Influence of Mn dopants on the structure and multiferroic properties of a Bi_0.90Ho_0.10FeO₃ thin film

Abstract

Mn substituted Bi_0.90Ho_0.10FeO₃ (BHFO) thin films having the compositions Bi_0.90Ho_0.10Fe_1−xMn_xO₃ (BHFMO) (x = 0, 0.01, 0.03, 0.05) were synthesized via chemical solution deposition. The influence of the Mn dopants on the structure and multiferroic properties of the Bi_0.90Ho_0.10FeO₃ thin film is systematically investigated. X-ray diffraction, Rietveld refinement and Raman spectra analysis reveal that the crystal structure of the BHFMO thin film is rhombohedral (R3c:H + R3̄m:R). More importantly, the percentage of the space groups R3c:H and R3̄m:R is changed with the substitution of Mn into the Fe-site in the BHFO thin film, which indicates the presence of a structural transition in BHFMO. (X-ray photoelectron spectroscopy) XPS analysis confirms that (Ho, Mn) co-doping decreases the oxygen vacancy concentration, showing less Fe²⁺ ions in the co-doped BHFMO thin films compared with the BHFO thin film. The leakage current of the BHFMO_x=0.01 thin film is 5.398 × 10⁻⁴ A cm⁻² at 400 kV cm⁻¹, which is reduced effectively compared with BHFO (2.63 × 10⁻³ A cm⁻²). Well saturated hysteresis loops with a large remanent polarization of ∼80 μC cm⁻² in the BHFMO_x=0.01 thin film is observed, due to the reduced leakage current and the structural transition. The BHFMO_x=0.01 thin film also shows enhanced ferromagnetism with saturated magnetization (M_s = 5.24 emu cm⁻³) due to the destruction of the antiferromagnetically ordered spins arising from a structural transition.

---

Introduction

Materials that have coupled electric, magnetic, and structural order parameters that result in simultaneous ferroelectricity, (anti)-ferromagnetism and ferroelasticity are known as multiferroics.^1,2 Multiferroic materials are promising candidates for designing emerging electronic devices like multiple-state memories, magnetic data-storage media, actuators, sensors, and spintronic devices for different technological applications.^3–5 Among those multiferroics, bismuth ferrite (BiFeO₃, BFO), which exhibits simultaneous ferroelectric and G-type antiferromagnetic orders, is almost the only series of single phase multiferroics at room temperature.⁶ BFO possesses a rhombohedrally distorted perovskite structure with the space group R3c (a = b = c = 5.63 Å, α = β = γ = 59.4°).^1,7 BFO is the most promising from an practical application point of view due to its high ferroelectric Curie (T_C ∼ 1103 K) and Neel (T_N ∼ 643 K) temperatures and its giant remanent polarization of P_r ≈ 100 μC cm⁻².^8–10

In contrast, the practical application of BFO has been hindered seriously by high leakage current density, likely a result of defects and nonstoichiometry.^11,12 On the other hand, BFO exhibits a very weak magnetism due to a residual moment from a canted spin structure.¹ To overcome these problems, various attempts have been performed to improve the multiferroic properties of BFO through Bi-site substitution with Ho, Gd, Tb^13–15 and Fe-site substitution with Cr, Mn, Cu, Ti.^16–18

Furthermore, since the ferroelectricity and magnetism in BFO originates from the lone pair electrons of Bi³⁺ ions and partially filled “d” orbital of Fe³⁺ ions, respectively, the A-site and B-site codoping such as La and Mn, Ca and Mn, Sr and Ti^19–21 etc., exhibits a further enhancement in multiferroics properties of BFO.

Ho substitution for the Bi-site is an effective way to decrease leakage current and enhance the magnetization of BFO.¹³ Mn substitution for the Fe-site contributed to improving ferroelectric property.^17,22 Thus, the co-doping of BFO thin film with Ho and Mn ions may further enhance multiferroics properties.

In present work, Bi_0.90Ho_0.10Fe_1−xMn_xO₃ (BHFMO) (x = 0, 0.01, 0.03, 0.05) were synthesized by chemical solution deposition. Influence of Mn dopants on the structure and multiferroic properties of Bi_0.90Ho_0.10FeO₃ thin film is systematically investigated.

Experimental

Bi_0.90Ho_0.10Fe_1−xMn_xO₃ (BHFMO) (x = 0, 0.01, 0.03, 0.05) thin films were successfully deposited on (FTO)/glass substrates by chemical solution deposition and sequential-layer annealing process. Bi(NO₃)₃·5H₂O, Ho(NO₃)₃·6H₂O, Fe(NO₃)₃·9H₂O and C₄H₆MnO₄·4H₂O were used as the raw materials to prepare the Bi_0.90Ho_0.10Fe_1−xMn_xO₃ precursor solution. These raw materials were mixed together with the atomic ratio of 0.90 : 0.10 : 1 − x : x (5% mol excess to compensate the Bi loss).

Then 2-methoxyethanol serves as solvent was added into the starting materials and stirred for 30 min to obtain a clear solution. The acetic anhydride with a volume ratio of one third to 2-methoxyethanol was added into the solution to dehydrate and adjust the pH value under constant stirring for 2 h to obtain a homogeneous precursor solution. The concentration of the solution was 0.3 mol L⁻¹.

The precursor solution was spin coated on FTO/glass substrates with 4000 rpm for 15 s. The name of the spin-coater used to coat the thin films was CHEMAT TECHNOLOGY SPIN-COATER (KW-4A). After spin-coating, the thin films were dried on a hot plate at 200 °C for 10 min, and then annealed in a rapid thermal process furnace at 550 °C for 10 min in the air atmosphere. These processes were repeated 15 times to obtain the desired film thickness. Circular Au electrodes of 0.8 mm in diameter were sputtered on the film surface through a shadow mask to investigate the electrical behaviors. After annealing at 285 °C for 20 min, the electrodes and the substrate could be completely contacted.

Microstructures of the thin films were confirmed by X-ray diffraction (XRD, D/MAX-2200, Rigaku) with Cu Kα radiation (λ = 0.154056 nm). Simulation of crystal structure based on the measured XRD data was performed by a Maud program. Raman spectra measurements were performed by a RENISHAW inVia Raman microscope with argon laser (532 nm). Surface and cross-sectional morphology of the thin films were characterized by a field emission scanning electron microscopy (FE-SEM, S4800, Hitachi). The valence states of ions were investigated on the surface of all the films by X-ray photoelectron spectroscopy (XPS, XSAM800, Kratos Ltd., Britain). Agilent B2901A was used to measure the leakage current density of the thin films. The electric hysteresis loops of the films were tested by an aixACCT TF-Analyzer 2000. The magnetic properties of the thin films were analyzed by the MPMS-XL-7 superconducting quantum interference magnetic measuring system.

Results and discussion

Fig. 1(a) shows the typical XRD patterns of the BFO, BHFO and BHFMO thin films deposited on the FTO/glass substrates. All peaks in the XRD patterns of the BFO thin films are indexed for a rhombohedrally distorted perovskite structure with space group R3c. From the inset of Fig. 1, it is clear that, with Ho-doping and (Ho, Mn) co-doping, the relative diffraction intensity of the (104) peak decreases while the intensity of the (110) peak increases and finally the (110) peak becomes much stronger than the (104) peak; meanwhile, the doublet (104)/(110) is gradually merged into a single broad peak. This observation clearly suggests the existence of structure transition in BHFO and BHFMO thin films^3,23 and this structure transition is due to the fact that the smaller Ho³⁺ (0.901 Å) and Mn²⁺ (0.46 Å) have entered into the BiFeO₃ lattice by substituting the larger Bi³⁺ (1.03 Å) and Fe³⁺ (0.645 Å).²⁴ In addition, the (110) peak of BHFMO thin film is shifted slightly toward lower angles compared with BHFO thin film, suggesting that the structure of BFO thin film is further changed with (Ho, Mn) co-doping. The broadening of (110) peak observed in BHFO and BHFMO thin films indicate that the grain size of BFO is reduced with Ho doping and (Ho, Mn) co-doping.

Fig. 1 (a) XRD patterns of BFO, BHFO and BHFMO thin films. Inset presents the magnified XRD patterns in the 2θ range of 31.0°–33.3°; (b) Rietveld refined XRD patterns of BFO, BHFO and BHFMO_x=0.01 thin films.

To further analyze the detailed subtle structural changes of BHFO and BHFMO thin films, the measured XRD patterns are simulated by Rietveld refinement using MAUD program,²⁵ shown in Fig. 1(b). Refined structural parameters along with profile R-factors are listed in Table 1. The fitting between the experimental spectra and the calculated values is relatively good based on the relatively lower R_w, R_wnb and R_b values. In the light of the refinement results, the schematics of the BFO crystal structure are shown in Fig. 2. XRD pattern of pure BFO thin film is refined with rhombohedral structure with space group R3c:H (Fig. 2(a)), while Rhombohedral structure with space group R3c:H and R3̄m:R (Fig. 2(b)) is obtained for BHFO and BHFMO thin films. This result indicates that the structure of BFO gradually changes from rhombohedral (R3c:H) to biphasic structure (R3c:H + R3̄m:R) with Ho doping and (Ho, Mn) co-doping. Furthermore, the percentage of space group R3̄m:R in BHFMO thin film is increased compared with BHFO thin film, implying that (Ho, Mn) co-doping contributes to further structural transition in BFO thin film.

Table 1 Lattice parameters obtained from the refinement of XRD data of BFO, BHFO and BHFMO thin films

Samples	Crystal structure	Space group	Lattice parameters (Å)	R-factors (%)
BFO	Rhombohedral	R3c:H	a = 5.5772	Rw = 8.70
			c = 13.8669	Rwnb = 7.93
				Rb = 6.56
BHFO	Rhombohedral	R3c:H(84.22%)	a = 5.5957	Rw = 9.19
			c = 14.1569	Rwnb = 10.92
		R3̄m:R(15.78%)	a = 3.9620	Rb = 6.95
			α = 89.4092
BHFMOx=0.01	Rhombohedral	R3c:H(61.07%)	a = 5.9995	Rw = 7.39
			c = 13.6738	Rwnb = 7.00
		R3̄m:R(38.93%)	a = 3.9474	Rb = 5.71
			α = 89.4436

---

Fig. 2 Schematics of BFO crystal structure (a) R3c:H space group and (b) R3̄m:R space group.

The structural transition of BHFO and BHFMO thin films is further reflected in the Raman spectra, shown in Fig. 3. According to group theory, BiFeO₃, a highly rhombohedrally distorted perovskite structure with R3c space group, should have 13 (Γ = 4A₁ + 9E) Raman active modes.^26,11 The Raman modes observed in pure BFO film are in good agreement with the rhombohedrally distorted (R3c) perovskite structure in terms of relative scattering intensity and mode frequency. However, in Ho-doped BFO film, the Raman spectra show quite different features compared with the pure thin film. A₁-1, A₁-2, and A₁-3 modes, related to Bi–O bond vibrations,⁹ are becoming broader with a significant reduction of intensity. The suppression and broadening of low frequency A₁-1, A₁-2, and A₁-3 modes are maybe due to the change of Bi–O bond length induced by the smaller Bi site substitution,²⁷ which is a clear indication of structural transition. In addition, the broadening of Raman peaks indicates BHFO has smaller grain sizes in comparison with pure BFO film,⁴ which is also confirmed by XRD studies.

Fig. 3 Raman spectra of BFO, BHFO and BHFMO thin films.

For BHFMO thin film, on one hand, the suppression and broadening of low frequency A₁-1, A₁-2, and A₁-3 modes also indicate the structural transition. On the other hand, the Raman spectra of BHFMO film show a strong and wide E-9 mode around 620 cm⁻¹ compared with BHFO thin film, which is associated with the Fe–O bonds.¹¹ The strengthening and broadening of E-9 mode indicates the structure of BHFMO thin film is further changed by (Ho, Mn) co-doping. Consequently, it can be inferred from the Raman spectra that both Ho and Mn ions affect BFO crystalline structure in the Ho and Mn co-doped BFO films.

Fig. 4 shows the surface morphologies images of BHFO and BHFMO thin films and a representative cross sectional micrograph of the BHFMO_x=0.03 thin film. The thickness of BHFMO_x=0.03 thin film is about 540 nm. The average crystallite size “D” was calculated from XRD peak broadening using the Debye Scherrer's equation:²⁸

D = 0.94λ/β cos θ (1)

where D is crystallite size, λ is the X-ray wavelength, β is the full width at half maxima, and θ is the Bragg angle. The average crystallite size calculated from Scherer's formula is found to be 20 nm for BHFMO_x=0.01 thin film. SEM observations show that the grain size of the BHFMO_x=0.01 thin film is about 80–100 nm. In general, the calculated value according to Scherrer's law is the crystallite size, which is usually smaller than the real grain size seen from SEM image. This difference in grain size and crystallite size obtained using Scherrer's formula are attributed to the stress in BFO films, which affects the FWHMs of the diffraction peaks.²⁹

Fig. 4 The surface morphology images of (a) BHFO, (b) BHFMO_x=0.01, (c) BHFMO_x=0.03, (d) BHFMO_x=0.05 and a typical cross sectional micrograph of (e) BHFMO_x=0.03 thin film.

Obviously, BHFMO thin films show decreased grain size compared with BHFO. The decrease of grain size of BHFMO thin films can be interpreted by the suppression of oxygen vacancy concentration, which will be discussed in the following by the XPS results. This suppression results in slower oxygen ion motion and consequently lower grain growth rate.³⁰

In addition, the reduction in the grain size is due to the substitution of Mn ions into the Fe site in BHFO. Both rare earth Ho and transition metal Mn ions may also act as nucleation centers for the perovskite structure, which increases the number of nucleus.³¹ Therefore, the grain growth rate is suppressed. Compared with BHFO, the surface morphology of BHFMO becomes more uniform and dense. However, when the Mn doping content increases to 0.05, some voids can be observed from the surface morphology, which is thought to be one of the causes of the high leakage current.

The XPS analysis was carried out to investigate the chemical states and binding energies of the thin films. Herein, the C 1s peak located at 284.6 eV is used as the criterion to rectify the binding energy of XPS spectra. Fig. 5(a) shows the XPS spectra of the Fe 2p region of the BHFO and BHFMO thin films. From Fig. 5(a), two main XPS peaks for Fe 2p_1/2 and Fe 2p_3/2 can be observed in all the thin films. The Fe³⁺ and Fe²⁺ ions coexisting in these four thin films are confirmed.

Fig. 5 XPS spectra of (a) Fe 2p for BHFO and BHFMO thin films (b) Mn 2p_3/2 for BHFMO_x=0.05 thin film (c) Ho 4d for BHFMO_x=0.01 thin film.

According to the ratio of the fitted peak areas, the concentration ratios of Fe³⁺ to Fe²⁺ ions in BHFO, BHFMO_x=0.01, BHFMO_x=0.03 and BHFMO_x=0.05 thin films are calculated as 56.21 : 43.79, 61.64 : 38.36, 63.47 : 36.53 and 58.50 : 41.5, respectively. This result indicates that the concentration of Fe²⁺ ions in the BHFMO films is less than that in the BHFO film. The electronic conduction in BFO occurs due to the hopping electron mechanism between Fe²⁺ and Fe³⁺ when oxygen vacancies are present in the lattice.⁹ The oxygen vacancies present in the lattice act as a bridge between Fe²⁺ and Fe³⁺ and significantly affect the leakage current of BFO.³² More Fe²⁺ ions imply more oxygen vacancies.³³ Therefore, the XPS analysis also confirms that oxygen vacancies in the BHFMO thin films are reduced compare to the BHFO counterpart.

Fig. 5(b) shows the fitted narrow-scan spectra of the Mn 2p_3/2 core level of BHFMO_x=0.05 thin film. The Mn²⁺ and Mn³⁺ coexistence in the BHFMO_x=0.05 film is confirmed by dividing the Mn 2p_3/2 peak into two subpeaks. In addition, due to different valence of Mn, the substitution of Mn at Fe site can convert Fe²⁺ to Fe³⁺ by the following process:³⁴

Fe²⁺ + Mn³⁺ ⇒ Fe³⁺ + Mn²⁺ (2)

The above process indicates that doping of Mn ions can restrain the formation of Fe²⁺. Therefore the ionic conduction resulting from the electron transfer between Fe²⁺ and Fe³⁺ can be greatly reduced. Fig. 5(c) shows the XPS spectra of the Ho 4d region of the BHFMO_x=0.01 thin film. The XPS results in Fig. 5(b) and (c) evidence the substitution of Ho and Mn into the Bi and Fe site in BFO, respectively.

Fig. 6 shows leakage current density versus electric field (J–E) characteristics of BHFO and BHFMO thin films measured at RT. The leakage current density in BHFO thin film increases sharply as the electric field increases beyond 100 kV cm⁻¹, which is a main reason why the polarization property can not be measured accurately at RT.¹⁷ The BHFO thin film exhibits the higher leakage current density due to its higher concentration of oxygen vacancies,³⁵ which is discussed in the XPS results. Normally, oxygen vacancies act as trapping centers for electrons to be mobile for conduction by the electric field.³³ So, a higher level of oxygen vacancies leads to a higher density of free carriers and hence, a higher leakage current in the BHFO thin film.³⁶

Fig. 6 The current density–electric field (J–E) characteristic of the BHFO and BHFMO thin film under positive biases (inset: schematic illustration of the test method for electrical properties in this study).

However, Mn substitution at Fe-site is indeed efficient in improving the leakage current density. The leakage current densities of BHFMO_x=0.01 and BHFMO_x=0.03 are 5.398 × 10⁻⁴ A cm⁻² and 3.725 × 10⁻⁴ A cm⁻² at 400 kV cm⁻¹ respectively, which are much lower than that of BHFO thin film (2.63 × 10⁻³ A cm⁻²). This reduction in the leakage current density is due to the reduction of oxygen vacancies. Another reason is the formation of defect complexes between and and the applied electric field can not overcome the electrostatic attraction force between and in BHFMO films.³⁷ Therefore, the is restricted by the and can not move to serve as the trapping centers for electrons. In addition, the co-doping of Ho and Mn results in significant reduction in grain size and hence provides large insulating boundaries between the grains, which act as a barrier for current conduction and consequently reduce leakage current. The improved microstructure is also beneficial to decrease the leakage current density of BHFMO thin films. Further increasing Mn doping content to 0.05, the leakage current density in BHFMO_x=0.05 thin film increases and is higher than that in BHFO thin film, which should be mainly due to the voids observed from the surface morphology.

Fig. 7 depicts P–E hysteresis loops of BHFO and BHFMO thin films measured at a frequency of 1 kHz. BHFO thin film exhibits nonsaturated P–E hysteresis loops and the remanent polarization values are difficult to be determined due to existence of the large leakage current density at high electric field. This film gets breakdown in the further increased electric field.

Fig. 7 Ferroelectric hysteresis loops of BHFO and BHFMO thin films measured at 1 kHz.

However, well saturated hysteresis loops with a large remanent polarization are obtained in BHFMO thin film indicating improved ferroelectric properties, and the remnant polarization (P_r) and coercive field (E_c) of BHFMO_x=0.01 thin film are 80 μC cm⁻² and 350 kV cm⁻¹ under the maximum applied electric field E_max of 833 kV cm⁻¹. The enhancement in ferroelectric properties of BHFMO thin film can be attributed to the low leakage current at high electric field and the improved anti-breakdown characteristic. On the other hand, the structure transition from rhombohedral (R3c:H (84.22%) + R3̄m:R (15.78%)) to rhombohedral (R3c:H (61.07%) + R3̄m:R (38.93%)) also contributes to enhancing the ferroelectric property. Furthermore, Ho³⁺ (0.901 Å) and Mn²⁺ (0.46 Å) ions have smaller ionic radius compared with Bi³⁺ and Fe³⁺, respectively, which may lead to a larger ion off-center in the Fe–O octahedral, so a larger ferroelectric polarization in the BFO unit cell could be obtained.

With the increase of Mn²⁺ ions doping content, P_r value (BHFMO_x=0.05) is gradually reduced to 60.26 μC cm⁻² and E_c is increased to 460 kV cm⁻¹. The ferroelectric polarization degradation in BHFMO_x=0.05 thin film should be due to the decrease of the effective electric field induced by the large leakage current as shown in Fig. 5. More importantly, this phenomenon can be explained by the higher content of in BHFMO_x=0.05 thin film, which may align along the direction of spontaneous polarization (P_S). The local internal fields produced by parallel to P_S will provide a driving force for domain backswitching, leading to the ferroelectric domains in BHFMO_x=0.05 film are difficult to be switched by the external field.³⁷ Therefore, P_S of BHFMO_x=0.05 film is decreased, accompanying with a relative higher coercive field. Moreover, the clamping effect association with oxygen vacancies hinders sufficient reorientation of ferroelectric domains during electrical poling and leads to a decrease of P_S.³⁸

Fig. 8 shows the magnetic hysteresis (M–H) loops of the BHFO and BHFMO thin films, the inset is an enlarged section of the loops. Generally, the local short-range magnetic ordering of BFO is G-type antiferromagnet and each Fe³⁺ spin is surrounded by six canted antiparallel spins on the nearest Fe neighbors. The spins are in fact not perfectly antiparallel, as there is a weak canting moment caused by the local magnetoelectric coupling to the polarization. This canting allows net magnetic moments.³⁹ However, a superimposed spiral spin structure with an incommensurate long wavelength period of 62 nm cancels the macroscopic magnetization.⁴⁰ Thus, pure rhombohedral BFO shows no macroscopic magnetization. In this work, BHFO thin film exhibits weak ferromagnetic behavior and the saturation magnetization (M_s) is 3.48 emu cm⁻³. In comparison with BHFO thin film, BHFMO_x=0.01 presents further enhanced magnetization with a saturation magnetization (M_s = 5.24 emu cm⁻³). In our work, there are four possible explanations for the origin and significant enhancement of spontaneous magnetization: (i) by substituting iron atoms at the B sites with Mn, a local ferrimagnetic spin configuration can be expected to form because of the differences in magnetic moment between the B-sites occupants.³⁴ In addition, the difference in magnetic moment of Mn²⁺ and Fe³⁺ interrupts the spiral spin structure, which results in enhancement of M_s for BHFMO_x=0.01 thin film;⁶ (ii) the content of Fe³⁺ estimated from XPS increases with the incorporation of Ho and Mn. Thus, the enhancement in the magnetization of the thin films is attributed to the presence of the Fe³⁺ ions and suppression of oxygen vacancies due to co-doping of Ho and Mn.⁸ (iii) Raman results have shown that Ho and Mn co-doping induces significant structure distortion of Fe–O bonds, which may result in a modification on the short-range canted G-type antiferromagnetic order of the nearest neighbor spins of Fe³⁺ and give a net magnetization.⁴¹ This structure distortion of Fe–O bonds may cause internal structure distortion related to the change in bond lengths of the Fe–O and bond angle of Fe–O–Fe. Fe–O–Fe angle is important because it controls both the magnetic exchange and orbital overlap between Fe and O, and as such it determines the magnetic ordering temperature and the conductivity.^3,26,40

Fig. 8 M–H curves of BHFO and BHFMO thin films and inset is the magnified M–H curve of each film.

In addition to all this, the structural transition in BHFMO_x=0.01 thin film destroys the spin cycloid and releases the latent magnetization locked within the cycloid structure. Hence, increasing in magnetization of the co-doping film may be also attributed to the collapse of space modulated spin structure.

Conclusions

Bi_0.90Ho_0.10Fe_1−xMn_xO₃ (BHFMO) thin films with x = 0, 0.01, 0.03, 0.05 were successfully deposited on FTO/glass substrates by chemical solution deposition. The influence of Mn ion doping on the structure, morphologies, leakage current, ferroelectric properties and magnetic properties of the BHFMO thin film is investigated. X-ray diffraction, Rietveld refinement and Raman spectra analysis both confirm the structural transition in BHFMO thin film. The grain size of BHFMO thin films is decreased as confirmed by the SEM measurements. XPS analysis confirms that the concentration of oxygen vacancy in BHFMO thin films is lower than that in the BHFO thin film, and it also evidences that Mn²⁺ ion doping can restrain the formation of Fe²⁺ ions. A proper amount of Mn doping can decrease the leakage current densities of the BHFMO thin film. BHFMO_x=0.01 thin film shows the enhanced ferroelectric properties with a large remnant polarization value (80 μC cm⁻²) due to the reduced leakage current and the structural transition. In addition, BHFMO_x=0.01 thin film also exhibits enhanced magnetization with a saturation magnetization (M_s = 5.24 emu cm⁻³). The enhanced magnetization is attributed to the destruction of the antiferromagnetically ordered spins, which is related to the structural transition compared with pure BHFO thin film.

Acknowledgements

This work is supported by the Project of the National Natural Science Foundation of China (Grant no. 51372145); the Academic Leaders Funding Scheme of Shaanxi University of Science & Technology (2013XSD06); Doctorate Scientific Research Initial Fund Program of Shaanxi University of Science & Technology (BJ4-13); the Graduate Innovation Fund of Shaanxi University of Science & Technology (SUST-A04).

References

1. J. Wang, J. B. Neaton, H. Zheng, V. Nagarajan, S. B. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. G. Schlom, U. V. Waghmare, N. A. Spaldin, K. M. Rabe, M. Wuttig and R. Ramesh, Science, 2003, 299, 1719.
2. J. Wu, J. Wang, D. Xiao and J. Zhu, ACS Appl. Mater. Interfaces, 2012, 4, 1182–1185.
3. Reetu, A. Agarwal, S. Sanghi, Ashima, N. Ahlawat and Monica, J. Appl. Phys., 2012, 111, 113917.
4. J. S. Park, Y. J. Yoo, J. S. Hwang, J.-H. Kang, B. W. Lee and Y. P. Lee, J. Appl. Phys., 2014, 115, 013904.
5. R. Das, G. Gopal Khan and K. Mandal, J. Appl. Phys., 2012, 111, 104115.
6. K. G. Yang, Y. L. Zhang, S. H. Yang and B. Wang, J. Appl. Phys., 2010, 107, 124109.
7. N. Li, J. Wu, Y. Jiang, Z. Xie, L. Zheng and Z.-G. Ye, J. Appl. Phys., 2013, 113, 054102.
8. K. Chakrabarti, K. Das, B. Sarkar, S. Ghosh, S. K. De, G. Sinha and J. Lahtinen, Appl. Phys. Lett., 2012, 101, 042401.
9. P. C. Sati, M. Arora, S. Chauhan, S. Chhoker and M. Kumar, J. Appl. Phys., 2012, 112, 094102.
10. J. Li, J. Wang, M. Wuttig, R. Ramesh, N. Wang, B. Ruette, A. P. Pyatakov, A. K. Zvezdin and D. Viehland, Appl. Phys. Lett., 2004, 84, 5261–5263.
11. B. Ramachandran and M. S. Ramachandra Rao, J. Appl. Phys., 2012, 112, 073516.
12. J. Wu and J. Wang, J. Phys. Chem. C, 2010, 114, 19318–19321.
13. N. Jeon, D. Rout, I. W. Kim and S.-J. L. Kang, Appl. Phys. Lett., 2011, 98, 072901.
14. G. D. Hu, X. Cheng, W. B. Wu and C. H. Yang, Appl. Phys. Lett., 2007, 91, 232909.
15. Y. Wang and C.-W. Nan, J. Appl. Phys., 2008, 103, 024103.
16. J. K. Kim, S. S. Kim, W.-J. Kim, A. S. Bhalla and R. Guo, Appl. Phys. Lett., 2006, 88, 132901.
17. S. K. Singh and H. Ishiwara, Appl. Phys. Lett., 2006, 88, 262908.
18. P. Kharel, S. Talebi, B. Ramachandran, A. Dixit, V. M. Naik, M. B. Sahanal, C. Sudakar, R. Naik, M. S. R. Rao and G. Lawes, J. Phys.: Condens. Matter, 2009, 21, 036001.
19. B. Li, C. Wang, G. Dou and Z. Wang, RSC Adv., 2014, 4, 55889–55896.
20. Ji-Z. Huang, Y. Shen, M. Li and C.-W. Nan, J. Appl. Phys., 2011, 110, 094106.
21. Reetu, A. Agarwal, S. Sanghi and Ashima, J. Appl. Phys., 2011, 110, 073909.
22. S. K. Singh, H. Ishiwara, K. Sato and K. Maruyama, J. Appl. Phys., 2007, 102, 094109.
23. P. Suresh and S. Srinath, J. Appl. Phys., 2013, 113, 17D920.
24. J. Liu, M. Li, L. Pei, B. Yu, D. Guo and X. Zhao, J. Phys. D: Appl. Phys., 2009, 42, 115409.
25. K. Chakrabarti, K. Das, B. Sarkar and S. K. De, J. Appl. Phys., 2011, 110, 103905.
26. S. K. Srivastav, N. S. Gajbhiye and A. Banerjee, J. Appl. Phys., 2013, 113, 203917.
27. J. Xu, D. Xie, C. Yin, T. Feng, X. Zhang, G. Li, H. Zhao, Y. Zhao, S. Ma, T.-L. Ren, Y. Guan, X. Gao and Y. Zhao, J. Appl. Phys., 2013, 114, 154103.
28. W. Qin, Y. Guo, B. Guo and M. Gu, J. Alloys Compd., 2012, 513, 154–158.
29. G. Yiping, S. Kazuyuki, N. Kaori, M. Takeshi and K. Kazumi, Jpn. J. Appl. Phys., 2006, 45, 855–859.
30. M. Kumar and K. L. Yadav, Appl. Phys. Lett., 2007, 91, 242901.
31. C. M. Raghavan, E. S. Kim, J. W. Kim and S. S. Kim, Ceram. Int., 2013, 39, 6057–6062.
32. S. J. Clark and J. Robertson, Appl. Phys. Lett., 2009, 94, 022902.
33. Y. Shuai, S. Zhou, D. Bürger, M. Helm and H. Schmidt, J. Appl. Phys., 2011, 109, 124117.
34. G. S. Arya and N. S. Negi, J. Phys. D: Appl. Phys., 2013, 46, 095004.
35. H. Yang, Y. Q. Wang, H. Wang and Q. X. Jia, Appl. Phys. Lett., 2010, 96, 012909.
36. X. Qi, J. Dho, R. Tomov, M. G. Blamire and J. L. MacManus-Driscoll, Appl. Phys. Lett., 2005, 86, 062903.
37. G. D. Hu, S. H. Fan, C. H. Yang and W. B. Wu, Appl. Phys. Lett., 2008, 92, 192905.
38. M. Zou, H. Fan, L. Chen and W. Yang, J. Alloys Compd., 2010, 495, 280–283.
39. G. Catalan and J. F. Scott, Adv. Mater., 2009, 21, 2463–2485.
40. Reetu, A. Agarwal, S. Sanghi, Ashima and N. Ahlawat, J. Appl. Phys., 2013, 113, 023908.
41. W. Maoa, X. Wang, Y. Han, X. 'ao Li, Y. Li, Y. Wang, Y. Ma, X. Feng, T. Yang, J. Yang and W. Huang, J. Alloys Compd., 2014, 584, 520–523.

---