, 21]. On the other hand, introduction of magnetically active Sm³⁺ ion into Bi-site was also reported to enhance the ferroelectric and magnetic prop- erties of BFO by reducing the concentration of oxygen vacan- cies and enlarging the spins canting in the corresponding sub-lattices [12]. Therefore, one can expect that the Sm and Sc co-doping may be able to greatly enhance the ferroelec- tric and magnetic properties simultaneously. Furthermore, the fabrication method of rapid liquid phase sintering was proved very effective for preparing pure phase BFO ceramics with good ferroelectric properties. For instance, Chen et al [8] reported a well saturated ferroelectric loop with a large remnant polarization of 25 μC cm⁻² in pure BFO ceramics.

This stimulated us to synthesise the novel single-phase Sm and Sc co-doped BFO ceramics by using rapid liquid phase sintered method and study in detail their crystalline, dielectric, ferroelectric, and magnetic properties. In this paper, we study dielectric, electric polarization, magnetization, and magneto- dielectric properties of Sm and Sc co-doped BFO ceramics.

It was found that Sm and Sc co-doped BFO ceramics exhibit enhanced room temperature electric polarization, magnetiza- tion, and magnetodielectric properties, which implies poten- tial applications in multiferroicial devices.

II. Experimental details

Sm and Sc co-doped Bi_1−xSmxFe_1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) and non-doped BFO (x = 0) ceramics were prepared by a rapid liquid phase sintered method. High purity Bi2O3, Fe2O3, Sm2O3 and Sc2O3 of analytical grade were used as starting materials, which were fully mixed using planetary milling with zirconia balls in ethanols for 24 h. After ball milling, the resultant powders were first granulated using 3 wt% poly vinyl alcohol (PVA) solutions and then uniaxi- ally pressed into pellets with 11.5 mm in diameter and 1 mm in thickness by applying a pressure of 9 MPa. The disk sam- ples were sintered by rapid liquid phase sintering process of 850 °C for 20 min at a heating rate of ~100 °C s⁻¹ and a cooling rate of ~10 °C s⁻¹.

The crystalline phase of non-doped and co-doped BFO sam- ples was determined by x-ray diffraction (XRD) (PANalytical X’Pert PRO diffractometer) with CuKa radiation. Raman spectroscopy was measured using inViaReflex micro-Raman system with 633 nm Ar ion laser source in backscattering

geometry. Dielectric characterizations were done by high performance frequency analyzer (Alpha-A, Novocontrol Technology) with the ac drive amplitude of 100 mV from 0.1 Hz to 2 MHz. The polarization hysteresis (P-E) loops was investigated by ferroelectric tester (Radiant Technology). The magnetic properties were carried out using a vibrating sample magnetometer incorporated into a physical properties meas- urement system (PPMS, Quantum Design).

III. Results and discussion

Figure 1(a) shows the XRD patterns of non-doped BFO and Bi_1−xSm_xFe_1−ySc_yO₃ (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics at room temperature. One can find that all the samples show good crystallinity without any second phase, indicating that the rapid liquid phase sintered method is suit- able for fabricating single-phase doped BFO ceramics. With increasing Sm concentration x, the intensities of some dif- fraction peaks, e.g. (0 0 6), (1 1 6) and (0 1 8), become weak and tends to disappear, starting to appear at x ⩾ 0.15. Further detailed analysis reveals that the XRD patterns for x < 0.15 were a rhombohedrally distorted perovskite structure (space group: R3c). Once x ⩾ 0.15, the structures become similar to that of orthorhombic SmFeO₃ (space group of Pbnm). It can also be seen from figure 1(b) that the sharp single peak (0 1 2) in the 2θ range of 21.5°–23° red shifts with increasing x. At x ⩾ 0.15, a new peak in the 2θ range of 21.5°–23° appears, meanwhile, the doubly splitting peaks in the 2θ range of 31°–33°overlap and form broadened peaks, see figure 1(c).

The observation shows a structural phase transition from R3c to Pbnm near x = 0.15. A similar phenomenon has also been observed previously in many rare earth doped BFO ceramics [9, 14, 15, 18].

Figure 1. XRD diffraction patterns for the non-doped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO₃ (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics.

20 30 40 50 60

22.0 22.5 23.0 31 32 33

x=0.20

x=0.15 x=0.10 x=0.05 x=0.00 x=0.05

x=0.20

x=0.15 x=0.10 x=0.00

(c) (b)

2

(Deg.) 2

(Deg.)

Intensity(a.u. )

Intensity(a.u. ) Intensity(a.u.)

2

(Deg.)

x=0.20 x=0.15 x=0.10 x=0.05

(012) (104) (110) (006) (202) (024) (116) (122) (018) (214)

(101) (200) (121) (022)(220) (040) (222)(301) (321)

(111)(020) (210)(002) (202) (212) (141) (123)

x=0.00

(a)

Room temperature Raman spectra of pure BFO and Bi1−xSmxFe1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics at wave number ranging from 100 to 650 cm⁻¹ are plotted in figure 2. The Raman active modes of the rhombo- hedral BFO with R3c structure can be given as: Г = 4A1 + 9E [22, 23]. From figure 2, four A1 modes and eight E modes can be found for all the samples. To clearly assign the observed peaks, the peak positions for non-doped BFO ceramics, which represent the active Raman modes, can be identified by fitting the measured spectra, in which the fitted spectra are decom- posed into individual Gaussian components. Four A1 peaks are observed at 142 cm⁻¹, 174 cm⁻¹, 223 cm⁻¹, and 437 cm⁻¹. The remaining eight E peaks at 121 cm⁻¹, 261 cm⁻¹, 282 cm⁻¹, 351 cm⁻¹, 373 cm⁻¹, 475 cm⁻¹, 529 cm⁻¹ and 630 cm⁻¹ are also identified in the spectra. These results are consistent with previous reports in polycrystalline BFO samples [20, 23]. With increasing x, the density of Raman active mode at 174 cm⁻¹ first increases and then decreases, while the density at 223 cm⁻¹ reduced monotonously. Hermet et al [24] sug- gested that the Bi-atoms are associatiated with the low fre- quency modes below 167 cm⁻¹, whereas the Fe atoms are mainly related to the modes between 152 and 262 cm⁻¹ by using first-principles calculation analysis. In our case, the former modes are related to the stereochemical activity of Bi³⁺ ions, which can affect the ferroelectric polarization. The latter modes can be ascribed to a significant destabilization of the FeO6 octahedron, which affects the canting of spins in irons via octahedral rotation [16]. Furthermore, the density of Raman peak around 625 cm⁻¹ is enhanced as increasing x.

Especially, the Raman peak is shifted toward a lower wave number once x ⩾ 0.15, which can be ascribed to the structural transition, which is consistent with the XRD results.

Figures 3(a) and (b) display the room temperature die- lectric constant (εr) and loss tangent (tan δ) as a function of frequency (0.1 Hz–2 MHz) for the non-doped BFO and Bi1−xSmxFe1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics. It is evident that the εr and tan δ values in all the sam- ples show a trend of decrease with increasing frequency at the lower frequency of <100 Hz, and remain almost unchanged at higher frequencies of >100 Hz. Generally, the charged

defects that come from the bismuth vacancies (V³_Bi⁻), oxygen vacancy (V_O²⁺), and Fe³⁺/Fe²⁺ fluctuation induced-localized charges in the BFO ceramics are able to follow the frequency of electric field, which contribute to the high dielectric con- stant and loss at low frequencies [25, 26]. In our samples, the increased εr and tan δ values at lower frequencies may result

Figure 2. Room temperature Raman scattering spectra of the non- doped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO3

(x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics.

100 200 300 400 500 600 700

E E E E E

E A-4¹

A-3¹ A¹-2 A¹-1 E

Intensity (a.u.)

Raman shift (cm

)

x=0.00 x=0.05 x=0.10 x=0.15 x=0.20

Figure 3. Frequency dependence of (a) dielectric constant and (b) dielectric loss for the non-doped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO₃ (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics.

80 120 160 200 240

10^-1 10⁰ 10¹ 10² 10³ 10⁴ 10⁵ 10⁶ 0.00

0.04 0.08 0.12 0.16

r

x=0.00 x=0.05 x=0.10 x=0.15 x=0.20

(a)

Frequency (Hz)

Ta n

x=0.00 x=0.05 x=0.10 x=0.15 x=0.20

(b)

Figure 4. (a)–(e) Polarization hysteresis (P-E) loops for non-doped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics, and (f) remnant polarization versus Sm-doping content of x.

-200 -100 0 100 200 -30

-20 -10 0 10 20 30

-200 -100 0 100 200 -30

-20 -10 0 10 20 30

-200 -100 0 100 200 -30

-20 -10 0 10 20 30

-200 -100 0 100 200 -30

-20 -10 0 10 20 30

-200 -100 0 100 200 -30

-20 -10 0 10 20 30

0.00 0.05 0.10 0.15 0.20 0

10 20 30

Polarization(µC/cm2 ) _{x=0.00 (a) (b)x=0.05}

(c)

Polarization(µC/cm2 )

x=0.10 (d)

Electric field (kV/cm) x=0.15

(e)

Polarization(µC/cm2 )

Electric field (kV/cm) x=0.20

(f)

Pr(µC/cm2 )

x

from these charged defects due to the Bi evaporation during sintering. In contrast, the weak dependence of εr and tan δ on frequency implies that intrinsic electrons/domains rather than the charged defects are dominating in the characteristics of εr and tan δ above 100 Hz. Moreover, the observed increase in εr with increasing in Sm content at all frequencies may be attributed to a possible increase in the ferroelectric polariza- tion dipoles. During heat treatment, the volatile nature of Bi might create defects in the system, such as V_O²⁺, V_Bi³⁻ and Fe³⁺/ Fe²⁺ fluctuation. These defects along with Bi 6s² lone pairs contribute to the charge polarization. The improved dielectric behaviors can be attributed to the reduction of these defects and favors the formation of intrinsic Bi 6s² lone pairs in the doped BFO ceramics.

Figures 4(a)–(e) plot polarization hysteresis (P-E) loops of non-doped BFO and Bi_1−xSmxFe_1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics at room temperature.

These loops are measured at 100 Hz with different elec- tric bias fields. Obviously, the well-saturated ferroelec- tric behaviors with large remanent polarization (Pr) are observed. The measured Pr are 10.6 μC cm⁻², 19.5 μC cm⁻², 20.8 μC cm⁻², and 16.5 μC cm⁻² for x = 0.00, 0.05, 0.10, 0.15, respectively, together with a weak polarization of 2.4 μC cm⁻² for x = 0.20. From the dependence of Pr

on the Sm doping concentration x, it can be found that Pr

is first increased and then decreased with increasing x. The improved ferroelectricity at the lower doping contents may be attributed to the enhanced stereochemical activity of Bi³⁺

ions due to Sm-substitution, as supported by Raman spectra analysis. While the degraded polarization properties at the higher doping level may be well explained by considering the effect of structural transition.

Figures 5(a)–(e) show the zero field cooled (ZFC) and field cooled (FC) magnetization curves of non-doped BFO and Bi_1−xSmxFe_1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics at a magnetic field of 500 Oe in the temperature

range of 5–400 K. For pure BFO, see figure 5(a), the ZFC and FC magnetization values reduce with decreasing the meas- ured temperature ranging from 50 K to 400 K, which shows conventional antiferromagnetic nature, whereas a small anomaly is seen at ~260 K, which exhibits a spin glass-like transition [27, 28]. Moreover, both ZFC and FC curves show sudden jump in magnetization below 50 K, indicating a weak ferromagnetic nature. For Sm/Sc doped BFO ceramics, see figures 5(b)–(e), both ZFC and FC curves exhibit a bump in the whole temperature range, which is similar to these reports for antiferromagnetic/ferri-ferromagnetic materials [27, 28].

With increasing x, the broad humps in the ZFC and FC curves are shifted towards lower temperature. Meanwhile, the mag- netization is also enhanced gradually. This means that the AFM nature in doped BFO is altered with the increase of Sm doped contents due to the stronger ferri-ferromagnetic interactions.

Room temperature magnetization-magnetic field (M-H) curves of non-doped and Bi_1−xSmxFe_1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics are shown in figure 6.

Clearly, no macroscopic magnetization can be observed in the pure BFO ceramics, which is consistent with other reports [12, 14, 16]. For the Sm/Sc doped BFO ceramics, as expected, the M-H curves exhibit clearly nonlinear behaviors. The measured remanent magnetization values (Mr) are 0.0209 emu g⁻¹, 0.0924 emu g⁻¹, 0.2020 emu g⁻¹, and 0.2568 emu g⁻¹ for x = 0.05, 0.10, 0.15, 0.20, respectively, which are remark- able higher than that of non-doped BFO (0.0008 emu g⁻¹). It is accepted that the enhanced ferromagnetism in doped BFO originates from the suppression or destruction of space modu- lated cycloidal spin structure by doping [9, 17]. For x < 0.15, the ferromagnetism increases slightly with rising amount of doping, meaning that a small amount of Sm can only slightly suppress but is still not enough to destroy the spin cycloid.

Once x ⩾ 0.15, a structural phase transition can be triggered, likely leading to the destruction of spin cycloids so that the

Figure 5. (a)–(e) ZFC and FC magnetization curves of the non-doped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics at a magnetic field of 500 Oe and a temperature range of 5–400 K.

0 100 200 300 400 0.0038

0.0040 0.0042 0.0044 0.0046 0.0048

0 100 200 300 400 0.0150

0.0165 0.0180 0.0195 0.0210

0 100 200 300 400 0.05

0.06 0.07 0.08

0 100 200 300 400 0.165

0.180 0.195 0.210

0 100 200 300 400 0.22

0.24 0.26 0.28 x=0.00

T(K)

M(emu/g)

FC ZFC H=500 Oe

(c)

(e) (d)

(b)

x=0.05 H=500 Oe

FC ZFC

M(emu/g)

T(K) (a)

x=0.10 H=500 Oe

FC ZFC

M(emu/g)

T(K)

x=0.15 H=500 Oe

FCZFC

M(emu/g)

T(K)

x=0.20 H=500 Oe

FC ZFC

M(emu/g)

T(K)

latent magnetization locked within the cycloid might be released [4–6, 12, 29]. Consequently, a significantly increased Mr value can be observed. Combining figures 4 and 6, it is found that the enhanced multiferroicial properties with Mr of 0.2020 emu g⁻¹ and Pr of 16.5 μC cm⁻² are obtained simul- taneously at an optimized composition of Bi0.875Sm0.125FeO3

(x = 0.15) ceramics, which should be attributed to the struc- tural phase transition.

Actually, the multiferroic properties have been reported in the Sm only doping BiFeO3 ceramics, for example Yuan et al, have reported simultaneously Pr of 15.09 μC cm⁻² and Mr of 0.071emu g⁻¹ in [12]. However, the magnetization is still weak.

Comparably, our study on Sm/Sc doped BiFeO3 ceramics shows better multiferroic properties, in which Mr of up to 0.2020 emu g⁻¹ along with Pr of 16.5 μC cm⁻² can be realized

at the optimized composition of Bi_0.85Sm_0.15Fe_0.97Sc_0.03O₃ ceramics. Therefore, Sc doping should play a major role in the enhanced multiferroical properties for the Sm/Sc doped BiFeO₃ ceramics.

To explore the coupling between ferroelectric and magnet- ization in the Sm/Sc doped BFO ceramics, we measured the variation of dielectric constant under various magnetic fields for an optimized composition, i.e. Bi_0.85Sm_0.15Fe_0.97Sc_0.03O₃ ceramics at room temperature. The magnetodielectric effect is defined as: MD = (ε(H) − ε(0))/ε(0) × 100%.

Figure 7(a) shows the dependence of MD on magnetic field at 100 kHz frequency. The magnetic field induced MD shift is ~0.46% at an applied field of 15 kOe. Similar observations were reported in other doped BFO ceramics. For example,

~0.04% for Sc-doped BFO nanoparticles [21]; ~1.05% for La/Zr doped BFO ceramics [30]. Generally, MD effect can be explained in the framework of the Ginzburg–Landau free energy theory for the second phase transition [31]. From the theory, one can obtain a quadratic dependence of the dielec- tric constant on magnetization i.e. MD ~ γΜ². Figure 7(b) shows the variation of MD versus M². It is found that the MD effect proportional to the M². Such a linear relation between MD and M² was also observed in many magnetoelectric mul- tiferroic materials [30, 31]. This MD effect in our samples reveals an interaction between polarization dipole and spin orders, which implies a potential application for multiferroic devices.

IV. Conclusions

We have successfully fabricated the Sm and Sc co-doped Bi_1−xSmxFe_1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics by using rapid liquid phase sintered method. XRD and Raman spectra confirmed that Sm-doping induced a struc- tural transform from rhombohedral to orthorhombic structure near at x = 0.15. Especially, the room temperature enhanced multiferroical properties with Pr of 16.5 μC cm⁻² and Mr of 0.2020 emu g⁻¹ were obtained at an optimized composition of Bi0.85Sm0.15Fe0.97Sc0.03O3 ceramics. Furthermore, a large magnetodielectric effect of 0.46% was observed at room tem- perature for the optimized composition. These results imply that the Sm/Sc doped BFO ceramics may have a potential application as multiferroic materials in the future.

Acknowledgments

This work was supported by the National Science Founda- tion of China (Grant Nos.: 51101063, 51272078, 51332007 and 51431006), the Program for Changjiang Scholars and Innovative Research Team in University (Grant No.

IRT1243), the Project for Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2014), and the Program for International Innovation Cooperation Platform of Guangzhou (No. 2014J4500016). The author C A Wang would like to thank the support by the Scientific Research Foundation of Graduate School of South China Normal University.

Figure 6. Room temperature magnetization versus magnetic field (M-H) hysteresis loops for the undoped BFO (x = 0.00) and Sm/Sc doped Bi1−xSmxFe1−yScyO3 (x = 0.05, 0.10, 0.15, 0.20; y = 0.03) ceramics.

-60 -40 -20 0 20 40 60

-0.9 -0.6 -0.3 0.0 0.3 0.6 0.9

Magnetization(emu/g)

Magnetic field (kOe)

x=0.00 x=0.05 x=0.10 x=0.15 x=0.20

Figure 7. Magneto-dielectric (MD) effect versus (a) magnetic field and (b) (magnetization)² for the Bi_0.85Sm_0.15Fe_0.97Sc_0.03O₃ ceramic. The inset of (b) shows a dependence of (magnetization)² on magnetic field for the Bi_0.85Sm_0.15Fe_0.97Sc_0.03O₃ ceramic.

-15 -10 -5 0 5 10 15 -0.5

-0.4 -0.3 -0.2 -0.1 0.0

0.00 0.03 0.06 0.09 0.12 -0.5

-0.4 -0.3 -0.2 -0.1 0.0

0 5 10 15

0.00 0.03 0.06 0.09 0.12

(b)

MD (%)

H(kOe) (a)

MD (% )

Magnetization

(emu/g)

Exp.

Linear Fit M2 (emu/g)2

H(kOe)

References

[1] Eerenstein W, Mathur N and Scott J 2006 Nature 442 759

[2] Ramesh R and Spaldin N A 2007 Nat. Mater. 6 21 [3] Hussain A, Xu X J, Yuan G L, Wang Y P, Yang Y, Yin J,

Liu J-M and Liu Z G 2014 Chin. Sci. Bull. 59 5161 [4] Singh A, Pandey V, Kotnal R and Pande D 2008 Phys. Rev.

Lett. 101 247602

[5] Dutta D P, Jayakumar O D, Tyagi A K, Girija K G, Pillai C G S and Sharma G 2010 Nanoscale 2 1149 [6] Wang K, Liu J-M and Ren Z 2009 Adv. Phys. 58 321 [7] Zhang J T, Lu X M, Zhou J, Sun H, Su J, Ju C C, Huang F Z

and Zhu J S 2012 Appl. Phys. Lett. 100 242413

[8] Chen X M, Zou Y H, Yuan G L, Zeng M, Liu J-M, Yin J and Liu Z H 2013 J. Am. Ceram. Soc. 96 3788

[9] Lennox R C, Price M C, Jamieson W, Jura M, Aladine A D, Murray C A, Tang C and Arnold D C 2014 J. Mater. Chem.

C 2 3345

[10] Singh H and Yadav K L 2015 J. Phys. D: Appl. Phys.

48 205001

[11] Rao T D and Asthana S 2014 J. Appl. Phys. 116 164102 [12] Yuan G and Or S W 2006 J. Appl. Phys. 100 024109

[13] Wang C A, Pang H Z, Zhang A H, Lu X B, Gao X S, Zeng M and Liu J-M 2015 Mater. Res. Bull. 70 595

[14] Zhang S T, Pang L H, Zhang Y, Lu M H and Chen Y F 2006 J.

Appl. Phys. 100 114108

[15] Kumar P, Panda C and Kar M 2015 Smart Mater. Struct.

24 045028

[16] Zou L Y, Yang R P, Lin Y B, Qin M, Gao X S, Zeng M and Liu J-M 2013 J. Appl. Phys. 114 034105

[17] Xu J et al 2013 J. Appl. Phys. 114 154103

[18] Xing Z B, Zhu X H, Zhu J M, Liu Z G and Al-Kassab T 2014 J. Am. Ceram. Soc. 97 2323

[19] Kalantari K, Sterianou I, Karimi S, Ferrarelli M C, Miao S, Sinclair D C and Reaney I M 2011 Adv. Funct. Mater.

21 3737

[20] Coondoo I, Panwar N, Bdikin I, Puli V S, Katiyar R S and Kholkin A L 2012 J. Phys. D: Appl. Phys. 45 055302 [21] Dutta D P, Mandal B P, Naik R, Lawes G and Tyagi A K 2013

J. Phys. Chem. C 117 2382

[22] Singh M K, Jang H M, Ryu S and Jo M H 2006 Appl. Phys.

Lett. 88 042907

[23] Bielecki J, Svedlindh P, Tibebu D T, Cai S Z, Eriksson S-G, Börjesson L and Knee C S 2012 Phys. Rev. B 86 184422 [24] Hermet P, Goffinet M, Kreisel J and Ghosez Ph 2007 Phys.

Rev. B 75 220102

[25] Gong Y F, Wu P, Hai X, Liu W F, Wang S Y, Liu G Y and Rao G H 2012 J. Phys. D: Appl. Phys. 45 355001 [26] Sharma H B, Nomita Devi K, Gupta V, Lee J H and Bobby

Singh S 2014 J. Alloys Compd. 599 32

[27] Ramachandran B and Rao R 2009 Appl. Phys. Lett. 95 142505 [28] Singh M K, Prellier W, Singh M P, Katiyar R S and Scott J F

2008 Phys. Rev. B 77 144403

[29] Marshall L G, Cheng J-G, Zhou J-S, Goodenough J B, Yan J-Q and Mandrus D G 2012 Phys. Rev. B 86 064417

[30] Kumar A, Yadav K L and Rani J 2012 Mater. Chem. Phys.

134 430

[31] Kumar A and Yadav K L 2010 Physica B 405 4650