Synthesis and structural characterization of Zn_xFe_3−xO₄ ferrite nanoparticles obtained by an electrochemical method

Abstract

A series of zinc ferrite nanoparticles were synthesized following a single-step electrochemical method in aqueous medium. This strategy allowed the control of both the size and chemical composition of the nanoparticles in an easy and reproducible manner by simply varying the intensity of the applied current. The obtained nanoparticles were morphologically and structurally characterized as a function of the particle size and the Zn content in the sample by X-ray diffraction (XRD), transmission electron microscopy (TEM), inductively coupled plasma emission spectroscopy (ICP) and Raman microscopy. The formation of Zn_xFe_(3−x)O₄ (x = 0.18–0.93) ferrite nanoparticles with crystal sizes in the range of 9 to 18 nm and with a homogeneous distribution of the Zn²⁺ cation in the crystalline structure was observed. However, following a thermal treatment, a migration of zinc cations was detected that led to the formation of two different crystalline phases, stoichiometric zinc ferrite and hematite. Raman microscopy revealed the formation of segregated micro-domains enriched within these crystalline phases. The study of the magnetic properties of the electro-synthesized ferrite nanoparticles with a homogeneous incorporation of Zn in the structure shows that the saturation magnetization and the coercively values are highly dependent on the chemical composition and crystal size.

---

1. Introduction

Spinel ferrites belong to a collection of magnetic materials that possess a wide array of technological and biomedical applications^1–4 due to their unique and versatile magnetic and electric properties, showing both high magnetization and resistivity. Spinel ferrites are described using the general formula (M_1−αFe_α)[M_αFe_2−α]O₄, where M correspond to a divalent cation (M = Fe, Co, Ni, Mn, Zn), and (M_1−αFe_α) and [M_αFe_2−α] represent tetrahedral and octahedral cation sites, respectively, in a face-centred cubic oxygen anion sublattice. The inversion degree is denoted by the parameter (α), leading to different structural orders, termed normal spinel if α is equal to zero, inverse spinel if α equals unit, and partially inverse spinel if 0 < α < 1.⁵ Ferrite nanoparticles generally display superparamagnetic behaviour maintaining their high saturation magnetization below a critical size.⁶ Hence, the nature of the cation, the cation distribution and the size of the particles all exert a decisive influence on the properties of the ferrite nanoparticles. Therefore, the study of this relationship between structure and properties is essential to achieve a better understanding of both the magnetic interaction and the specific physicochemical and magnetic properties required for certain applications.

Among the spinel ferrites, zinc ferrite has long been the subject of studies due to its unique properties such as its chemical and thermal stability⁷ and the reduced toxicity of Zn over other metals.⁸ Zinc ferrite encompasses a large number of technological applications, such as magnetic materials,⁹ gas sensors,¹⁰ photocatalysts,¹¹ high density magnetic recording devices¹² and biomedical applications such as MRI contrast agents¹³ or magnetic fluid hyperthermia,¹⁴ among others. Furthermore, its remarkable nature is also due to its unusual and versatile magnetic behaviour in nanoregime.

Bulk spinel zinc ferrite possesses a normal crystalline structure, where Zn cations occupy the tetrahedral sites, showing an antiferromagnetic order with a Néel temperature of 10 K.¹⁵ On the contrary, the nanosized counterparts present a partially inverse spinel structure, influencing the physical and magnetic properties of the material.¹⁶ Such magnetic properties are also affected by factors such as particle size, homogeneity, crystalline structure, microstructure and chemical composition, which are sensitive to the preparation methodology used in the nanoparticles synthesis.¹⁷

Numerous methods have been employed for synthesizing zinc ferrite nanoparticles, such as solvothermal,¹⁸ co-precipitation,⁷ thermal decomposition,¹⁹ hydrothermal synthesis²⁰ or ball milling.²¹ In this study a new methodology has been established for the synthesis of zinc ferrite nanoparticles based on the electrochemical method previously developed by Mazario et al. for cobalt and manganese ferrite nanoparticles.^22,23 The advantage of this method over those previously applied is the versatility of the synthesis by means of the modulation of certain parameters. The variation of the current intensity applied or the reaction temperature, allows monodisperse size ferrite nanoparticles with a controlled chemical composition and particle size to be obtained.

The motivation behind this work was to synthesize a series of zinc ferrite nanoparticles with a monodisperse size and defined chemical composition in a controlled and reproducible manner. Following this, a detailed structural characterization was performed with the aim of investigating how the intensity current affects the crystal structure of the nanoparticles and, consequently, their properties.

2. Experimental

2.1. Synthesis of zinc ferrite nanoparticles

Zinc ferrite nanoparticles were prepared following a one-step electrochemical synthesis method previously reported for the preparation of other ferrites nanoparticles.^22,23 Two iron and zinc foils, both 2 cm², were used as anodes and placed parallel to each other whilst a cylindrical iron counter-electrode possessing an area of 120 cm² was employed as the cathode. These materials were purchased from Goodfellow with purities of 99.5% and 99.9% for Fe and Zn, respectively. The electrodes were immersed in a 0.04 M aqueous solution of tetrabutylammonium bromide (nBu₄NBr) (Merck), which was used simultaneously as a supporting electrolyte and as a surfactant to avoid the agglomeration of the nanoparticles. The synthesis was performed under constant stirring for 30 minutes, whilst varying the current applied to the electrolytic cell using an AMEL model 549 potentiostat/galvanostat. The current applied to the Fe anode was fixed at 100 mA, whilst it was varied between 5 and 50 mA for the Zn anode. The powder obtained was magnetically separated, washed repeatedly with water and dried under vacuum at 60 °C for 12 h (samples named as Zn_xFe_(3−x)O₄). Following this the samples were thermally treated up to 580 °C for 12 hours in order to study possible changes in their microstructure (samples named as Zn_xFe_(3−x)O₄Cal).

2.2. Characterization

Chemical analysis was performed through inductively coupled plasma emission spectroscopy (ICP) using a Perkin Elmer model Optima 2100 DV system. Transmission electron microscopy (TEM) images of the synthesized nanoparticles were collected in a JEOL JEM 1010 operating at acceleration voltage of 100 kV in order to study the size and the shape of the nanoparticles. High-resolution images of the nanoparticles and their spectroscopic analyses were performed in an FEI Titan XFEG 60–300 kV spherical aberration-corrected microscope operated at 300 kV. This microscope was equipped with a monochromator, an EDAX EDS detector, a Gatan Tridiem Energy Filter for spectroscopy measurements, and a CEOS corrector for electron probe-aligned prior observations, allowing an image resolution of 0.8 Å. The crystalline phase and the crystal size of the resulting ferrite nanoparticles were investigated via X-ray diffraction. X-ray diffractograms were recorded between 5° and 80° 2θ in a D5000 diffractometer, equipped with a secondary monochromator and SOL-X Bruker detector with Cu Kα radiation and analyzed using the FullProf Suite program,²⁴ based on the Rietveld method. The crystalline phase of the nanoparticles was also studied by Raman spectroscopy. Raman spectra were collected in a confocal Raman microscopy integrated with atomic force microscopy (AFM) on a CRM-Alpha 300 RA microscope (WITec, Ulm, Germany) equipped with Nd:YAG dye laser (maximum power output of 50 mW power) at 532 nm. In particular, power of the incident laser used to obtain the Raman spectra was fixed at 0.2 mW to avoid possible undesirable sample modifications. The collected spectra were then analyzed using WITec Control Plus Software, in which Raman mode positions were fitted using Lorentzian functions. The Raman images of 7 × 7 μm consisted of 440 spectra of 10 s of integration time each performed at 0.2 mW. Magnetic characterization was performed using a Vibrating Sample Magnetometer (VSM) MLVSM9 MagLab 9T, Oxford Instruments and a SQUID magnetometer (Quantum Design MPMS XL-5) with Reciprocating Sample Option (RSO). The magnetization hysteresis loops M(H) were measured at the VSM at room temperature, applying magnetic fields up to 1 T. The temperature dependence of the magnetization, M(T), measured at SQUID, was determined for each sample under 1000 Oe and over a temperature range between 5 and 400 K. Zero-field-cooling (ZFC) and field-cooling (FC) procedures were also carried out. A ZFC magnetization curve was obtained by cooling the sample in a zero external magnetic field from 400 to 5 K. Subsequently, a magnetic field was applied allowing the magnetization to be measured whilst heating the sample. The same process was employed for the FC curve but the magnetic field was also applied while cooling the sample from 400 to 5 K.

3. Results and discussion

In the present work a series of zinc ferrite nanoparticles with controlled chemical composition were easily synthesized following an electrochemical method previously developed by our research group for the preparation of other ferrite nanoparticles such as CoFe₂O₄,²³ NiFe₂O₄ ²⁵ or MnFe₂O₄,²⁶ introducing several modifications.

The amount of Fe and Zn cations generated during the electrolysis, hence the relative ion quantity or [Fe/Zn] fraction in the reaction medium, is directly related with the oxidation velocity of the metal foils, which is dependent on the current intensities applied to the electrolytic cell. Thus this allows the effortless control of the relative content of Zn and Fe cations in the resulting ferrite. In principle this is done by tuning the current applied to each electrode. Concretely ZnFe₂O₄ produced a minimum [Fe/Zn] fraction of 2 for a ferrite because of the stoichiometry of the material allowing it to obtain higher relative ion proportions. This control of the stoichiometry of the ferrite nanoparticles is of great importance as it plays a determinant role in the magnetic properties of the nanomaterial.

Several samples were synthesized applying a fixed current intensity of 100 mA to the Fe anode whilst varying it for the Zn anode (5, 10, 20, 30, 40 and 50 mA). The chemical analysis calculated by ICP of the samples reveals that, in effect, there is a linear dependence between the current intensity applied to the Zn anode and the quantity of Zn incorporated into the ferrite structure (x), defining the zinc ferrite formula as Zn_xFe_(3−x)O₄ (see Fig. 1). The content of Zn²⁺ in the structure increases from x = 0.18 ± 0.02 to an almost stoichiometric composition, x = 0.93 ± 0.14, when the current augments between 5 and 50 mA.

Fig. 1 Representation of the dependence between the amount of Zn²⁺ incorporated into the ferrite structure (x) with a general formula Zn_xFe_(3−x)O₄ and the current intensity (I) applied to the Zn anode.

Subsequently, XRD measurements were performed on all of the samples prepared with variable chemical compositions to confirm the formation of crystalline zinc ferrite nanoparticles. Fig. 2a shows the diffractograms for this set of samples, named as Zn_xFe_(3−x)O₄. The presence of a pure single phase can be clearly observed in all patterns, where the peaks can be indexed according to the Fd3̄m: 227 cubic spinel structure. The crystal cell parameters of each sample were determined by Rietveld refinement and are given in Table 1. Remarkably, a linear variation of its value with the Zn content of the nanoparticles was observed, increasing the cell parameters as the Zn content in the ferrite augments. This demonstrates the presence of Zn inside the spinel structure (Fig. 2b) as the cell parameter augments with the incorporation of a larger cation in the cubic structure.

Fig. 2 (a) X-ray diffractograms of Zn_xFe_(3−x)O₄ set of samples. Bragg positions are marked (|). (b) The linear dependence between crystal cell parameter, a, of Zn_xFe_(3−x)O₄ set of samples (magnetite included) and the amount of Zn²⁺ incorporated into the ferrite structure, x.

Table 1 The relationship between the current intensity applied to the Zn anode and the quantity of Zn²⁺ incorporated into the ferrite structure (x), and therefore the stoichiometry of the zinc ferrite nanoparticles. Crystal cell parameters calculated by the Rietveld method using the FullProf Suite program. Crystal sizes estimated by the Debye–Scherrer equation. Particle sizes measured from the TEM micrographs

I (mA)	ZnxFe(3−x)O4	a (10^−2 Å)	Crystal size (nm)	Particle size (nm)
5	Zn0.18Fe2.82O4	8.401 (1)	18.8 (4)	15 (2)
10	Zn0.3Fe2.7O4	8.409 (1)	16.3 (5)	14 (1)
20	Zn0.5Fe2.5O4	8.418 (1)	12.7 (3)	12 (1)
30	Zn0.64Fe2.36O4	8.435 (1)	10.6 (4)	8.5 (5)
40	Zn0.84Fe2.16O4	8.445 (1)	8.9 (2)	7.8 (4)
50	Zn0.93Fe2.07O4	8.463 (2)	9.0 (2)	6.2 (2)

---

Interestingly, the average crystal size estimated from the Debye–Scherrer equation also varies with the Zn content in the structure, decreasing as the x value increases, as shown in Table 1. This behaviour has been reported for other ferrites and attributed to thermodynamic factors rather than nucleation features.²⁷

The crystal size values obtained by XRD were in good agreement with the particle size determined by TEM, indicating that the nanoparticles can be considered monocrystalline. Fig. 3a and b show the micrographs of two of the spinel zinc ferrites synthesized relatively uniform in size, but isotropic in shape.

Fig. 3 TEM micrographs of Zn_0.64Fe_2.36O₄ (a) and Zn_0.93Fe_2.07O₄ (b) NPs and their particle size distributions (insets).

The synthesized nanoparticles were further investigated to corroborate the homogeneous incorporation of the Zn²⁺ in the structure. Fig. 4 presents the C_s-corrected STEM-HAADF analysis of the Zn_0.64Fe_2.36O₄. Fig. 4a depicts a high-resolution image of the various nanoparticles, confirming the good crystallinity of the material. In the centre of the image a nanoparticle can be seen sitting along its [110] zone axis. The FFT diffractograms (inset) were indexed assuming Fd3̄m symmetry with a unit cell of ≈8.38 Å. EELS spectrum image analysis was acquired of the nanoparticles presented in Fig. 4b. The extracted signal maps for oxygen, iron and zinc, confirm the homogeneous distribution of the three elements along the entire nanoparticles.

Fig. 4 (a) Atomic resolution C_s-corrected STEM-HAADF image of nanoparticles orientated along the along their [110] zone axis with the FFT shown inset. (b) EELS spectrum image analysis acquired of various nanoparticles, extracted signal maps for oxygen, iron and zinc (right of the image).

The Zn_xFe_(3−x)O₄ samples were then thermally treated up to 580 °C in an oven under static air atmosphere for 12 hours, to investigate the influence of the annealing process in the crystallinity of the structures.

The samples obtained after this calcination were thus named Zn_xFe_(3−x)O₄Cal. In general, all the samples present a common tendency to increase in size as a result of the annealing process at high temperatures. Fig. 5 shows a TEM image and the XRD pattern of the sample Zn_0.93Fe_2.07O₄Cal, which was compared with the results obtained before calcination. An increase in particle size up to ∼30 nm can be clearly observed, calculated using TEM, the Debye–Scherrer equation and the narrowing of the diffraction peaks. Fig. 5b also indicates the sample maintains its spinel structure after the thermal treatment as there is no evidence of structural changes.

Fig. 5 (a) TEM micrograph of Zn_0.93Fe_2.07O₄Cal and particle size distribution in the inset. (b) X-ray diffractograms of Zn_0.93Fe_2.07O₄ and Zn_0.93Fe_2.07O₄Cal, presenting both of their spinel structures. Bragg positions are marked (|).

The XRD patterns of the remaining non-stoichiometric ferrites are shown in Fig. 6, where evident differences can be observed. They present lower maximum diffraction intensities corresponding to the spinel structure, while new diffraction maximums are detected belonging to the hematite (α-Fe₂O₃) structure which is the new phase generated after this thermal treatment.

Fig. 6 X-ray diffractograms of Zn_xFe_(3−x)O₄Cal set of samples. Bragg positions are marked in the upper position for the spinel structure (|) and in lower position for the hematite structure (+).

The intensity of the peaks corresponding to the hematite phase progressively decrease with Zn content, concomitantly with the increase of the peaks related to the ferrite spinel structure. Table 2 shows the Zn quantity in the Zn_xFe_(3−x)O₄ set of samples previous to the annealing, obtained by the ICP analysis, compared to the percentage of zinc ferrite phase after the thermal treatment. It has been expressed in terms of the percentage of x in Zn_xFe_(3−x)O₄. It is noteworthy that the percentages of Zn calculated by ICP analysis before calcination in each sample are almost equivalent to the percentages of the zinc ferrite phase generated after the thermal treatment.

Table 2 Zn and ZnFe₂O₄ percentages in each sample and crystal cell parameters (a) of Zn_xFe_(3−x)O₄ and Zn_xFe_(3−x)O₄Cal sets of samples, before and after thermal treatment; and crystal cell parameters (a and c) of the α-Fe₂O₃ phase generated, all obtained by X-ray Rietveld refinement

Sample	% Zn	% ZnFe2O4	ZnFe2O4	Fe2O3
			a (Å)	a (Å)	c (Å)
Zn0.18Fe2.82O4Cal	19 (2)	22 (11)	8.440 (9)	5.036 (3)	13.749 (14)
Zn0.3Fe2.7O4Cal	31 (2)	32 (10)	8.440 (5)	5.035 (2)	13.749 (9)
Zn0.5Fe2.5O4Cal	49 (9)	55 (7)	8.439 (3)	5.035 (1)	13.750 (6)
Zn0.64Fe2.36O4Cal	64 (9)	78 (7)	8.439 (2)	5.035 (2)	13.751 (7)
Zn0.84Fe2.16O4Cal	84 (3)	98 (10)	8.438 (4)	5.039 (1)	13.738 (6)
Zn0.93Fe2.07O4Cal	93 (14)	100 (3)	8.439 (1)	—	—

---

Considering the crystal cell parameter values, it is noticeable that there is a change following the thermal treatment in all the non-stoichiometric samples, tending all of them to acquire the stoichiometric ZnFe₂O₄ ferrite crystal cell parameter value, 8.44 Å.²⁸ The new hematite phase generated after the thermal treatment also presents its crystal cell parameters in good agreement with the ones described in literature, a = 5.034 Å and c = 13.749 Å.²⁹

These facts seem to indicate that the material was initially synthesized with the Zn²⁺ cations homogeneously situated along the entire crystalline structure, whilst a Zn segregation mechanism takes place during the thermal treatment, generating stoichiometric zinc ferrite and pure hematite phases. The percentage of each phase depends on the initial chemical composition of the non-stoichiometric zinc ferrite.

Subsequently, Raman spectroscopy was employed to obtain further insight into the structure of the zinc ferrites nanoparticles synthesized with a variable chemical composition, Zn_xFe_(3−x)O₄ (0 ≤ x≤ 0.93). As previously discussed, all the samples presented a pure cubic spinel phase independent of the different amounts of Zn²⁺ cations distributed along the crystalline structure. These cubic spinel zinc ferrites belong to the O_h⁷ (Fd3̄m) space group with eight formula units per unit cell, and according to the group theory, five Raman active modes are typically observed at ambient temperatures (A_1g + E_g + 3F_2g). However, as shown in Fig. 7, only three Raman bands are clearly observed in the range between 200 and 1000 cm⁻¹. Specifically, these three first-order Raman modes at 336–342, 500–511 and 667–696 cm⁻¹ correspond to F_2g(2), F_2g(3) and A_1g symmetries, respectively. The observed Raman modes and their assignments are collected in Table 3 for each synthesized sample containing different contents of zinc, being in concordance with the values found in literature.³⁰

Fig. 7 Raman spectra of the Zn_xFe_(3−x)O₄ set of samples and magnetite, each performed from 10 discrete spectra.

Table 3 Raman modes observed for the synthesized spinel zinc ferrites and magnetite modes fitted by a Lorentzian function, and compared with other values reported in literature³⁰

Assignments	ZnFe2O4 (ref. 30)	Raman modes (cm^−1) of ZnxFe(3−x)O4
		x = 0.93	x = 0.84	x = 0.64	x = 0.5	x = 0.3	x = 0.18	x = 0
F2g(1)	221	—	—	—	—	—	—	—
Eg	246	—	—	—	—	—	—	—
F2g(2)	355	336	336	357	333	333	342	343
F2g(3)	451	500	496	495	501	502	503	511
A1g	647	667	673	673	680	680	685	696

---

In contrast, the F_2g(1) and E_g modes are absent in all of the zinc ferrites nanoparticles. This behaviour has been previously observed in other zinc ferrites systems,³⁰ and attributed to the small crystalline size of the nanoparticles and also to possible cation inversion, typically found in Zn ferrites nanoparticles.¹⁶ These features provoke a broadening of the modes, which are intrinsically of low intensity, and would hinder their observation. It is well known that the zinc ferrites, at the nanoscale, present a partially inverted spinel structure which implies an alteration of the distribution of Fe³⁺ and Zn²⁺ ions among the octahedral and tetrahedral sites. Therefore, at the same site two types of cations can vibrate leading to two different Raman modes, very close in frequencies, and consequently, the appearance of broad peaks in the spectra are usually observed.³¹ In fact, the three active modes visible in the spectra exhibit clear broadening characteristics.

Remarkably, the A_1g mode progressively shifts toward lower wavenumbers as the content of Zn increases in the ferrite, from 696 to 667 cm⁻¹, whereas the F_2g modes practically do not vary with the chemical composition of the ferrites, the shift does not seem to follow any trend (Fig. 7 and Table 3). In cubic spinels, such as ferrites, the modes above 600 cm⁻¹ are attributed to the motion of oxygen in tetrahedral AO₄ groups, while the other low wavenumber modes represent the characteristics of the octahedral BO₆ sites.³⁰ Thus, it is suggested that the Zn²⁺ cations are preferentially located in tetrahedrally coordinated A sites as there is a shift in the A_1g mode as the Zn is incorporated into the structure. The incorporation of Zn²⁺ progressively increases the cell parameters as demonstrated by XRD, which leads to a reduction of the bond force constant and consequently, to a softening of the Raman frequencies of the corresponding bands.³² Nevertheless, we cannot totally discard a contribution of the variation of the crystal size in the Zn_xFe_(3−x)O₄ on the observed peak shift. The XRD and TEM measurements indicated that the incorporation of Zn to the structures decreases the crystal and particle size that accordingly could augment the crystalline disorder and the presence of grain boundaries. However, this explanation seems to be less plausible because the F_2g modes do not follow this trend.

The samples thermally treated at 580 °C were then analysed by Raman spectroscopy to investigate the effect of the chemical composition on the crystalline phase transition and to get a further insight of the zinc ferrites nanoparticle structure. Fig. 8 displays the Raman spectra of the Zn_xFe_(3−x)O₄Cal nanoparticles that clearly show the appearance of new phase peaks associated to the active Raman modes of hematite structures (*) in the samples with lower content of Zn, 0.18 ≤ x ≤ 0.64. Moreover, an intense band at 1320 cm⁻¹ associated to the hematite is observed which can be assigned to an overtone or second order scattering process.

Fig. 8 Average Raman spectra of the Zn_xFe_(3−x)O₄Cal set of samples (* main hematite phase peaks) each performed from 10 discrete spectra.

Table 4 summarizes the positions of the Raman modes fitted by assuming Lorentz peak shapes. It is worthy to mention that the positions of the bands corresponding to the zinc ferrite nanoparticles do not change with respect to the values observed for the almost stoichiometric Zn_0.93Fe_2.07O₄ before the thermal treatment. In addition, the Raman shifts obtained in calcined samples are very similar between them, independently of the chemical composition of the nanoparticles.

Table 4 Raman modes observed in the zinc ferrites samples after the thermal treatment

Assignments	Ref. 30 and 33	Raman modes of ZnxFe(3−x)O4Cal
		x = 0.93	x = 0.92	x = 0.64	x = 0.5	x = 0.3	x = 0.26
F2g(2) (ZnFe2O4)	355	354	357	354	356	—	—
F2g(3) (ZnFe2O4)	451	444	450	448	443	441	444
A1g (ZnFe2O4)	647	660	668	668	668	665	665
A1g (α-Fe2O3)	225	—	—	225	225	225	225
Eg (α-Fe2O3)	247	—	—	245	245	245	245
Eg (α-Fe2O3)	293	—	—	294	294	294	294
Eg (α-Fe2O3)	412	—	—	410	410	410	410
A1g (α-Fe2O3)	498	—	—	500	500	500	500
Eg (α-Fe2O3)	613	—	—	613	613	613	613
Overtone (α-Fe2O3)	1320	—	—	1320	1320	1320	1320

---

These results are in agreement with the conclusions obtained from the XRD measurements. The zinc ferrite nanoparticles synthesized with varying contents of Zn²⁺ cations exhibit a pure spinel structure in which the cell parameter increases with the proportion of Zn incorporated to the structure. In turn, this provokes a shift in the A_1g mode towards lower wavenumbers. Subsequently the occurrence of a homogeneous distribution of the Zn²⁺ cations in the spinel structure can be suggested. However, upon thermal treatment migration and redistribution of the Zn cations may be induced that can lead to the formation of two different phases, hematite and stoichiometric zinc ferrites. As expected, in the prepared samples with almost stoichiometric composition (x = 0.93), the calcination process only provokes a slightly increase in the particles size rather than a detectable crystalline phase transition.

This redistribution of the Zn²⁺ cation within the non-stoichiometric samples, induced by increasing the temperature, was analysed more in detail. In particular, Raman spectra of the sample Zn_0.64Fe_2.36O₄ were recorded at different temperatures (Fig. 9).

Fig. 9 Raman spectra of the Zn_0.64Fe_2.36O₄ nanoparticles recorded at various temperatures (−173 °C to 500 °C).

The appearance of the peaks belonging to hematite due to the partial phase transition from ferrite to hematite can be clearly observed starting at about 330 °C, with the concomitant formation of stoichiometric ferrite as a consequence of this redistribution of the Zn in the sample. At 500 °C the crystal phase transition had finished and the sample was cooled down to room temperature, wherein the spectrum exhibits peaks of both phases, zinc ferrite and hematite.

Subsequently, to investigate the grade of segregation of theses phases upon thermal treatment, Raman mappings of non-stoichiometric samples were later performed. Fig. 10 shows the Raman images obtained for the samples Zn_0.93Fe_2.07O₄Cal, Zn_0.64Fe_2.36O₄Cal and Zn_0.18Fe_2.82O₄Cal, with differing contents of Zn²⁺ cations.

Fig. 10 Raman images of spatial phase distribution of samples after thermal treatment: (a) Zn_0.93Fe_2.07O₄Cal, (b) Zn_0.64Fe_2.36O₄Cal and (c) Zn_0.18Fe_2.82O₄Cal. The Raman spectra associated with each region are shown below each image (the blue region representing the zone enriched in zinc ferrite and the red region are demonstrates the areas enriched with hematite).

The previous XRD measurement and the average Raman spectrum of the Zn_0.93Fe_2.07O₄Cal showed only the ferrite phase without any evidence of hematite. Fig. 10a shows that although zinc ferrite is the predominant phase, smaller regions enriched in hematite can also be seen. The Raman image of Zn_0.64Fe_2.36O₄Cal (Fig. 10b) also exhibit two clear regions, one consisting of almost pure hematite whilst the other shown being enriched in ferrite, however, and as expected, the hematite region augments with respect to the Zn_0.93Fe_2.07O₄Cal sample. On the other hand, in the sample possessing the lowest concentration of zinc, Zn_0.18Fe_2.82O₄Cal, only a hematite phase was observed within the entire image, although from its XRD pattern 20% of ZnFe₂O₄ was estimated. This could be attributed to overlapping of the weak Raman bands of the ferrite with the strong bands of the hematite, which may have caused difficulties in resolving the spectrum.

Therefore, it was demonstrated the migration of zinc cations from homogeneous ferrites when a thermal treatment is performed and the formation of segregated micro-regions consisting of different crystalline phases, Zn ferrite and hematite.

It is well known that the magnetic properties of zinc ferrites nanoparticles are highly dependent on their synthesis process which ultimately varies their chemical composition, inversion degree and particle size.³⁴ Next, the magnetic properties of the synthesized zinc ferrites nanoparticles by this electrochemical method were further investigated. The zero field cooling-field cooling, ZFC–FC, curves between 5 and 350 K and under a 1000 Oe external magnetic field of the samples with x = 0.18, 0.5, 0.64, 0.84, are depicted in Fig. 11. As it can be seen, all samples display similar magnetic behaviour, presenting a maximum in the ZFC curve, except the sample x = 0.18 in which the maxima is expected to be above 350 K. This maximum is entitled the blocking temperature, T_B, and is related to the blocking/freezing of magnetic moments. As expected, the T_B increases when Zn content decreases, from 100 K for the sample where x = 0.93 to 300 K for the sample where x = 0.5. Furthermore the maximum of ZFC is broadened when there is a decrease in the content of Zn. Results also demonstrated that the samples x = 0.18, 0.5 show ferromagnetic or superferromagnetic behaviour with a magnetic phase transition above 350 K, whereas samples x = 0.64, 0.84 exhibit superparamagnetic behaviour, with a broad particle size distribution. The maxima observed in the FC curves could be associated with a Morin transition,³⁵ however it is not correlated with the Zn content, or with a spin-glass process in the shell of the nanoparticles.³⁶

Fig. 11 ZFC (black) and FC (red) magnetic susceptibility for Zn_xFe_3−xO₄, (x = 0.18, 0.5, 0.64, 0.84) samples at H = 1000 Oe.

Fig. 12 shows magnetization plotted against the applied magnetic field measured at room temperature for all the different samples synthesized.

Fig. 12 The M vs. H dependences measured at 290 K for all the synthesized zinc ferrites nanoparticles with variable chemical composition, magnetite included. The inset shows an augment of the plots.

All samples show the characteristic hysteresis loop of single domain uniaxial nanoparticles. Moreover, none of these samples present evidence for the co-existence of magnetic phases (two phases of two different nanoparticle sizes), in accordance with the X-ray diffraction analysis. Saturation magnetization (M_s) values decreased with the increase in Zn content, as also previously observed in the ZFC–FC curves. In addition, a decrease in the coercivity field values (H_c) as Zn increases in the structure was also observed. The values of these magnitudes are summarised in Table 5.

Table 5 Magnetic parameters of the zinc ferrites samples

Sample	Ms (emu g^−1)	Hc (Oe)
Fe3O4	82	95.5
Zn0.18Fe2.82O4	65	33.2
Zn0.30Fe2.7O4	77	28.5
Zn0.50Fe2.5O4	55	7.0
Zn0.64Fe2.36O4	13	7.5
Zn0.84Fe2.16O4	24	12.0
Zn0.93Fe2.07O4	7	12.0

---

Although the magnetic behaviour could well be correlated with Zn concentration, one cannot overlook how the crystal size diminishes with the amount of Zn in the ferrite. In contrast, Zn_0.3Fe_2.7O₄ and Zn_0.84Fe_2.16O₄ nanoparticles exhibit higher M_s values in comparison with other samples with lower contents of Zn. This seems to indicate a certain inversion degree present in the electro-synthesized nanoparticles that affect their magnetic properties, increasing the magnetic response, as previously and extensively observed in others zinc ferrites.¹⁶ As mentioned before, the magnetic behaviour of spinel depends on the substitutional metal percentage, cation inversion, shape and size of the nanoparticles and all of these parameters are dependent on the synthesis method. This assortment of dependences prevents an accurate determination of the factors governing the magnetic behaviour. Further investigations need to be performed in this concern, by the electrochemical synthesis of zinc ferrites nanoparticles with different contents on Zn and similar particle sizes.

4. Conclusions

In this study, zinc spinel ferrite nanoparticles were synthesized by an electrochemical method presenting a high level of versatility and enabling control of the composition and size of the nanoparticles by varying the intensity current applied in a reproducible manner. It was concluded that the amount of Zn²⁺ in the ferrite linearly augments with the current applied to the electrolytic cell. Several Zn_xFe_(3−x)O₄ (x = 0.2–0.93) ferrite nanoparticles have been obtained with particles sizes in the range of 6 to 15 nanometers with a homogeneous distribution of the Zn²⁺ cation in the structure. However, a Zn segregation process occurs following thermal treatment of the sample in which the Zn²⁺ cations homogeneously situated along the entire crystalline structure diffuse towards the formation of two different crystalline structure; stoichiometric zinc ferrite and pure hematite. In addition, Raman mapping of the sample reveals that this annealing process produces micro-regions, consisting of domains enriched in Zn ferrite and hematite crystalline phases. From the analysis of the magnetic properties, it can be concluded that although all the samples display hysteresis behaviour, the saturation magnetization and the coercivity significantly vary with the concentration of Zn in the ferrite, decreasing as the contents augment. Although some exceptions were observed, these have been attributed to the inversion degree in the samples.

Acknowledgements

The authors gratefully acknowledge the MINECO (Projects MAT2012-37109-C02-02 and MAT2013-48009-C4-1-P) for financial support. A. Muñoz-Bonilla also thanks the MINECO for her Ramon y Cajal contract.

References

1. S. Kalia, S. Kango, A. Kumar, Y. Haldorai, B. Kumari and R. Kumar, Colloid Polym. Sci., 2014, 292, 2025–2052.
2. D. Stanicki, L. Vander Elst, R. N. Muller and S. Laurent, Curr. Opin. Chem. Eng., 2015, 8, 7–14.
3. B. Antic, M. Perovic, A. Kremenovic, J. Blanusa, V. Spasojevic, P. Vulic, L. Bessais and E. S. Bozin, J. Phys.: Condens. Matter, 2013, 25, 086001.
4. E. Petrova, D. Kotsikau and V. Pankov, J. Magn. Magn. Mater., 2015, 378, 429–435.
5. A. Milutinović, Z. Lazarević, Č. Jovalekić, I. Kuryliszyn-Kudelska, M. Romčević, S. Kostić and N. Romčević, Mater. Res. Bull., 2013, 48, 4759–4768.
6. C. Pereira, A. M. Pereira, C. Fernandes, M. Rocha, R. Mendes, M. P. Fernández-García, A. Guedes, P. B. Tavares, J. M. Grenèche, J. P. Araújo and C. Freire, Chem. Mater., 2012, 24, 1496–1504.
7. R. Raeisi Shahraki, M. Ebrahimi, S. a. Seyyed Ebrahimi and S. M. Masoudpanah, J. Magn. Magn. Mater., 2012, 324, 3762–3765.
8. J. Wan, X. Jiang, H. Li and K. Chen, J. Mater. Chem., 2012, 22, 13500.
9. L. Tung, V. Kolesnichenko, G. Caruntu, D. Caruntu, Y. Remond, V. Golub, C. O'Connor and L. Spinu, Phys. B, 2002, 319, 116–121.
10. S. Tyagi, N. Batra and A. K. Paul, Trans. Indian Inst. Met., 2015, 68, 707–713.
11. R. Rahimi, M. Heidari-Golafzani and M. Rabbani, Superlattices Microstruct., 2015, 85, 497–503.
12. H. Malik, A. Mahmood, K. Mahmood, M. Y. Lodhi, M. F. Warsi, I. Shakir, H. Wahab, M. Asghar and M. A. Khan, Ceram. Int., 2014, 40, 9439–9444.
13. C. Bárcena, A. K. Sra, G. S. Chaubey, C. Khemtong, J. P. Liu and J. Gao, Chem. Commun., 2008, 2224–2226.
14. Y. Yang, X. Liu, Y. Yang, W. Xiao, Z. Li, D. Xue, F. Li and J. Ding, J. Mater. Chem. C, 2013, 1, 2875.
15. M. Mozaffari and H. Masoudi, J. Supercond. Novel Magn., 2014, 27, 2563–2567.
16. R. Sai, S. D. Kulkarni, S. S. M. Bhat, N. G. Sundaram, N. Bhat and S. a. Shivashankar, RSC Adv., 2015, 5, 10267–10274.
17. Z. Ž. Lazarević, Č. Jovalekić, V. N. Ivanovski, A. Rečnik, A. Milutinović, B. Cekić and N. Ž. Romčević, J. Phys. Chem. Solids, 2014, 75, 869–877.
18. V. Blanco-Gutierrez, E. Climent-Pascual, M. J. Torralvo-Fernandez, R. Saez-Puche and M. T. Fernandez-Diaz, J. Solid State Chem., 2011, 184, 1608–1613.
19. O. Kaman, T. Kořínková, Z. Jirák, M. Maryško and M. Veverka, J. Appl. Phys., 2015, 117, 17C706.
20. R. M. Freire, T. S. Ribeiro, I. F. Vasconcelos, J. C. Denardin, E. B. Barros, G. Mele, L. Carbone, S. E. Mazzetto and P. B. a. Fechine, J. Nanopart. Res., 2013, 15, 1616.
21. T. F. Marinca, I. Chicinaş, O. Isnard and V. Pop, Optoelectron. Adv. Mater., Rapid Commun., 2011, 5, 39–43.
22. E. Mazario, J. Sánchez Marcos, N. Menéndez, M. Cañete, A. Mayoral, S. Rivera-Fernandez, J. M. De La Fuente and P. Herrasti, J. Phys. Chem. C, 2015, 150304123957002.
23. E. Mazarío, P. Herrasti, M. P. Morales and N. Menéndez, Nanotechnology, 2012, 23, 355708.
24. J. Rodríguez-Carvajal, Phys. B, 1993, 192, 55–69.
25. R. Galindo, E. Mazario, S. Gutiérrez, M. P. Morales and P. Herrasti, J. Alloys Compd., 2012, 536, S241–S244.
26. E. Mazario, J. Sánchez Marcos, N. Menéndez, M. Cañete, A. Mayoral, S. Rivera-Fernandez, J. M. De La Fuente and P. Herrasti, J. Phys. Chem. C, 2015, 150304123957002.
27. B. T. Naughton and D. R. Clarke, J. Am. Ceram. Soc., 2008, 91, 1253–1257.
28. M. A. Hakim, M. Manjurul Haque, M. Huq and P. Nordblad, Phys. B, 2011, 406, 48–51.
29. M. Lübbe, A. M. Gigler, R. W. Stark and W. Moritz, Surf. Sci., 2010, 604, 679–685.
30. J. P. Singh, R. C. Srivastava, H. M. Agrawal and R. Kumar, J. Raman Spectrosc., 2011, 42, 1510–1517.
31. Z. Wang, D. Schiferl, Y. Zhao and H. S. C. O'Neill, J. Phys. Chem. Solids, 2003, 64, 2517–2523.
32. L. Malavasi, P. Galinetto, M. C. Mozzati, C. B. Azzoni and G. Flor, Phys. Chem. Chem. Phys., 2002, 4, 3876–3880.
33. D. L. A. de Faria, S. V. Silva and M. T. de Oliveira, J. Raman Spectrosc., 1997, 28, 873–878.
34. S. J. Stewart, I. a. Al-Omari, F. R. Sives and H. M. Widatallah, J. Alloys Compd., 2010, 495, 506–508.
35. R. Zysler, D. Fiorani and A. Testa, J. Magn. Magn. Mater., 2001, 224, 5–11.
36. S. Bedanta and W. Kleemann, J. Phys. D: Appl. Phys., 2009, 42(1BC), 013001.

---