Magneto-electric/dielectric and fluorescence effects in multiferroic xBaTiO₃–(1 − x)ZnFe₂O₄ nanostructures

Abstract

Multiferroic composites of xBaTiO₃–(1 − x)ZnFe₂O₄ (BTZF) [x = 0.25 (BTZF2575), 0.35 (BTZF3565), 0.45 (BTZF4555), 0.50 (BTZF5050) and 0.75 (BTZF7525)] nanostructure have been synthesized by a sol–gel method. Different types of nanostructural shapes and sizes have been obtained by the effect of ionic radii, surface energy and poly vinyl alcohol, which enhances the magneto-electric/dielectric interaction between BT/ZF phases. The crystalline phases of BTZF composite are confirmed by X-ray diffraction, and nanostructural dimensions and shape by transmission electron microscopy. The improvement in magnetization of BTZF is dependent upon the size and shape of the nanostructure, stoichiometric ratio and strength of occupation of cations at octahedral and tetrahedral sites. The chemical states of Fe in BTZF are analyzed by X-ray photoelectron spectroscopy. The ferroelectric property is explained by the nano size effect, 1D nanostructure shape, lattice distortion and epitaxial strain between two phases. The magnetoelectric coefficient is measured at room temperature under an applied dc magnetizing field and show different types of behavior in each sample. The magnetocapacitance is measured and explained on the basis of Maxwell–Wagner space charge and magnetoresistance, and relates to theoretical investigation, which proves that the enhancement not only depends on the size/shape of nanostructure but also the strain-induced phase transition where out-of-plane polarization appears in the composite. The photoemission of BTZF is observed by fluorescence spectroscopy.

---

1. Introduction

Multiferroic (MF) magnetoelectric (ME) nanostructural composites aimed at understanding new physics – such as product properties, dynamic coupling between electric and magnetic dipole moments,¹ and microwave damping² – have attracted significance for use in magnetoelectronic and spintronic devices.^3–5 The ME coupling in MF composites is strain mediated from electrical order in one phase to magnetic order in the other, creating the desired ME phenomena.⁶ The spin-current-based multiferroicity of an MF composite is an interfacing ME coupling: interfacing ferromagnetic (FM) with ferroelectric (FE) triggers in the FM low-energy (coherent magnonic) excitations near the interface, which builds up a spatially inhomogeneous, spin-current-carrying surface magnetic order within the spin-diffusion length (on the order of nm).⁷ These ME couplings contribute as: (i) quasistatic screening spins, which follow adiabatically the direction of local magnetic moments, yielding a linear ME effect; (ii) deviations from adiabaticity give rise to an exponential spiral spin density which goes along with an emergent electric response associated with the intrinsic spin current of the spiral.⁸ In a comparison of an ME nanostructural composite with its bulk form, the FE and FM phases could be tuned and controlled at room temperature.⁹ In these composite, the two different phases can be combined at the atomic level, and thus interface losses could be reduced significantly, which results in an epitaxial or superlattice structure resides stress/strain.¹⁰

The ME coupling manifests itself by electric properties on an applied magnetic field, called direct ME effect, and is usually quantified in terms of the polarization and voltage coefficients. This polarization defines the change δP by an applied magnetic field δH as the magnetic-field control of electric polarization, which is termed the dynamic ME coefficient: α_E = ∂P/∂H.¹¹ The associated output voltage δV induced by δH affects the dielectric properties of ME as well; additional important information on the degree of ME coupling is provided by the magnetodielectric coefficients. This can be easily evaluated by measuring the dependence of capacitance C on magnetic-field intensity is so called magnetocapacitance (MC).^11,12 The MC effect can be understood on the basis of Maxwell–Wagner space charge effect at the interface of composite and electrode or grain boundaries,¹³ and magnetoresistance (MR).

Recently, perovskite based MF composites such as BaTiO₃–CoFe₂O₄,^3,13 BaTiO₃/Fe,⁶ ZnFe₂O₄–BaTiO₃,¹² BiFeO₃–CoFe₂O₄,¹⁴ CoFe₂O₄–PbTiO₃,¹⁵ LaCaMnO₃/BiFeO₃,¹⁶ Pb(Zr_0.52Ti_0.48)O₃–La_0.65Sr_0.35MnO₃,¹⁷ FeGaB/Pb(La,Sn,Zr,Ti)O₃,¹⁸ BaTiO₃–Ni_0.7Zn_0.3Fe₂O₄,¹⁹ Ba_0.8Ca_0.2TiO₃–Ni_0.2Cu_0.3Zn_0.5Fe₂O₄ (ref. 20) etc. have been widely investigated. Among them, the perovskite BaTiO₃ (BT) is a well-known²¹ room-temperature ferroelectric that is tetragonal with a = b = 3.99 Å and c = 4.03 Å. At higher temperatures it is cubic and thus not ferroelectric, but on cooling it becomes tetragonal below the Curie temperature of 410 K,²² orthorhombic below 290 K and rhombohedral below 190 K. It is given in a recent report²³ that the polycrystalline tetragonal BT nanostructure has enhanced polarization due to its twin structure in which stress residing near grain boundaries can easily control depolarization field and long range interactions. The nanostructural ZnFe₂O₄ (ZF) involves the redistribution of Fe³⁺ to tetrahedral and Zn²⁺ to octahedral interstices. This cationic rearrangement leads to the formation of two (A) and (B) magnetic sublattices, which are then responsible for the enhanced magnetization displayed when compared with bulk ZF.^24,25 In order to obtain a high ME response in MF composites, the ferrite phase should be highly magnetostrictive possessing high resistivity, which is possible in the ZF nanostructure.

In comparison with nanoparticles and thin films, one dimensional (1D) nanostructures (nanorods, nanowires etc.) have special physical properties.²⁶ This is due to their anisotropic property and unique size effect. Glinchuk et al.²⁷ pointed out that ferroic nanorods or nanowires will show giant ME effects as their radii decrease, and then induce an abnormal increase in the dielectric tunability and dramatic phase transition. Moreover, M. Q. Cai et al.²⁸ used first-principles calculations for the enhancement of spontaneous polarization as the polarization is induced along the length direction of the wire, by which the tension induced by the nanowire surface curvature counteracts the near-surface depolarization effect.

In this paper we have synthesized MF composites xBaTiO₃–(1 − x)ZnFe₂O₄ (BTZF) [x = 0.25 (BTZF2575), 0.35 (BTZF3565), 0.45 (BTZF4555), 0.50 (BTZF5050) and 0.75 (BTZF7525)] nanostructures by a sol–gel method using poly vinyl alcohol (PVA) as surfactant. The structural, microstructural, ferroelectric, ferromagnetic, ME coefficient, magnetodielectric and fluorescence properties have been investigated.

2. Experimental details

2.1 Preparation of BTZF nanostructures

For the synthesis of BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 nanostructures, a typical simple chemical process, i.e., sol–gel using PVA as surfactant was used. The precursor solution was prepared from barium acetate, tetra-n-butyl orthotitanate, zinc acetate and ferric chloride with desired molar concentrations. In this method, ethanol and acetic acid were mixed in the ratio of 3 : 1. The tetra-n-butyl orthotitanate was added to it and stirred for 2 h using a magnetic stirrer and denoted as solution A. Barium acetate, zinc acetate and ferric chloride were dissolved in distilled water separately to give a clear solution denoted as solution B. The final solution M was prepared by mixing A and B dropwise and stirring for another 3 h. This precursor solution M was added to a PVA solution in the molar ratio of M : PVA, 5 : 2. The solution was dried at 250 °C and annealed at 700 °C for 2 h to crystallize.

2.2 Characterization

Atomic absorption spectroscopy analysis was used to check the composition of Ba, Ti, Zn and Fe content in BTZF using a Perkin-Elmer system and both experimental and theoretical values of composition are well matched in each FE BT and FM ZF composite. The crystalline structure was analyzed by X-ray diffraction (XRD) by using an X’Pert PRO PANalytical system and microstructure by transmission electron microscopy (TEM) using a HITACHI H-7500. For electrical measurements, the BTZF crystalline powder was pressed into pellets of thickness ∼0.5 mm by the cold isostatic pressing method with a pressure of 5 bar for 10 min and then sintered at 1000 °C for 5 h. Polarization under the influence of an applied electric field was measured using a Radiant Technologies FE test system. Magnetization of BTZF was measured at room temperature using a Lakeshore 7304 vibrating sample magnetometer. X-ray photoelectron spectroscopy (XPS) was performed using a Perkin-Elmer model 1257 equipped with a non-monochromatized Al Kα X-ray source. The ME coupling coefficient was measured by the dynamic field method used a system built in-house. A buffer and high-pass filter circuit to reduce the background voltage noise were employed in the measurement set-up to detect the induced ME voltage across the sample thickness. The room temperature magnetodielectric performed on BTZF samples at frequency of 1 kHz by change in relative dielectric permittivity as a function of an applied magnetic field strength was measured by combining a Wayne Kerr 6500B Precision Impedance analyzer with the Lakeshore 7304. The photoemission is shown by fluorescence spectrum of BTZF was recorded using an Edinburgh FLSP920 spectrofluorophotometer at room temperature.

3. Result and discussion

Fig. 1 shows the XRD patterns of all samples of BTZF composites measured at room temperature. The pattern confirms the existence of two phases, namely perovskite and spinel together corresponding to BT and ZF, respectively. The polycrystalline BT with tetragonal phase at diffraction angles 2θ = 21.74, 22.66, 31.71, 39.27, 44.73, 51.75 and 56.01 correspond to (001), (100), (101), (111), (200), (102/201) and (112/211) respectively, observed. The reflections 001/100 and 102/201 are the splitting of cubic into tetragonal phase. Meanwhile, 2θ = 29.69, 35.27, 36.84, 42.78, 48.73, 53.45 and 58.67° correspond to (220), (311), (222), (400), (331), (422) and (511), respectively, indicates the cubic spinel phase of ZF. Using the CHEKCELL lattice constant refinement program, the lattice parameters of both phases are summarized in Table 1. The distortion ratio c/a of tetragonal BT is 1.0494, 1.0087, 1.0076, 1.0089 and 1.0097, respectively, calculated for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composite. The distortion, c/a when compared with pure BT (c/a ∼ 1.01),²¹ 75%BT in BTZF has slightly smaller value of c/a and decreases with decreasing BT to 45%BT. The smaller c/a value of the BT phase in the composite than in the pure BT ceramic suggests the diffusion of Fe³⁺ or Zn²⁺ into the BT lattice from the ZF phase. BaTiO₃ is known to enable substitution of different sizes of ions including Fe³⁺ and Zn²⁺ and the doping/substitution site depends on the size of the ions.²⁹ The ionic radii of Fe³⁺ (0.69 Å) and Zn²⁺ (0.88 Å) are more comparable to the ionic radius of Ti⁴⁺ (0.745 Å) than Ba²⁺ (1.49 Å) and therefore, Fe³⁺ and Zn²⁺ substitution will occur at the Ti-site.³⁰ Meanwhile, substitution of Fe³⁺ can result in a decrease of BT lattice parameters because of its smaller radius than Ti⁴⁺.¹⁸ Zn²⁺ substitution, on the other hand, can increase the BT lattice parameters due to it having a larger ionic radius than Ti⁴⁺. Therefore, with 25%BT in BTZF, the large value of lattice distortion occurs because Zn²⁺ is reach and has a larger ionic radius than both Fe³⁺ and Ti⁴⁺. Moreover, substitution of Zn²⁺ and Fe³⁺ at Ti⁴⁺ sites will cause the formation of oxygen vacancies to maintain the electroneutrality³⁰ and these oxygen vacancies contribute to the increase of the lattice parameters.

Fig. 1 XRD patterns of xBaTiO₃–(1 − x)ZnFe₂O₄ (BTZF) [x = 0.25 (BTZF2575), 0.35 (BTZF3565), 0.45 (BTZF4555), 0.50 (BTZF5050) and 0.75 (BTZF7525)] nanostructure composites measured at room temperature.

Table 1 Values of lattice parameters a and c of BT and a of ZF, and diameter (d), length (l), saturation magnetization (M_s), remanent magnetization (M_r), magnetic coercive field (H_c), spontaneous polarization (P_s), remanent polarization (P_r), electric coercive field (E_c), magnetoelectric coefficient (α_E) at 3 kOe and magnetocapacitance (MC) at 3 kOe of BTZF nanostructures

BTZF	Lattice constant			Multiferroic properties
	BT		ZF	BTZF
	a (Å)	c (Å)	a (Å)	d (nm)	l (nm)	Ms (emu g^−1)	Mr (emu g^−1)	Hc (Oe)	Ps (μC cm^−2)	Pr (μC cm^−2)	Ec (kV cm^−1)	αE (mV cm^−1 Oe^−1)	MC (%)
2575	4.0013	4.1992	8.448	6	>100	0.19	0.056	1068	69.61	18.44	13.79	122.9	2.84
3565	3.992	4.0274	8.438	33	103	15.41	3.19	319	41.95	5.47	1.78	30.5	1.24
4555	3.992	4.0222	8.432	54	120	8.71	1.38	255	26.31	5.36	3.61	29.2	0.78
5050	3.995	4.0304	8.422	59	141	0.15	0.038	249	31.66	6.41	12.81	51.7	1.42
7525	3.999	4.0378	8.351	84	—	0.25	0.047	116	26.10	12.51	12.93	43.9	1.12

---

Fig. 2(a and b) show the TEM images (left) and HRTEM images (right) of BTZF nanostructural composites measured at different magnifications because the shape and size of nanostructure (nanoparticles, nanowires and nanorods) in each sample is quite different hence creates difficulty in visualizing clear pictures of the shape and size at same magnification for each sample. The BTZF2575 composite forms nanowire-like structures of average diameter (d) 6 nm and length of greater than 100 nm. But with increasing BT concentration the dimensions along the long side decrease and the diameter shows an increment, i.e., the formation of nanorod-like structures. The average value of d is 33, 54 and 59, and l is 103, 120 and 141 nm, respectively, measured for BTZF3565, BTZF4555 and BTZF5050 nanorods. However, the BTZF7525 composite forms nanoparticles of cubic shape and the average value of d is 84 nm. The mechanism of the formation of nanoparticles, nanorods and nanowires is explained in the next paragraph. It has been also observed in the TEM images that there are random distribution of dark and bright borders, which may be the formation of FE BT and FM ZF phases alternatively into a single nanocrystal, as is clearly shown by the HRTEM images in the right panels of Fig. 2(a and b). Large regions of periodically ordered material are no longer observed, indicating random filling of phases. This type of behavior imagining the two phases in a single crystal is described recently by Sreenivasulu et al.³¹ The planes with interplanar spacing d ∼ 0.28 nm of the most intense peak (101) of BT in the XRD pattern is observed in all HRTEM images and d ∼ 0.25 of the most intense peak (311) of ZF is also seen. In BTZF2575, the most frequent lattice fringes are oriented parallel to the major axis of the nanowires. The high crystallinity of the particles is evident from the selected area electron diffraction (SAED) patterns shown in the insets of the HRTEM images and show that all the composites are of polycrystalline phase.

Fig. 2 (a) TEM images (left) and HRTEM images (right) of BTZF2575, BTZF3565 and BTZF4555 samples, respectively. Insets of HRTEM images are the SAED pattern of each respective sample. (b) TEM images (left) and HRTEM images (right) of BTZF5050 and BTZF7525 samples, respectively. Insets of HRTEM images are the SAED patterns of each respective sample.

The formation of different types of nanostructural (nanoparticles, nanorods and nanowires) formations is explained on the basis of ionic radii, surface energy and PVA acting as surfactant, as well as orientation during the growth process of large aggregates of metal ions. The variation in the size of grains is due to different ionic radii of metal ions and their stoichiometric ratio is different in each BTZF composite. The variation in the shape of BTZF nanostructures is explained on the basis of the surface energy of Zn²⁺ ions, which is higher and forms agglomerations during chemical reaction with OH⁻ ions.³² The Zn²⁺ ions in OH⁻ medium normally take the form of divalent metal ions and their solubility increases with appropriate concentration and temperature. During heating, the surfactant PVA helps in synthesizing nanostructures. Under heat treatment of PVA, there exist two sites, namely hydrophobic (Hb) and hydrophilic (Hp), of PVA and they form clusters. The Hb sites keep the materials separate due to their repulsive forces which slow down the rate of reaction, whereas Hp sites keep the metal intact and retard the growth process. Therefore, higher concentrations of Zn²⁺ ions form large agglomerations with OH⁻ and when this product reacts with PVA, the Hp site gives the orientation to grow continuously in the confined reverse micelle process, causing formation of nanorods and nanowires. Moreover, the Ostwald ripening theory³² is also responsible for this type of crystal growth process.

Fig. 3(a–e) shows the FM behavior of BTZF composites by measuring magnetization versus applied magnetizing field (M–H_dc) hysteresis at room temperature. The values of saturation magnetization (M_s) are 0.19, 15.41, 8.71, 0.15 and 0.25 emu g⁻¹, values of remanent magnetization (M_r) are 0.056, 3.19, 1.38, 0.038 and 0.047 emu g⁻¹, and values of coercive field (H_c) are 1068, 319, 255, 249 and 116 Oe, measured for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composites, respectively, and are also given in Table 1. These values of ferromagnetism showed an improvement over reported work on BT based MF composites.^{12,19,20,33–35} The variation of observed ferromagnetism with BT concentration in BTZF is shown in Fig. 3(f). The value of magnetization is slightly varied with 25, 50 and 75% of ZF concentration in BTZF but increases abruptly with 55 and 65% of ZF. This type of behavior is explained on the basis of size and shape of nanostructure, and stoichiometric ratio of the MF BTZF composite. The slight variation in ferromagnetism is the general behavior and depends upon the size of nanograins.³² The shape of nanostructure is also responsible for improvement in M_s since the nanostructure is increased in the strength of magnetic exchange interaction by surface spins contributed by their large surface to volume ratio and involves a similar behavior given by Callsen et al.³⁶ for nonmagnetic organic molecules on the FM surface using theoretical first principle calculations. In the BTZF nanostructures, the nonmagnetic nano boundaries between nanograins are present instead of organic molecules. It is demonstrated that the possibility of a hybrid nano boundary–FM interface acts as a local embedded surface spin based magnetic unit due to a magnetic hardening of BT in the composite and exhibits spin-filter functionality with sharp spin split molecular-like electronic features at the nano boundary site. This surface spin induced local magnetic hardening leads to exchange interactions and therefore a larger magnetic anisotropic energy results in a stable magnetization axis. This type of magnetism by surfaces and interfaces of nonmagnetic nano boundaries is also described by Neeleshwar et al.³⁷ for Ge quantum dots. As discussed above, it is concluded that the orbital momentum induced at the surface conduction electrons is crucial to understand the observed giant anisotropy.

Fig. 3 (a–e) Magnetization (M) versus magnetizing field (H_dc) of BTZF nanostructures (insets shows the polarization (P) versus electric field (E) of the respective samples). (f) Plot of M and remanent polarization, P_r with varying the concentration of BT in BTZF composite.

The abrupt increment of M_s in BTZF3565 and BTZF4555 (Fig. 3(f)) composites is also seen in various ferrites and MF systems based on ZF.^25,38 Generally in ZF, the magnetic properties are strongly dependent on the strength of occupation of various cations at octahedral and tetrahedral sites. It is well known that in nanostructural ZF, the antiferromagnetic ordering temperature is much higher (T_N = 450 K)²⁵ and can be raised by increasing the Fe³⁺ occupation at the tetrahedral sites through mechanical activation. This increase of magnetization originates in the stronger inter sub-lattice (A–B) superexchange interaction as compared to intra sub-lattice (A–A and B–B) interactions. At lower ZF concentration, A–A and B–B interactions are dominant. But with higher concentrations, i.e., 50–60% ZF in BTZF, Fe³⁺ occupies Ti⁴⁺ B sites of perovskite via oxygen vacancy and hence stronger superexchange A–B interactions occur, which results in an abrupt increment in magnetization. With sufficient higher concentration, 75% of ZF, weakening of the A–B exchange interaction occurs and the parallel orientation of the magnetic moments in B-site is altered so that they compensate each other only partially, which leads to a progressive decrease of saturation magnetization.

In order to find the contribution of Fe on ferromagnetism of BTZF composites, the XPS spectra are given in Fig. 4, measured at room temperature for Fe 2p states in the binding energy region 700–727 eV. It was reported³⁹ that Fe 2p photoelectron peaks from oxidized iron are associated with satellite peaks, which is important for identifying the chemical states. The Fe²⁺ and Fe³⁺ 2p_3/2 peaks always show satellite peaks at 6 eV and 8 eV above the principal peaks at 709.5 eV and 711.2 eV, respectively.³⁹ In Fig. 4, the satellite peaks were found in the energy region of 6–8 eV above the 2p_3/2 principal peak. It indicates that in this BTZF MF system Fe coexists in both Fe²⁺ and Fe³⁺ states. When Fe³⁺ ions coexist with Fe²⁺ in ZF, the Fe³⁺ distorts the lattice structure more than the Fe²⁺ ion due to the effects of charge and ionic size.⁴⁰ The difference in energy between the Fe 2p_3/2 and 2p_1/2 peaks resulting from spin–orbit coupling is ∼13.5 eV, which is very close to the reported value of Fe 2p.³⁹ The peaks related to 2p_3/2 are 710.16, 709.83, 710.07, 710.05 and 710.16 eV, and 2p_1/2 are 723.67, 723.34, 723.59, 723.37 and 723.92 eV, respectively, observed for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composites. These results suggest that the Fe ions have mixed valence state of 2+ and 3+ and must be contributing ferromagnetism via oxygen vacancies.

Fig. 4 XPS spectra of Fe 2p of BTZF nanostructures measured at room temperature.

The insets of Fig. 3(a–e) show the polarization versus electric field (P–E) hysteresis curves at room temperature on poled samples of BTZF composites with 50 Hz being the frequency of polarization. The values of spontaneous polarization, P_s = 69.61, 41.95, 26.31, 31.66 and 26.18 μC cm⁻², remanent polarization, P_r = 18.44, 5.47, 5.36, 6.41 and 12.51 μC cm⁻², and coercive field, E_c = 13.79, 1.78, 3.61, 12.81 and 12.93 kV cm⁻¹, respectively, measured for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composite and are also given in Table 1. These values of polarization show improvement over reported work on BT based MF composites.^{12,20,33,41,13,17} The values of P_r are plotted in Fig. 3(f) with varying BT concentration and show an abrupt decrement for BT > 0.25 and again enhanced for BT ≥ 0.5. This improvement and increment in the value of P_r is dependent upon different factors, i.e., nanosize, 1D nanostructural formation, lattice distortion and epitaxial strain^23,28 and the resulting mechanism is described by the following arguments: the suggestion by Lu et al.⁴² also applies to the BTZF nanostructural composite which is associated with multidomains, the ZF certainly induced modifications of the FE properties of BT, which might change the depolarization field including change in interfacial strain via magnetostriction effect. However, the stability of ferroelectricity in nanoscale BT is proposed by Spanier et al.⁴³ on the basis of polarization-induced surface charges. Incomplete screening of surface charges by nanostructures results in a depolarization field that opposes the bulk polarization, thereby suppressing ferroelectricity. When the nanostructural formation occurs in the forms of rods and wire, the capacitive contribution is eliminated between the tip and the length so that the resulting map exhibits only the contribution from the surface charges associated with a local electric polarization. Moreover, the BTZF nanostructure is synthesized by a sol–gel based chemical process where there may be the possibility of the presence of OH ions on the surfaces. These molecular adsorbate OH ions can compensate the surface polarization charges, providing a mechanism for reducing the depolarizing fields.⁴³ In the BTZF nanostructure exist both Fe²⁺/Fe³⁺ ions (Fig. 4), which creates the possibility that each Fe²⁺ with its surrounding Fe³⁺ will form a dipole with strong local polarization. Under an external electric field, the hopping of electrons between Fe²⁺ and Fe³⁺, which is equivalent to the reversal of the dipole, contributes to the large polarization and involves a similar behavior observed in nano MF YFeO₃.⁴⁴ Furthermore, the jumping of electrons in Fe²⁺/Fe³⁺ dipoles is also a switching process of the dipole, which contributes to magnetocapacitance effect (Fig. 6) due to the jumping becoming easier in a magnetic field.⁴⁴

The ME coupling coefficient was determined as a function of dc magnetic field using the equation:¹¹

where t is the sample thickness and V_out is the induced ME voltage. The variations in α_E with H_dc under an ac magnetic field frequency of 1093 Hz and ac magnetizing field of 5 Oe for BTZF nanostructures is shown in Fig. 5. In these cases, the ME effect results from the interaction between different orderings of the two phases in the composite. This is an interface coupled electrical and magnetic ordering via elastic interaction between the piezoelectric and piezomagnetic phases.⁴⁵ There is an elastic coupling at the interface (randomly distributed BT and ZF): an applied magnetic field produces an elastic strain in the magnetostrictive phase (ZF), which is stress coupled to that of the piezoelectric (BT), resulting in an induced voltage. Moreover, Bary’achtar et al.⁴⁶ reported that an electric polarization appears in the vicinity of the inhomogeneous magnetic media due to the lowering of the magnetic symmetry group of the crystal. In this context, the absence of a symmetry centre in the near-interface should lead to an extra ME effect at the interfaces in the composite. The value of α_E at H_dc = 3 kOe is 122.9, 30.5, 29.2, 51.7 and 43.9 mV cm⁻¹ Oe⁻¹, respectively, measured for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composites. These values of α_E show improvement over reported work on BT based MFs composites,^{3,19,20,33,34,47,48} which depends upon: (i) the surface spins of nanostructure enhances magnetostriction constant (λ), λ_total = λ_bulk + λ_surf/t (ref. 49) due to the reduced symmetry of atoms on the surface. The large surfaces of nanorods and nanowires results in giant electric field tuning of the magnetic anisotropy, FM resonance field, coercive field and anisotropic magnetoresistance being demonstrated. (ii) In 1D nanostructures, long range interactions are truncated due to the lack of periodicity and short range interactions are significantly modified near the surface boundary.⁵⁰ The compressive stress induced by surface curvature of nanorods/wires would produce an effective tensile in the length direction, leading to a big off-center displacement which enhances induced ME voltage. (iii) BTZF nanostructures of highly distorted BT polycrystalline phase contribute a twin structure where the stress resides near grain boundaries, leading to enhance ME polarization.¹⁰ And, (iv) the epitaxial strain by lattice mismatch of FE and FM phases. The observed ME response varies non-linearly and does not have an abrupt decay of α_E by varying H_dc, which is explained by the mechanism given by A. Sukhov et al.⁶ in BT composites of tetragonal phase. It is suggested that this MF contact consists of a thin FM ZF layer and a thick FE BT. The thin FM layer is responsible for spin waves as well as to avoid a decay of the ME coupling in long FMs, whereas the thick FE part stabilizes the FE polarization. Because of the interfacial screening charges due to epitaxial growth with highly distorted BT tetragonal phase, the principal axis of ZF is rotated by 45°.⁶

Fig. 5 ME voltage coefficient (α_E) with H_dc (0–6 kOe) under the influence of H_ac = 5 Oe and frequency of 1093 Hz of BTZF nanostructures.

The origin of ME coupling of BTZF composites is the coupling of FE and FM degrees of freedom due to the influence of the FE matrix on the exchange coupling constant via screening of the intragrain and intergrain Coulomb interactions. Therefore, a possible mechanism for ME coupling of this FE/FM (BTZF) interface is based on screening effects.⁵¹ The spin-polarized charge density formed in the FM in the vicinity of the FM/FE interface⁵² acts with a torque on the magnetic moments in the FM, resulting in a noncollinear magnetic ordering.⁵³ Hence, electric polarization emerges that couples the FM to the FE part. Additionally, the FE polarization (and electric field) stems actually from the FE surface, which triggers the spin spiral in a FM. The latter carries a spin current with an associated Aharonov–Casher effect and/or Dzyaloshinkii–Moriya (DM) interaction.⁵⁴ In this sense the ME coupling is caused by an emergent inverse DM interaction at MF interfaces.⁷

Recently, MF composites have been reported with MC or magnetodielectric (MD) effects,^11,12 where for composites with inhomogeneous structures in the interfacial layers for different resistivity and other electric structures, the MR value could be calculated with a simple equivalent circuit of two parallel RC elements in series. MR depends upon core-based and the interfacial (nanograins boundaries) of the MF composite.¹² For an MF composite under an external magnetic field, the magnetostriction in the magnetic phase produces stresses transferred in the FE phase, resulting in an electric polarization via the ME effect.¹² As a result, the dielectric behavior is modified. The MC/MD is the change in dielectric constant (ε), [ε(H) − ε(0)]/ε(0), where ε(H) and ε(0) denote the dielectric constants at applied magnetic field H and zero field, respectively. Fig. 6 shows the change in the real part of the relative dielectric permittivity of the nanostructural BTZF composites at 1 kHz as a function of magnetic field H_dc and measured at room temperature. Clearly, the capacitance decreases with increasing applied magnetic field (negative value of MC) and a similar behavior is also observed by Anjum et al.⁵⁵ for their MF system. The value of MC at H_dc = 3 kOe is 2.84, 1.24, 0.78, 1.42 and 1.12%, respectively, calculated for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 composites.

Fig. 6 MD responses of the BTZF nanostructures under the influence of H_dc (0–6 kOe) at 1 kHz of frequency.

Pertsev et al.¹¹ gives theoretical calculations of strain-mediated MC of BaTiO₃ film deposited on FM substrate showing a similar behavior to that observed in the present BTZF nanostructure where ME coupling is mediated via lattice strain. It has been observed for BT films that the lattice mismatch S^ε_αα(u_m = −1%) ≅ 43 and S^ε_αα = (u_m = +1%) ≅ −58, where S^ε_αβ = S^ε_αβ(u_m) of the out-of-plane permittivity ε₃₃ as a function of the initial biaxial strain u_m. In the present case, BT has tetragonal phase with higher tetragonal distortion and ZF shows cubic phase. Therefore, these values of S^ε_αα may be improved for different types of nanostructural formations of BTZF.⁵⁶

For this strain-mediated ME effect in a MF composite, the coefficient can be calculated as

where u_αβ (α, β = 1, 2) are the in-plane composite strains and deformation, respectively, and perfect mechanical coupling at the FE/FM (BT/ZF) interface is assumed (δ^u_αβ = δ^S_αβ). Since the deformation response of a FM material is weak in measuring magnetic field δH which becomes linear under sufficient bias field H, the second term in (2) can be replaced by the interface field-dependent piezomagnetic coefficients d^m_jαβ = ∂u^S_αβ/∂H_j. It should be emphasized that the local deformation response of the interface region beneath the BT nanostructure generally depends upon the geometry of interface.

Indeed, when the magnetic field is not parallel to the composite interface, the demagnetising field eventually makes d^m_3αβ smaller than the piezomagnetic coupling where the internal magnetic field is equal to the applied one. Consider BTZF as a parallel plate capacitor, where only the out-of plane composite permittivity ε₃₃ is important. Introducing the strain sensitivities S^ε_αβ of the permittivity ε₃₃ via the relation δ^ε_αβ = ∂ln ε₃₃/∂u_αβ, the resulting product is obtained as:

This eqn (3) is valid for weak fields δH ≪ H that do not change the initial piezomagnetic coefficients and strain sensitivity significantly. Therefore, sensitivities S^ε_αβ can be calculated using the non-linear thermodynamic theory^57,58 by neglecting the in-plane lattice strain.¹¹ When the composite stabilizes in the tetragonal c phase (P₁ = P₂ = 0, P₃ ≠ 0) the strain-mediated MC appears to be significant and there is no direct proportionality between the MC and the ME voltage coefficient. Therefore, it gives a contrast between the MC and ME voltage coefficient which may become negligible at P₃ = 0 in the conventional parallel plate capacitor geometry of BTZF nanostructures where BT and ZF are distributed randomly, separated by a buffer layer of nano grain boundaries.

In the case of aa phase (|P₁| = |P₂|, P₃ = 0) forming an isotropic biaxial strain (u₁₁ = u₂₂ = u_m, u₁₂ = 0), the measuring magnetic field δH does not induce shear strain u₁₂ in the composite⁹ and is sufficient to calculate the strain mediated MC using (3).

Therefore, to evaluate the MC of an ME composite, we need to know only the longitudinal piezomagnetic coefficient characterizing the deformation u^S₁₁ induced along the field direction and the transverse coefficient governing the deformation u^S_⊥ in the direction orthogonal to δH. Considering d^m_‖ and d^m_⊥, we obtain

where φ is the angle between the magnetic field δH and the [100] crystallographic axis of BT. The theoretically calculated MC/δH of the BT interface is ∼10⁻⁴ Oe⁻¹. The measured values of MC from Fig. 6 given in Table 1 are consistent with the above theoretical observation. This is the giant MD effect that has potential applications in advanced electronic devices.

Fig. 7 shows the fluorescence spectra of BTZF nanostructures measured at room temperature. Absorption bands corresponding to green centred at 555.48 and 563.33 nm, and red at 646.42 and 674.84 nm are observed. The green luminescence is distinctly enhanced in BTZF 2575, 7525 and 5050 nanostructures. However, the emission intensity is enhanced for BTZF7525. The generation of oxygen vacancies in the lattice lead to the shrinking of the lattice.^59,60 The distortion of the local asymmetry along B-sites of BT increases, and the transition probabilities that govern various intra-shell transitions giving rise to both green and red emission are increased.⁵⁹ As a whole, the enhanced luminescence is conjectured to result from the distortion of the local asymmetry and oxygen vacancies generated by Fe³⁺ ion incorporation in the Ti⁴⁺ lattices.⁶⁰ The absorption bands, i.e., 279.57, 340.78 and 352.58 nm lie in the UV emission region, which is related to near band emission attributed to free-exciton recombination.⁶¹

Fig. 7 Fluorescence spectra of BTZF 2575 (black), 3565 (red), 4555 (dark cyan), 5050 (violet) and 7525 (pink) nanostructure measured at room temperature.

4. Conclusion

The MF BTZF composites were synthesized by a sol–gel process at 700 °C 2 h⁻¹ using PVA as surfactant, resulting in nanorods, nanowires and nanoparticle type nanostructural formations, which have enhanced ME coefficient and MC response. The XRD results show coexistence of polycrystalline phases, i.e., tetragonal and spinel, respectively, for BT and ZF. The tetragonal distortion of the BT phase depends upon the size and concentration of composite ions, and oxygen vacancies. The nanostructures of BTZF2575 form nanowires while BTZF3565, BTZF4555 and BTZF5050 are nanorods and BTZF7525 form nanoparticles. The mechanism in the formation of nanoparticles, nanorods and nanowires is explained on the basis of ionic radii, surface energy, PVA and Ostwald ripening process. The XPS analysis results show mixed valence state of 2+ and 3+ of Fe in each BTZF sample. The values of M_s = 0.19, 15.41, 8.71, 0.15 and 0.25 emu g⁻¹, P_s = 69.61, 41.95, 26.31, 31.66 and 26.18 μC cm⁻², α_E (at H_dc = 3 kOe) = 122.9, 30.5, 29.2, 51.7 and 43.9 mV cm⁻¹ Oe⁻¹, and MC (at 1 kHz frequency and H_dc = 3 kOe) = 2.84, 1.24, 0.78, 1.42 and 1.12% were measured for BTZF2575, BTZF3565, BTZF4555, BTZF5050 and BTZF7525 nanostructural composites, respectively. The enhancement in ME and MD is possible by strain-mediated magnetocapacitance and exists on the basis of nonlinear thermodynamic theory can be exploited in magnetic-field-controlled ferroelectric capacitors, namely magnetovaractors implemented in electronic oscillators, where the oscillation frequency is tuned by a dc magnetic field remotely (via the change of capacitance) in contrast to conventional voltage-controlled oscillators.

References

1. W. Eerenstein, M. Wiora, J. L. Prieto, J. J. Scott and N. D. Mathur, Nat. Mater., 2007, 6, 348.
2. C. Brosseau, V. Castel and M. Potel, J. Appl. Phys., 2010, 108, 024306.
3. Y. Shen, J. Sun, L. Li, Y. Yao, C. Zhou, R. Su and Y. Yang, J. Mater. Chem. C, 2014, 2, 2545.
4. J. Wu and J. Wang, J. Appl. Phys., 2009, 105, 124107.
5. J. Wu and J. Wang, Electrochem. Solid-State Lett., 2009, 12(10), G61.
6. A. Sukhov, P. P. Horley, C. L. Jia and J. Berakdar, J. Appl. Phys., 2013, 113, 013908.
7. C. L. Jia, T. L. Wei, C. J. Jiang, D. S. Xue, A. Sukhov and J. Berakdar, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 90, 054423.
8. M. Mostovoy, Phys. Rev. Lett., 2006, 96, 067601.
9. Y. Wang, J. Hu, Y. Lin and C. W. Nan, NPG Asia Mater., 2010, 2(2), 61.
10. M. H. Frey and D. A. Payne, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 3158.
11. N. A. Pertsev, S. Prokhorenko and B. Dkhil, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 134111.
12. M. Lorenz, M. Ziese, G. Wagner, J. Lenzner, C. Kranert, K. Brachwitz and M. Grundmann, CrystEngComm, 2012, 14, 6477.
13. J. S. Andrew, J. D. Starr and M. A. K. Budi, Scr. Mater., 2014, 74, 38.
14. F. Yan, G. Chen, L. Lu, P. Finkel and J. E. Spanier, Appl. Phys. Lett., 2013, 103, 024906.
15. C. Y. Tsai, H. R. Chen, F. C. Chang, H. H. Kuo, H. M. Cheng, W. C. Tsai, Y. H. Chu, C. H. Lai and W. F. Hsieh, J. Appl. Phys., 2014, 115, 134317.
16. D. Yi, J. Liu, S. Okamoto, S. Jagannatha, Y. C. Chen, P. Yu, Y. H. Chu, E. Arenholz and R. Ramesh, Phys. Rev. Lett., 2013, 111, 127601.
17. X. Lv, C. Cheng, Y. Yiao, M. Tang, Z. Tang, H. Cai, Y. Zhou and R. Li, Mater. Lett., 2013, 100, 7.
18. Z. Zhou, X. Y. Zhang, T. F. Xie, T. X. Nan, Y. Gao, X. Yang, X. J. Sun and D. Z. Sun, Appl. Phys. Lett., 2014, 104, 012905.
19. R. F. Zhang, C. Y. Deng, L. Ren, Z. Li and J. P. Zhou, Mater. Res. Bull., 2013, 48, 4100.
20. K. Sadhana, S. R. Murthy, S. Jie, Y. Xie, Y. Liu, Q. Zhan and R. W. Li, J. Appl. Phys., 2013, 113, 17C731.
21. H. F. Kay and P. Vousden, Philos. Mag., 1949, 40, 1019.
22. A. M. Bratkovsky and A. P. Levanyuk, Phys. Rev. Lett., 2005, 94, 107601.
23. K. C. Verma, J. Kaur, N. S. Negi and R. K. Kotnala, Solid State Commun., 2014, 178, 11.
24. K. C. Verma, V. P. Singh, M. Ram, J. Shah and R. K. Kotnala, J. Magn. Magn. Mater., 2011, 323, 3271.
25. M. Atif, S. K. Hasanain and M. Nadeem, Solid State Commun., 2006, 138, 416.
26. B. Liu, B. Hu and Z. Du, Chem. Comm., 2011, 47, 8166.
27. M. Glinchuk, E. Eliseev, A. Morozovska and R. Blinc, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 024106.
28. M. Q. Cai, Y. Zheng, B. Wang and G. W. Yang, Appl. Phys. Lett., 2009, 95, 232901.
29. D. Ghosh, H. Han, J. C. Nino, G. Subhash and J. L. Jones, J. Am. Ceram. Soc., 2012, 95, 2504.
30. M. T. Buscaglia, V. Buscaglia, M. Viviani and P. Nanni, J. Am. Ceram. Soc., 2001, 84(2), 376.
31. G. Sreenivasulu, M. Popov, F. A. Chavez, S. L. Hamilton, P. R. Lehto and G. Srinivasan, Appl. Phys. Lett., 2014, 104, 052901.
32. R. K. Kotnala, V. Gupta and K. C. Verma, Curr. Appl. Phys., 2014, 14, 749.
33. G. Sreenivasulu, M. Popov, F. A. Chavez, S. L. Hamilton, P. R. Lehto and G. Srinivasan, Appl. Phys. Lett., 2014, 104, 052901.
34. M. Lorenz, V. Lazenka, P. Schwinkendorf, F. Bern, M. Ziese, H. Modarresi, A. Volodin, M. J. V. Bael, K. Temst, A. Vantomme and M. Grundmann, J. Phys. D: Appl. Phys., 2014, 47, 135303.
35. K. C. Verma, V. Gupta, J. Kaur and R. K. Kotnala, J. Alloys Compd., 2013, 578, 5.
36. M. Callsen, V. Caciuc, N. Kiselev, N. Atodiresei and S. Blugel, Phys. Rev. Lett., 2013, 111, 106805.
37. S. Neeleshwar, C. L. Chen, C. B. Tsai, Y. Y. Chen and C. C. Chen, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 71, 201307.
38. S. K. Upadhyay and V. R. Reddy, J. Appl. Phys., 2013, 113, 114107.
39. J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben, Handbook of X-ray Photoelectron Spectroscopy Physical Electronics, reissue edn., 1995.
40. S. Y. Seo, C. H. Kwak, Y. B. Lee and S. H. Kim, J. Korean Phys. Soc., 2008, 52, 805.
41. C. X. Li, B. Yang, S. T. Zhang, D. Q. Liu, R. Zhang, Y. Sun and W. W. Cao, J. Alloys Compd., 2014, 590, 346.
42. X. Lu, Y. Kim, S. Goetze, X. Li, S. Dong, P. Werner, M. Alexe and D. Hesse, Nano Lett., 2011, 11, 3202.
43. J. E. Spanier, A. M. Kolpak, J. J. Urban, I. Grinberg, L. Ouyang, W. S. Yun, A. M. Rappe and H. Park, Nano Lett., 2006, 6, 735.
44. Z. X. Cheng, H. Shen, J. Y. Xu, P. Liu, S. J. Zhang, J. L. Wang, X. L. Wang and S. X. Dou, J. Appl. Phys., 2012, 111, 034103.
45. J. Ma, J. Hu, Z. Li and C. W. Nan, Adv. Mater., 2011, 23, 1062.
46. V. G. Bary’achtar, V. A. L’vov and D. A. Jablonskii, JETP Lett., 1983, 37, 673.
47. E. V. Ramana, F. Fiqueiras, M. P. F. Graca and M. A. Valente, Dalton Trans., 2014, 43, 9934.
48. R. C. Kambale, K. M. Song and N. Hur, Curr. Appl. Phys., 2013, 13, 562.
49. M. Liu, O. J. Lou, S. Li, X. Xing, G. Yang and N. X. Sun, J. Appl. Phys., 2011, 109, 07D913.
50. E. A. Eliseev, A. N. Morozovska, M. D. Glinchuk and R. Blinc, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 165433.
51. O. G. Udalov, N. M. Chtchelkatchev and I. S. Beloborodov, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 89, 174203.
52. L. Chotorlishvili, R. Khomeriki, A. Sukhov, S. Ruffo and J. Berakdar, Phys. Rev. Lett., 2013, 111, 117202.
53. N. Sedlmayr, V. K. Dugaev and J. Berakdar, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 174422.
54. H. Katsura, N. Nagaosa and A. V. Balatsky, Phys. Rev. Lett., 2005, 95, 057205.
55. G. Anjum, S. Mollah, D. K. Shukla and R. Kumar, Mater. Lett., 2010, 64, 2003.
56. H. Zheng, J. Wang, S. E. Lofland, Z. Ma, L. M. Ardabili, T. Zhao, L. S. Riba, S. R. Shinde, S. B. Ogale, F. Bai, D. Viehland, Y. Jia, D. G. Schlom, M. Wuttig, A. Roytburd and R. Ramesh, Science, 2004, 303, 661.
57. N. A. Pertsev, A. G. Zembilgotov and A. K. Tagantsev, Phys. Rev. Lett., 1998, 80, 1988.
58. A. G. Zembilgotov, N. A. Pertsev, U. Bottger and R. Waser, Appl. Phys. Lett., 2005, 86, 052903.
59. X. Chen, Z. Liu, Q. Sun, M. Ye and F. Wang, Opt. Commun., 2011, 284, 2046.
60. Q. Sun, X. Chen, Z. Liu, F. Wang, Z. Jiang and C. Wang, J. Alloys Compd., 2011, 509, 5336.
61. X. Li, Y. Hou, Q. Zhao and L. Wang, J. Colloid Interface Sci., 2011, 358, 102.

---