NiTiO₃/MnFe₂O₄ ferrite composites: synthesis, optical, ferroelectric and magnetic properties

Monohydrate citric acid (C₆H₈O₇·H₂O) was purchased from Guangdong Xilong Chemical Co., China. Nickel (II) nitrate hexahydrate (Ni(NO₃)₂·6H₂O), iron (III) nitrate nonahydrate (Fe(NO₃)₃·9H₂O), manganese (II) chloride tetrahydrate (MnCl₂·4H₂O) and tetraisopropoxytitanium (Ti[OCH(CH₃)]₄) were purchased from Sigma Aldrich and used without further purification.

The (1 − x)NiTiO₃-xMnFe₂O₄ (where x = 0, 0.05 and 0.10) composites were synthesized by a sol-gel method and referred to as NTO-yMFO, where y stands for MnFe₂O₄(MFO) content. Citric acid solution was used as a solvent and a chelating agent. First, citric acid was dissolved into distilled water to make a solution with molar concentration of 1.5 M. Tetraisopropoxytitanium (TTIP) was dropped into the above solution under vigorous stirring for about 1 h at 70°C. After that, nickel nitrate was introduced into the solution and the mixed solution was kept stirring for around 1 h. Then, iron nitrate and manganese chloride were added with appropriate molar portions of metal nitrates fixed at Mn:Fe ratio of 1:2. The solutions were continuously stirred at 70°C until a gel was formed. The gel precursor was dried at 100°C for a day to remove excess water and turned into a foam block of xerogel. The xerogel was heated at 400°C for 2 h and annealed at 800°C in air for 3 h with a heating rate of 10°C/min. After cooling to room temperature, it was ground in an agate mortar to obtain composite powders. The overall synthesis procedure of (1 − x)NiTiO₃-xMnFe₂O₄ composite samples is schematically represented in Figure 1.

Fig. 1

For P-E loop measurement, (1 − x)NiTiO₃-xMnFe₂O₄ composite powders were well mixed with 3 wt.% polyvinyl alcohol (PVA, 5%) and then uniaxially pressed into a disc with a diameter of 10 mm and ∼ 1 mm under pressure of 150 MPa. The pressed pellets were sintered at 1200°C with a heating and a cooling rate of 5°C/min in air for 5 h in a box furnace. After sintering, the sintered pellets were polished, cleaned, and a silver electrode was deposited on both sides of the pellet.

X-ray diffraction (XRD) pattern was measured by an X-ray diffractometer (Philips-X’PertPro) using Cu Kα (λ = 1.54056 Å) radiation. The Raman spectra were employed by a micro Raman spectrophotometer (JASCO Raman NRS-3000). Microstructure of the samples was observed by scanning electron microscopy (Tabletop Microscope HITACHI TM4000Plus) with acceleration voltage of 15 kV. The room temperature optical reflectance data of the synthesized samples were investigated using a UV–Vis Spectrometer (JASCO V-750). Ferroelectric hysteresis loops (P–E) were tested by a ferroelectric tester (Radiant Precision Premier II). The magnetic properties of the NiTiO₃ and (1 − x)NiTiO₃-xMnFe₂O₄ composites were recorded at room temperature in range of −10 kOe ≤ H ≤ 10 kOe by a vibrating sample magnetometer (Lakeshore 7400).

Figure 2 depicts the XRD patterns of synthesized (1 − x)NiTiO₃-xMnFe₂O₄ (x = 0, 0.05 and 0.10) composites annealed at 800°C for 3 h. All samples included the observed diffraction peaks at 2θ = 24.03, 33.16, 35.72, 40.76, 49.54, 53.90, 57.35, 62.35, and 64.06, which corresponds to the (hkl) lattice planes of (012), (104), (110), (113), (024), (116), (018), (124), and (300), respectively. It belongs to the rhombohedral structure with space group of NiTiO₃ structure. The observed peaks and corresponding planes are well matched with standard JCPDS 33-0960. It is noted that the expected diffraction peaks of MFO in composite materials can be detected diffraction peaks at 2θ = 30.31, 37.32 which exhibited a cubic spinel crystal structure with space group (JCPDS card number 74-2403). The relative intensity of the MnFe₂O₄ diffraction varied corresponding to the MnFe₂O₄ content. The X-ray diffraction spectra of the synthesized powders indicated that mixed crystalline spinel–ilmenite phases have been formed without any trace of foreign phases. These results exhibited the successful formation of the NTO and MFO crystal phases in the composites, and no prominent chemical reactions happened during the synthesis process, which obtained the individual separation of ferroelectric and ferromagnetic phases. Rietveld refinement was carried out for analysis of crystal structure. For refinement work, constituent information phases were used from the crystallography open database. The refined crystal parameters of the constituent phases and their composites are shown in Table 1. The presence of both phases in composites has not influenced the crystal structure of the individual phase, showing that no additional chemical interactions between ferroelectric and ferromagnetic phases form the impurity phase and and that their symmetry is maintained in the synthesized composite materials. Moreover, the addition of MnFe₂O₄ phase resulted in a slight change in lattice parameters of NiTiO₃ phase. The a and c lattice constants of NiTiO₃ increase slightly with an increase of MnFe₂O₄ content. Further, the a lattice constant of MFO had the opposite trend to the lattice constant of NiTiO₃. The addition of MFO caused an expansion of NTO lattice that induced strain, which can affect the electrical characteristics of the composite samples. It can be clearly seen that the diffraction peak intensity of the MFO phase increased with the increasing of MFO concentration, which confirmed the two phases coexisted in composites. The cell volume of NTO phase increased with an increase of MFO content. The increasing of unit cell volume of NTO can be ascribed to the expansion strain of the NTO phase. In contrast, the cell volume of the MFO phase increased, which means that the MFO lattice was expanded. The strain formation can be one of the critical parameters which influence the optical, electric, and magnetic properties of composite samples.

Fig. 2

Figure 3 shows the Lorentzian deconvoluted Raman spectra of (1 − x)NiTiO₃-xMnFe₂O₄ composites monitored in the range of 150–1000 cm⁻¹ at room temperature. The pristine NiTiO₃ presented ten Raman active modes (5A_g + 5E_g) of ilmenite structure, which are in agreement with reports in the literature [22]. However, the intensity of these vibration modes of Raman signals of composite samples showed the breadth of Raman modes. These results were due to the overlapping of Raman modes of MFO with vibration modes of ilmenite structure. The intensity of Raman modes of MFO was very weak in comparison with those of vibration modes in NiTiO₃ material. Therefore, these vibration modes were not observed clearly in Raman spectra.

Fig. 3

However, the overlapping of the Raman vibration modes of both phases caused the broadening Raman vibration modes. We can observe these modes by deconvolution of the Raman spectrum of the composite samples. Lorentzian least square fitting was used to fit the experimental data to analyze the line width. Ten Raman vibration modes corresponding to pure NiTiO₃ materials were observed from the fitting curves. In addition, the Raman vibration mode of MnFe₂O₄ at 653 cm⁻¹ was observed in the Raman spectra due to the symmetric stretching of oxygen atoms along Fe–O bonds in the tetrahedral sites in MnFe₂O₄ material [23]. The intensity of the Raman vibration band of MFO phase increased with an increase of MFO content, which revealed the MFO phase existed in the synthesized samples. No modes of impurity phase were detected from Raman spectra, which indicated the high purity of the prepared composite samples.

The SEM images of the (1 − x)NiTiO₃-xMnFe₂O₄ ceramics are presented in Figure 4. All images show remarkably clear boundaries; however, the grain size is not homogeneous and grains are in a wide range of size distribution. The average grain size of the samples was estimated using IMAGE-J software [24]. It is clear from the figures that the pure NiTiO₃ samples have smaller grains and more pore in comparison with composite samples. The average grain size of pure sample is about 3.7 μm. It increases to 4.7 μm and 5.0 μm for composite samples with 5 and 10 mol. % MnFe₂O₄, respectively. Moreover, the pores decrease with introducing MnFe₂O₄. The addition of MnFe₂O₄ phase in NiTiO₃ material can work as a sintering aid, which increases the rate of diffusion and enhances the performance of the densifying mechanism.

Fig. 4

The optical properties of synthesized (1 − x)NiTiO₃-xMnFe₂O₄ composite powders were studied by UV–Vis diffuse reflectance spectra, which is presented in Figure 5. The color of (1 − x)NiTiO₃-xMnFe₂O₄ composites changed from the light yellow color of pure NiTiO₃ to the black color of NiTiO₃-MnFe₂O₄ composites. The absorption spectrum of NiTiO₃ showed two main absorption regions: 300 nm – 550 nm (the transition ³A_2g(3F) → ³T_1g(3P)) and 650 nm – 900 nm (the transition ³A_2g(3F) → ³T_1g(3F)) [25, 26]. The optical absorption of NiTiO₃-MnFe₂O₄ composites exhibited a broad band in 300 nm – 550 nm region, which indicates that the band edge absorption has been extended to the visible light region when the MnFe₂O₄ was introduced into NiTiO₃ to form composites. The optical band gap was estimated using the Wood-Tauc method [27]. The optical absorption of materials was presented by equation (αhν) ∼ (hν-E_g)ⁿ, where α is the linear absorbance coefficient, hν is the photon energy, E_g is the optical band gap energy, and n is a constant which relates to different types of electronic transition (n = 2 and 1/2 for an indirect and direct bandgap material). The plot of (αhν)² as a function of photon energy (hν) of the synthesized powders was exhibited in Figure 5b. The optical band gap values were determined by extrapolating the linear portion of (αhν)² versus hν plot to the point at (αhν)² = 0 [28]. The relationship between bandgap energy and MFO content is illustrated in the inset of Figure 5b. The E_g value decreased with an increase of MFO content. It decreased from 2.23 eV to 1.38 eV for pure NiTiO₃ and NiTiO₃-MnFe₂O₄ composites with 10 mol.% MFO, respectively. These results revealed that the optical bandgap of NiTiO₃ were significantly changed by the MnFe₂O₄ phase. The bandgap energy of MnFe₂O₄ material is narrower than that of NiTiO₃ and this could result in decreasing the band gap value of the composites. The decrease of bandgap values of host materials due to ferrite phase is in agreement with the results which reported from previous studies [29, 30]. Moreover, there are many other factors that can be responsible for decreasing bandgap energy, such as strain, grain size effect, etc. [31, 32]. This result indicated the ability to modify the optical bandgap value by simple manipulation of the MFO content.

Fig. 5

The ferroelectric properties of prepared samples have been determined from the P-E loop recorded at room temperature under the constant field of 10 kV/cm, which is shown in Figure 6a–d. The P-E loops of prepared samples showed a typical P-E loop, indicating the ferroelectric properties of these samples. However, the shape of P-E loops depended on the ratios of phases between ferroelectric phase and ferrite phase. The pure NiTiO₃ without ferrite phase exhibited a small hysteresis loop. Increasing the ferrite phase to 5%, the P-E loop of this sample is well saturated with the calculated values of remanent polarization (+P_r) of 0.063 μC/cm² and saturation polarization (+P_s) of 0.108 μC/cm². These values are higher than those of the pure NiTiO₃ sample with remanent polarization (+P_r) of 0.017 μC/cm² and saturation polarization (+P_s) of 0.056 μC/cm².

Fig. 6

It showed that both saturation and remanent polarization values of synthesized composites increased with an increase of ferrite content. The value of P_s increased with MnFe₂O₄ content, which could be related to the space charge contribution of the ferrite phase [33]. The addition of a magnetic phase into a ferroelectric phase generally caused ferroelectric properties to decline because the presence of the ferromagnetic phase resulted in a leakage current in the composite, consequently the ferroelectric properties of the composite decreased [34, 35]. However, the addition of the ferromagnetic phase in the ferroelectric material can enhance the ferroelectricity at certain content due to attribution of coherent interfaces. The ferroelectric properties rise when ferroelectric content increases to a threshold [36]. In our work, we observed that ferroelectric properties increased at 5% MnFe₂O₄ content. However, the lossy current occurred with decreasing of polarization at a high electric field when MnFe₂O₄ content was 10%. An increase of coercivity value and a round-shaped P-E loop at high MFO content can be attributed to higher loss and a rise in conductivity in the composites [30, 36]. Our results showed similar trends with some other system composites of the ferroelectric and ferromagnetic phase reported in the literature, such as Na_0.5Bi_0.5TiO₃ − BaFe₁₂O₁₉ [33], CoFe₂O₄-Pb_0.7Ca_0.3TiO₃ [37]. Sharma et al. observed that ferroelectric behaviors of CoFe₂O₄-Pb_0.7Ca_0.3TiO₃ composites at room temperature were strongly dependent on ferrite concentration [37]. The polarization value of CoFe₂O₄-Pb_0.7Ca_0.3TiO₃ composite increased with the increasing of ferrite content. Pattanayak et al. showed that the values of saturation polarization and remanent polarization of Na_0.5Bi_0.5TiO₃-BaFe₁₂O₁₉ composites increased with BaFe₁₂O₁₉ content up to 30 wt.% and then dramatically decreased because of an increase of leakage current through ferrite paths [33]. Jain et al. reported on the ferroelectric and magnetic properties of Ba_0.83Ca_0.10Sr_0.07TiO₃-MnFe₂O₄ multiferroic composites [19]. Ba_0.83Ca_0.10Sr_0.07TiO₃ material showed a ferrimagnetic hysteresis loop with considerable saturation magnetization; however, it exhibited a slight reduction in dielectric and ferroelectric with MnFe₂O₄ addition [19].

Figure 7 shows the leakage current of pristine NiTiO₃ and (1 − x)NiTiO₃-xMnFe₂O₄ composites as a function of an applied electric field. The leakage current of all samples increased with the increasing of an applied field. The increment of leakage current versus rising of the electric field is non-linear, which suggested non-Ohmic characteristic of this system. In addition, it also increased with increasing of MnFe₂O₄ content. The result may be due to the attribution of lower resistive MnFe₂O₄ material in a higher resistive NiTiO₃ phase.

Fig. 7

Figure 8 showed the magnetization vs. magnetic field hysteresis loops for the spinel–ilmenite (1 − x)NiTiO₃-xMnFe₂O₄ composite samples at room temperature. The pure NiTiO₃ consisted of paramagnetic materials, which is consistent with reports in the literature [5]. The loops of NiTiO₃-MnFe₂O₄ composites clearly exhibited the existence of an ordered magnetic structure in the ilmenite-ferrite composite. Magnetic parameters of synthesized composites are presented in Table 2. The magnetic behavior of (1 − x)NiTiO₃-xMnFe₂O₄ composites showed similar behavior, with a clear S-shaped ferromagnetic hysteresis loop. The magnetic saturation magnetization (M_s), remanent magnetization (M_r), and coercive field (H_c) of the composites have a strong dependence on the addition of the MFO. As the analysis in the XRD above, results implied that the MnFe₂O₄ ferrite phase formed in the composite samples, which explained why the magnetic properties of these samples were enhanced and exhibited the ferromagnetic behavior. The increase in magnetic properties of synthesized composites is expected because magnetic properties of composites depend on the ferrite amount added into composites. The remanent and saturation magnetization increased with an increase of MFO content, which indicated a significant increase in magnetic characteristics with the MFO addition in NTO. The addition of MFO phase resulted in an increase of M_s and M_r, which play a crucial role in determining the magnetization of synthesized composites because NiTiO₃ is a nonmagnetic material. The higher saturation magnetization in composite samples with higher MFO content could be explained by the individual ferrite grain distribution in composite, which acts as a center of the magnetization and the saturation magnetization of the composites is the vector sum of these individual contributions. The clear P–E and M–H hysteresis loops of the synthesized composite samples showed that both ferroic properties in (1 − x)NiTiO₃-xMnFe₂O₄ composites existed at room temperature.

Fig. 8