Hydrothermal synthesis and structure evolution of metal-doped magnesium ferrite from saprolite laterite

Abstract

Spinel metal-doped magnesium ferrite (MgFe₂O₄) was synthesized using an atmospheric hydrochloric acid leaching process and hydrothermal synthesis process from saprolite laterite. The effects of the hydrothermal preparation conditions, such as pH value of the acid leaching solution, hydrothermal temperature and time, on the formation of magnesium ferrite were systematically investigated. It was shown that pure magnesium ferrite could be obtained when the pH value of the leaching solution was controlled at 12.0 and the hydrothermal reaction was conducted at 160 °C for 6 h. More importantly, the structural evolution of the as-prepared magnesium ferrite was investigated in detail using X-ray diffraction (XRD), X-ray fluorescence (XRF), Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), and vibrating sample magnetometry (VSM). The results indicated that the metal-doped magnesium ferrite displayed soft ferrimagnetic behavior and the magnesium ions migrated from the tetrahedral site to the octahedral site as the calcination temperature increased. The formed magnesium ferrite was indeed mixed with traces of other magnesium-containing compounds.

---

1. Introduction

Spinel ferrites have been of great interest in recent years due to their properties and potential applications such as magnetic applications, high-density data storage, catalysts, gas sensors, and rechargeable lithium batteries.^1–5 Magnesium ferrite (MgFe₂O₄) has a cubic spinel-type structure and is well known as a soft magnetic n-type semi-conductive material⁶ with high resistivity and low magnetic and dielectric losses.

In general, the ferrite formula can be expressed as AB₂O₄, where A represents a tetrahedral site and is a divalent ion (Mg²⁺, Fe²⁺, Ni²⁺, Mn²⁺, Co²⁺ or a combination of these) and B represents an octahedral site and mainly stands for Fe³⁺ but can be substituted by any trivalent ion (Al³⁺ and Cr³⁺). In the normal spinel structure, divalent ions only occupy the A-sites and trivalent spinel ions only occupy the B-sites. The sites occupied by ions are influenced not only by their properties but also by the synthesis method.

Among various preparation methods for magnesium ferrite (MgFe₂O₄),^7–14 the hydrothermal approach has been regarded as one of the most promising and environmentally-friendly synthesis methods. In general, ferrites or metal-doped ferrites were synthesized from pure chemical reagents, wherein the Fe to M (Ni, Co, Mn and Mg) molar ratio (R_Fe/M) was strictly controlled at 2.0 (the stoichiometric ratio). Shan et al.¹² synthesized NiFe₂O₄ with different morphologies, wherein R_Fe/M was at the stoichiometric ratio of 2.0, by the hydrothermal method without adding any surfactant. Andersen et al.¹³ also prepared CoFe₂O₄, where R_Fe/M = 2.0, via the hydrothermal approach. However, this method was not used for the preparation of magnesium ferrite. Shen et al.¹⁴ synthesized MgFe₂O₄ nanospheres via a solvothermal reduction method in which R_Fe/Mg was 0.67. Sasaki et al.¹⁵ found that at the stoichiometric ratio for MgFe₂O₄ (R_Fe/M = 2.0), both MgFe₂O₄ and α-Fe₂O₃ were obtained using a supercritical hydrothermal reaction. However, when R_Fe/M was controlled at 1.0 and 0.67, the product obtained was the desired single-phase MgFe₂O₄. Thus, the nonstoichiometric R_Fe/M played an important role in preparing pure MgFe₂O₄, namely, a relatively lower R_Fe/M (R_Fe/M < 2.0, with a moderate excess of magnesium) benefited the formation of pure MgFe₂O₄.

On the other hand, about 70% of global land-based nickel resources occur as laterite ores. In the conventional hydrometallurgical process for the laterite ores, atmospheric acid leaching, as a highly attractive and promising technology, produces leach liquor with significant concentrations of iron and other impurity ions (such as Mg, Mn and Al), which complicates downstream processing and creates some problems such as a lower Ni and Co recovery rate, difficult separation and increased costs.^16,17 In addition, iron and many ions in the laterite ores are not utilized, which generates large amounts of environmental contamination and wastes resources. Therefore, achieving the maximum utilization of resources is our aim. To the best of our knowledge, no research has been done on the synthesis of magnesium ferrite from saprolite laterite, especially the production of metal-doped magnesium ferrite directly from saprolite laterite by a hydrothermal approach. Furthermore, no data in the literature mentioned the structural evolution and existence of magnesium in metal-doped magnesium ferrite.

In this study, metal-doped magnesium ferrite was synthesized from saprolite laterite using an atmospheric hydrochloric acid leaching process and hydrothermal synthesis process. The atmospheric hydrochloric acid leaching experiments were conducted to investigate the effects of the hydrochloric acid concentration, liquid–solid ratio (mL g⁻¹) and reaction time on the leaching efficiencies of metals from saprolite laterite. During the hydrothermal process, the influencing factors, such as the pH value of the acid leaching solution, hydrothermal temperature and time, on the synthesis of ferrite were studied in detail. Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), and vibrating sample magnetometry (VSM) were utilized to elucidate the microstructure and the state of magnesium in the as-prepared metal-doped magnesium ferrite.

2. Experimental

2.1 Materials

Saprolite laterite used in this study was supplied by the Beijing Research Institute of Mining and Metallurgy. Saprolite laterite was first dried overnight at 100 °C and then ground into powders with a particle size smaller than 150 μm. The chemical analysis and XRD pattern of saprolite laterite are shown in Table 1 and Fig. 1. The main minerals in saprolite laterite were lizardite (Mg₃Si₂O₅(OH)₄), goethite (FeO(OH)), chlorite ((Mg,Al)₆Si₄O₁₀(OH)₈), quartz (SiO₂) and hematite (Fe₂O₃).

Table 1 Chemical compositions of saprolite laterite analyzed by XRF

Compositions	Fe	Mg	Ni	Co	Mn	Ti	Al	Cr	Ca	Si
Content (wt%)	21.74	11.92	1.98	0.05	0.32	0.30	1.79	0.56	0.19	13.04

---

Fig. 1 XRD pattern of saprolite laterite.

Analytical reagent (AR) grade sodium hydroxide (NaOH), glacial acetic acid (HAc) and hydrochloric acid (36%–38%) were purchased from the Beijing Reagent Factory of China and were used without further purification.

2.2 Methods and procedure

Fig. 2 shows a flow chart of the synthesis of metal-doped magnesium ferrite from saprolite laterite. It can be observed from Fig. 2 that the experimental procedure included two parts: the atmospheric acid leaching process and the hydrothermal synthesis process.

Fig. 2 A flow chart of the synthesis of metal-doped magnesium ferrite from saprolite laterite.

2.2.1 Atmospheric acid leaching process. In the acid leaching process, saprolite laterite was added into a 500 mL round bottomed flask and then mixed with 2.5–3.0 M of 120–220 mL HCl solution. The flask was placed in the electric sets to ensure a temperature of 100 °C. A condenser was used to avoid evaporation during the leaching process. Various parameters, such as the hydrochloric acid concentration (2.5–3.0 M), liquid–solid ratio (6–11 mL g⁻¹), and leaching duration (10–90 min), were investigated to optimize the leaching conditions. After the process of leaching, the slurry was separated in a RJ-TDL-50A centrifuge with a speed of 4000 rpm for 15 min and washed with deionized water several times. Then, a solution containing various ions was obtained.

2.2.2 Hydrothermal synthesis process. The acid leaching solution obtained under the optimum leaching conditions was utilized to synthesize metal-doped magnesium ferrite. First, sodium hydroxide solution was added dropwise into the stirring solution until a pH value of 8.0–12.0 was obtained. Then, the precipitates and aqueous solution were removed into a Teflon-lined autoclave to react under hydrothermal conditions. The temperature was controlled from 160 to 200 °C and the reaction was performed from 3 to 12 h. The final products were washed and filtered several times with distilled water until their pH was close to neutral and then dried at 80 °C for 12 h. In addition, a portion of the samples was calcined at 500 and 900 °C for 2 h.

2.2.3 Analysis and characterization. The metal concentrations of the leaching solutions were measured on an Inductively Coupled Plasma Atomic Emission Spectrometer (ICP-AES, America, Varian). The phase and chemical composition of powder specimens were determined using X-ray diffraction (XRD, Japan, Rigaku) and X-ray fluorescence (XRF-1800, Japan). Room temperature Raman spectra were obtained using a Raman spectrometer equipped with an Ar⁺ laser (532 nm, 10 mW) excitation source and a CCD detector. X-ray photoelectron spectroscopy studies were performed using an AXIS ULTRA spectrometer with monochromatic Al Kα (1486.6 eV) radiation. Magnetic characterization was carried out using a vibrating sample magnetometer (VSM, America, 9T(EC-II)). Hysteresis curves were measured at room temperature in the range of 0–6.0 kOe.

3. Results and discussion

3.1 Atmospheric hydrochloric acid leaching from saprolite laterite

3.1.1 Effect of hydrochloric acid concentration on the metal leaching efficiency. The effect of the concentration of hydrochloric acid (2.5–3.0 M) on the metal leaching efficiency from saprolite laterite was investigated, as shown in Fig. 3. The concentration of hydrochloric acid (2.5–3.0 M) had little impact on the leaching efficiencies of Fe and Ni, which were around 96.5% and 98.3%. The leaching efficiencies of Mn, Co, Mg and Al increased from 93.8% to 97.3%, 94.3% to 97.3%, 74.6% to 79.3% and 74.7% to 79.3%, respectively, as the concentration of HCl solution was increased from 2.5 to 3.0 M. The metal leaching efficiencies were closely related to the metal distributions in saprolite laterite and mineral dissolution behaviors. Ni is present in isomorphic lizardite (Mg₃Si₂O₅(OH)₄) and goethite (FeO(OH)), which are easily dissolved in acid. Magnesium-rich chlorites ((Mg,Al)₆Si₄O₁₀(OH)₈) are difficult to dissolve in acid,¹⁸ which is why the metal leaching efficiencies of Mg and Al are not high. Considering the metal leaching efficiency and the subsequent additional amount of sodium hydroxide, the optimal acid concentration of the following leaching experiments was controlled at 2.75 M.

Fig. 3 Effect of the concentration of hydrochloric acid on the metal leaching efficiency (other leaching conditions: liquid–solid ratio, 10 mL g⁻¹; temperature, 100 °C; time, 90 min).

3.1.2 Effect of liquid–solid ratio on the metal leaching efficiency. The effect of the liquid–solid ratio (6–11 mL g⁻¹) on the metal leaching efficiency from saprolite laterite was tested and the results are shown in Fig. 4. It can be observed that the metal leaching efficiencies increased as the liquid–solid ratio increased until the liquid–solid ratio reached 9 mL g⁻¹; furthermore, the leaching efficiency remained constant. These results can be attributed to the fact that the extent of contact between saprolite laterite and hydrochloric acid increased with the increasing liquid–solid ratio. Moreover, the leaching efficiencies of Ni and Fe, Co and Mn, and Al and Mg exhibited a correlation with the liquid–solid ratio, which meant that Ni and Fe, Co and Mn, and Al and Mg were dissolved simultaneously. This can be explained by the metal distributions in saprolite laterite: Ni was mainly isomorphic in Mg₃Si₂O₅(OH)₄ and FeO(OH), Co was mainly embedded in the crystalline structure of high manganese compositions and Al was mainly isomorphic in (Mg,Al)₆Si₄O₁₀(OH)₈. Therefore, the optimal liquid–solid ratio was 9 mL g⁻¹.

Fig. 4 Effect of the liquid–solid ratio on the metal leaching efficiency (other leaching conditions: HCl concentration, 2.75 M; temperature, 100 °C; time, 90 min).

3.1.3 Effect of leaching time on the metal leaching efficiency. The effect of the leaching time on the metal leaching efficiency from saprolite laterite was investigated and the results are presented in Fig. 5. All the metal leaching efficiencies increased rapidly during the initial stage (0–10 min) and increased slowly with the consumption of H⁺ during the second stage (10–30 min). After 30 min, the metal dissolution reactions had nearly reached equilibrium and the maximum leaching efficiencies of Fe, Mg, Ni, Mn, Co and Al reached 96.5%, 78.4%, 98.3%, 96.4%, 97.0% and 77.1%, respectively. These results indicated that a leaching time of 30 min was favorable for the metal leaching.

Fig. 5 Effect of leaching time on the metal leaching efficiency (other leaching conditions: HCl solution concentration, 2.75 M; liquid–solid ratio, 9 mL g⁻¹; temperature, 100 °C).

According to the abovementioned experiments, it can be concluded that the optimum leaching conditions were as follows: an HCl solution concentration of 2.75 M, liquid–solid ratio of 9 mL g⁻¹, leaching time of 30 min and leaching temperature of 100 °C. The chemical analysis results for the leaching residue and the solution are listed in Table 2. The Fe to M (M = Mg, Ni, Mn or Co) molar ratio (R_Fe/M) in the metal leaching solution was calculated to be about 0.878. The main phases in the leaching residue were silica and less amount of chlorite.

Table 2 Chemical analysis results for the acid leaching residue and leaching solution under the optimum leaching conditions (HCl solution concentration, 2.75 M; liquid–solid ratio, 9 mL g⁻¹; leaching time, 30 min and leaching temperature, 100 °C); solubility product constants and pH values are for completely precipitated metal hydroxides at 25 °C ([M] = 10⁻⁶ M)

Element	Leaching residue (wt%)	Leaching solution concentration (g L^−1)	Solubility product constant	pH value of completely precipitated metal hydroxide
Fe	1.46	14.08	4.0 × 10^−38	3.5
Mg	5.69	6.34	1.8 × 10^−11	11.6
Ni	0.05	1.30	2.0 × 10^−15	9.7
Mn	0.017	0.21	1.3 × 10^−13	10.6
Co	0.003	0.034	1.6 × 10^−15	9.6
Al	0.917	0.93	3.0 × 10^−34	5.5

---

3.2 Hydrothermal synthesis of metal-doped magnesium ferrite from the solution (precursor solution)

To maximize the use of metal elements in the solution, a pH value of 12.0 was chosen to realize the complete precipitation of Fe, Mg, Ni, Mn and Co, as shown in Table 2. Thermodynamic equilibrium calculations were performed to investigate the thermodynamic stability of related species under hydrothermal conditions. Considering the different metal contents and the complex system, the Fe–Mg system was considered. Fig. 6 shows the variation in the standard Gibbs free energy¹⁹ (ΔG^o) with temperature for the main reactions ((1)–(6)) during the hydrothermal synthesis process.

2Fe(OH)₃ → Fe₂O₃ + 3H₂O (1)

Mg(OH)₂ → MgO + H₂O (2)

Mg(OH)₂ + 2Fe(OH)₃ → MgFe₂O₄ + 4H₂O (3)

Mg(OH)₂ + Fe₂O₃ → MgFe₂O₄ + H₂O (4)

MgO + 2Fe(OH)₃ → MgFe₂O₄ + 3H₂O (5)

MgO + Fe₂O₃ → MgFe₂O₄ (6)

Fig. 6 Variation of ΔG^o with temperature for different reactions.

From a thermodynamic viewpoint, the reaction may occur when the value of ΔG^o is negative; therefore, the dehydration from Mg(OH)₂ to MgO (reaction (2)) could not occur, resulting in reactions (5) and (6) hardly occurring due to a lack of the reactant MgO in the system. It can be also observed from Fig. 6 that the dehydration from Fe(OH)₃ to Fe₂O₃ (reaction (1)) could occur only at temperatures higher than 153 °C, and as the temperature increased, the dehydration reaction could occur more easily, whereas the reaction between Fe₂O₃ and Mg(OH)₂ to form MgFe₂O₄ (reaction (4)) took place only to a small extent. Therefore, magnesium ferrite could be synthesized mainly by the reaction between Mg(OH)₂ and Fe(OH)₃ (reaction (3)). Thus, the hydrothermal temperature had a major impact on the formation of pure MgFe₂O₄.

3.2.1 Effect of pH value of the precursor solution. Considering the different pH values of completely precipitated metals (Table 2), it is necessary to study the effect of pH of precursor solution on the hydrothermal synthesis of metal-doped MgFe₂O₄.

Fig. 7 shows the XRD patterns of the as-prepared products formed at different pH values. It can be observed from Fig. 7(A) that metal-doped MgFe₂O₄ with poor crystallinity, accompanied by a small amount of layered double hydroxides (LDH), was obtained when the pH value of the precursor solution was controlled at 12.0. When the pH value was decreased to 10.0 and 8.0, ferrite accompanied by Fe₂O₃ appeared, suggesting that the pH value of the precursor solution played a key role in determining the purity of the as-prepared products during the hydrothermal process. To enhance the degree of crystallinity, the products were calcined at 500 °C for 2 h under an air atmosphere. It can be clearly observed from Fig. 7(B) that the dehydration of layered double hydroxides (LDH) occurred and the peak at 2Theta of around 22° disappeared, the crystallinity of all the products after calcination increased considerably and the spinel ferrite was concurrently formed with Fe₂O₃ at pH = 8.0 and 10.0. The iron oxide originating from the dehydration of Fe(OH)₃ and the incomplete precipitation of divalent metal ions (Ni²⁺, Mn²⁺ and Co²⁺) reacted with Fe(OH)₃ at pH = 8.0 and 10.0.

Fig. 7 XRD patterns of the as-prepared products using different precursor solution pH values: (A) direct hydrothermal products, (B) products calcined at 500 °C for 2 h; (other hydrothermal conditions: temperature, 160 °C; time, 6 h).

3.2.2 Effect of hydrothermal temperature. Fig. 8 shows the XRD patterns of the as-prepared products synthesized at different reaction temperatures. It can be observed from Fig. 8(A) that relatively pure ferrite was obtained, accompanied by a small amount of layered double hydroxides (LDH) when the hydrothermal temperature was controlled at 160 °C. After calcination at 500 °C for 2 h, only the ferrite with improved crystallinity was produced, as shown in Fig. 8(B), further confirming that a temperature of 160 °C was beneficial to the formation of pure ferrite. As the hydrothermal temperature increased to 180 °C and 200 °C, the ferrite accompanied by Fe₂O₃ and Mg(OH)₂ appeared and increased accordingly. This phenomenon may be mainly ascribed to a larger quantity of Fe₂O₃ being generated at high temperatures of 180 °C and 200 °C. After calcining the hydrothermal products at 500 °C for 2 h, the diffraction peaks of Mg(OH)₂ disappeared, due to its dissolving into a MgFe₂O₄ solid solution, leaving the other two phases (ferrite and Fe₂O₃) coexisting in the calcined products, as shown in Fig. 8(B).

Fig. 8 XRD patterns of the as-prepared products with different reaction temperatures: (A) direct hydrothermal products, (B) products calcined at 500 °C for 2 h; (other hydrothermal conditions: pH value, 12.0; time, 6 h).

3.2.3 Effect of hydrothermal time. Fig. 9 shows the XRD patterns of the as-prepared products formed after different reaction times. It can be observed from Fig. 9(A) that relatively pure ferrite was obtained, accompanied by only a small amount of layered double hydroxides (LDH) when the reaction time was controlled for 6 h and after calcination at 500 °C for 2 h, only the ferrite with improved crystallinity can be found, as shown in Fig. 9(B). As the time increased to 9 h and 12 h, ferrite accompanied by Fe₂O₃ appeared and increased accordingly. From Fig. 6, it is known that reaction (3), forming ferrite and reaction (1), forming Fe₂O₃, were coexisting and competing reactions. As the hydrothermal time increased, the two reactions ((1) and (3)) would proceed completely, which produced more ferrite and Fe₂O₃. The reaction between Fe₂O₃ and Mg(OH)₂ to form ferrite could occur only to a small extent, based on the analysis of Fig. 6; therefore, ferrite and Fe₂O₃ were both obtained as the time increased to 9 and 12 h.

Fig. 9 XRD patterns of the as-prepared products for different reaction times: (A) direct hydrothermal products, (B) products calcined at 500 °C for 2 h; (other hydrothermal conditions: temperature, 160 °C; pH value, 12.0).

According to the abovementioned hydrothermal synthesis experiments, pure metal-doped MgFe₂O₄ could be synthesized when the pH value of the precursor solution was controlled at 12.0 and the hydrothermal temperature and time were 160 °C and 6 h, respectively. It should be noted that the R_Fe/M of the precursor solution used in the hydrothermal process was 0.878 (the stoichiometric ratio of R_Fe/M for the synthesis of metal-doped MgFe₂O₄ is 2.0). Thus, the amount of magnesium used in the hydrothermal process was larger than that of iron; therefore, it is necessary to analyze the existing form of the excess magnesium in the pure metal-doped MgFe₂O₄.

3.3 Characterization and analysis of the as-prepared metal-doped magnesium ferrite

3.3.1 Chemical analysis. Table 3 gave the chemical analysis results of the hydrothermal products washed with deionized water and diluted acetic acid (5%, vol%). The R_Fe/M of the hydrothermal product washed with diluted acetic acid became larger (1.437) compared with that washed with deionized water (1.042), suggesting that some components were dissolved in acetic acid. Considering that spinel ferrite is very stable in diluted acids and the amount of magnesium was in a greater excess compared with that of iron in the ferrite, it is reasonable to assume that some other magnesium-containing compounds such as magnesium oxide/magnesium hydroxide coexisted in the as-prepared metal-doped magnesium ferrite.

Table 3 Chemical analysis results of the direct hydrothermal products after washing with deionized water (sample a) and diluted acetic acid (sample b)

Element	Fe	Mg	Ni	Mn	Co	Al	RFe/M
Sample a, mol	0.549	0.469	0.048	0.009	0.001	0.066	1.042
Sample b, mol	0.662	0.393	0.056	0.010	0.002	0.076	1.437

---

3.3.2 Raman characterization and analysis. Raman spectroscopy is a powerful probe to reveal the vibrational and structural properties of the materials. Fig. 10 illustrates the room temperature Raman spectra of the products obtained in the range of 100–1000 cm⁻¹. Theories predict the following optical phonon distribution: 5T_1u + A_1g + E_g + 3T_2g. The 5T_1u modes are IR active, whereas the other five (A_1g + E_g + 3T_2g) are Raman-active modes composed to the motion of O ions and both the A-site and B-site ions in the spinel structure.²⁰ The A_1g mode is due to symmetric stretching of the oxygen anion, whereas the E_g mode occurs due to symmetric bending of the oxygen anion, and the T_2g mode is the result of asymmetric stretching of the oxygen anion with respect to the A-site and B-site cations.

Fig. 10 Raman spectra of the metal-doped MgFe₂O₄ ferrites: (a) direct hydrothermal product, (b) product calcined at 500 °C for 2 h, (c) product calcined at 900 °C for 2 h.

From Fig. 10, it can be observed that the spectra can be deconvoluted into seven components at ∼205, ∼280, ∼335, ∼475, ∼550, ∼635 and ∼700 cm⁻¹ for the metal-doped MgFe₂O₄ ferrites (Table 4). For the direct hydrothermal product, it has been observed that a Raman band at around 699 cm⁻¹ shows a shoulder like feature at the lower wavenumber side (632 cm⁻¹) against the reported single band for Fe₃O₄.²¹ These bands were assigned to A_1g(1) and A_1g(2) modes, reflecting the stretching vibrations of the Fe–O and M–O chemical bonds in tetrahedral sites. The other low frequency modes (209, 334, 473, and 550 cm⁻¹) were assigned to the T_2g and E_g modes, reflecting the vibration of the spinel structure. For the calcined ferrites, there were similar situations in comparison with the direct hydrothermal product, only a little shift owing to the ion redistribution between both the sites. Furthermore, by comparing the relative intensities of the A_1g(1) and A_1g(2) modes that represented the quantitative surface compositions, it can be concluded that a number of Mg²⁺ cations transferred from the A-site to B-site (the same number of Fe³⁺ cations that migrated from the B-site to A-site) as the calcination temperature increased. More importantly, apart from these characteristic models for the spinel ferrite, there was also a band at around 283 cm⁻¹ (280 and 278 cm⁻¹ after calcination at different temperatures) that was assigned to the stretching vibration of the Mg–O chemical bond, as reported elsewhere.²² This means that magnesium oxide/magnesium hydroxide indeed existed in the metal-doped magnesium ferrite. Therefore, it is reasonable to assume that Mg in the products existed mainly in the form of Mg–O in tetrahedral and octahedral sites in the spinel structure, as well as trace amounts in the form of Mg–O in magnesium-containing compounds such as magnesium oxide/magnesium hydroxide.

3.3.3 XPS characterization and analysis. To obtain detailed structural information, XPS analysis was used to elucidate the elemental composition and valence state of the three products obtained, as illustrated in Fig. 11. Wide-scan XPS spectra of the products evidenced Fe, Mg and O and other elements, such as Ni, Al, Mn, Si, and C, that are not shown in Fig. 11(A). The high-resolution narrow-scan XPS spectra of the Fe 2p, Mg 2p and O 1s peaks of the metal-doped MgFe₂O₄ ferrites are shown in Fig. 11(B)–(D). In the Fe 2p spectra of the ferrites, there were two nonequivalent Fe 2p_3/2 peaks at ∼710.73 eV, consistent with Fe_B³⁺ ions in octahedral sites and at ∼712.43 eV, consistent with Fe_A³⁺ ions in tetrahedral sites,^23–25 accompanied by satellite lines at ∼714.62 and ∼719.5 eV, indicating only the presence of Fe³⁺ cations.²⁶ However, the contributions at ∼724.32 and ∼726.13 eV were assigned to the chemical state Fe³⁺ of Fe 2p_1/2.²⁷ By calculating the relative intensities of the spectral components for Fe 2p, it can be concluded that a number of Fe³⁺ cations migrated from the B-site to the A-site as the calcination temperature increased.

Fig. 11 XPS spectra of the metal-doped MgFe₂O₄ ferrites: survey of the samples (A) and HRXPS spectra of Fe 2p peak (B), Mg 2p peak (C) and O 1s peak (D); (a) direct hydrothermal product, (b) product calcined at 500 °C for 2 h, (c) product calcined at 900 °C for 2 h.

In the Mg 2p spectra of the ferrites, the signal could be deconvoluted into three peaks at ∼49.4, ∼50.1 and ∼50.35 eV, consistent with different environments of Mg²⁺ ions: octahedral sites (Mg_B²⁺) in the spinel structure, Mg–O sites and tetrahedral sites (Mg_A²⁺) in the spinel structure.^28–30 By calculating the relative intensities of spectral components for Mg 2p, it can be concluded that most of the Mg²⁺ ions occupied the tetrahedral sites in the direct hydrothermal product and a number of Mg²⁺ cations migrated from the A-site to the B-site as the calcination temperature increased (the corresponding number of Fe³⁺ cations from the B-site to the A-site). Apart from the magnesium in the spinel structure, the amount of excess magnesium combining with oxygen to form the Mg–O chemical bond in magnesium-containing compounds was basically unchanged as the calcination temperature increased, as shown by the unchanged relative intensities in Fig. 11(C).

In the O 1s spectra of the ferrites, the signal could be deconvoluted into three peaks, ∼530.05, ∼531.3 and ∼531.96 eV, representing three different types of surface oxygen species: the metal–oxygen bonds, hydroxyl bonded to metal and adsorbed H₂O.³¹ The intensity of the O 1s peak at ∼531.96 eV was high for the direct hydrothermal product of the ferrite because of the multiplicity of physical and chemisorbed water at the surface and reduced as the calcination temperature increased, indicating a decrease in adsorbed H₂O. The intensity of the O 1s peak at ∼530.05 eV increased as the calcination temperature increased, indicating that the relative content of the metal–oxygen compound increased. However, the intensity at 531.3 eV was basically unchanged, which meant that part of the hydroxide coexisted in the ferrite.

3.3.4 Magnetic properties. It is well known that the magnetic properties of ferrites depend on their composition, cation distribution, average size and crystallinity. The chemical composition, crystallinity and size of the ferrite play critical roles in determining the magnetic properties of the nanoparticles and possible applications.^32,33

The magnetic hysteresis loops of the products obtained by vibrating sample magnetometry (VSM) at room temperature are shown in Fig. 12 and the magnetic parameters are summarized in Table 5. Curves obtained for coercive field strength H_C and saturation magnetization M_S values explained the soft ferrimagnetic nature of the metal-doped MgFe₂O₄ ferrites. The saturation value of 45.5 emu g⁻¹ obtained for the ferrite was higher than the values of 33.4 emu g⁻¹ for bulk MgFe₂O₄,³⁴ 21.9 emu g⁻¹ for sol–gel/combustion-synthesized MgFe₂O₄ (ref. 33) or 12.9 emu g⁻¹ for co-precipitation-synthesized MgFe₂O₄.²⁰ The results indicated that the preparation method of ferrite affected its magnetic properties.^35–38 For the product calcined at 500 °C, the saturation magnetization was reduced to 8.5 emu g⁻¹. This phenomenon may be ascribed to the following reasons: the hydrothermally synthesized metal-doped MgFe₂O₄ ferrite may be accompanied by some magnetite or maghemite, which may also contain a small fraction of Mg (Mg/Fe ⋘ 0.5). Both magnetite and maghemite possess much larger magnetization than magnesium ferrite;^39,40 therefore, a high value of M_S (45.5 emu g⁻¹) was obtained. In addition, some of the Mg may be incorporated into the layered double hydroxide (LDH) phase. During the calcination process, the Mg hydroxide reacted with the magnetite and magnesium ferrite (with a larger fraction of Mg) started to form, leading to a decrease in the magnetization of the product calcined at 500 °C. For the product calcined at 900 °C, the saturation magnetization increased to 39.2 emu g⁻¹. The improvement in the crystallinity and the larger crystallite size might be the main reasons for the increased value of M_S.^41,42

Fig. 12 Room temperature hysteresis loops of the metal-doped MgFe₂O₄ ferrites: (a) direct hydrothermal product (*), (b) product calcined at 500 °C for 2 h (*), (c) product calcined at 500 °C for 2 h (#), (d) product calcined at 900 °C for 2 h (*) (other hydrothermal conditions: temperature, 160 °C; time, 6 h; pH value, 12.0). Remark: (*) product washed with deionized water; (#) product washed with diluted acetic acid.

Table 4 Raman parameters of the metal-doped MgFe₂O₄ ferrites

Assignment	Raman modes (cm^−1)
	Direct hydrothermal product	Product calcined at 500 °C for 2 h	Product calcined at 900 °C for 2 h
A1g(1)	699	694	702
A1g(2)	632	625	640
T2g(3)	550	551	563
T2g(2)	473	478	479
Eg	334	337	331
	283	280	278
T2g(1)	209	201	207

---

Table 5 Magnetic parameters (remanent magnetization M_R, saturation magnetization M_S and coercive field strength H_C) of the metal-doped MgFe₂O₄ ferrites. Remark: (*) product washed with deionized water; (#) product washed with diluted acetic acid

Ferrite samples	MR (emu g^−1)	MS (emu g^−1)	HC (Oe)
Direct hydrothermal product (*)	6.8	45.5	65.8
Product calcined at 500 °C for 2 h (*)	0.5	8.5	23.8
Product calcined at 500 °C for 2 h (#)	2.5	15.9	39.1
Product calcined at 900 °C for 2 h (*)	9.7	39.2	108.4

---

The saturation magnetization of the product washed with diluted acetic acid was higher than that of the product washed with deionized water, as evidenced by comparing the magnetic hysteresis loops, as shown in Table 5. Considering that the magnesium-containing compounds are in the non-ferromagnetic phase, which is detrimental to the M_S,^43–45 it can be concluded that there were indeed some other magnesium-containing compounds, such as magnesium oxide/magnesium hydroxide, coexisting in the metal-doped magnesium ferrite.

4. Conclusions

We succeeded in synthesizing metal-doped magnesium ferrite via an atmospheric hydrochloric acid leaching process and a hydrothermal synthesis process from saprolite laterite. As for the leaching process, it is shown that the leaching efficiencies of Fe, Mg, Ni, Mn and Co were 96.5%, 78.4%, 98.3%, 96.4% and 97.0%, respectively, under the optimum conditions: an HCl solution concentration of 2.75 M, a liquid–solid ratio of 9 mL g⁻¹, a leaching time of 30 min and a leaching temperature of 100 °C. As for the hydrothermal synthesis process, the pH value of the leaching solution, the hydrothermal temperature and time played important roles in the formation of metal-doped magnesium ferrite. Furthermore, the metal-doped magnesium ferrite was synthesized at a temperature of 160 °C for 6 h, at a pH value of 12.0 from a leaching solution with an R_Fe/M of 0.878. From the structural analysis of the products, it can be concluded that most of the magnesium ions occupied tetrahedral A-sites in the direct hydrothermal product and migrated from the tetrahedral A-site to the octahedral B-site as the calcination temperature increased. There were indeed some other magnesium-containing compounds, such as magnesium oxide/magnesium hydroxide, coexisting in the metal-doped magnesium ferrite.

Acknowledgements

The study was financially supported by the National Science Foundation of China (No. 51272025 and 51072022) and the National Basic Research Program of China (No. 2014CB643401, 2013AA032003).

References

1. Q. Chen and Z. J. Zhang, Appl. Phys. Lett., 1998, 73, 3156.
2. E. P. Naiden, V. A. Zhuravlev, V. I. Itin, O. G. Terekhova, A. A. Magaeva and Y. F. Ivanov, Russ. Phys. J., 2006, 49, 946.
3. Y. H. Lee, G. D. Lee, S. S. Park and S. S. Hong, React. Kinet. Catal. Lett., 2005, 84, 311.
4. K. Mukherjee, D. C. Bharti and S. B. Majumder, Sens. Actuators, B, 2010, 146, 91.
5. S. Permien, S. Indris and M. Scheuermann, J. Mater. Chem. A, 2015, 3, 1549.
6. R. J. Willey, P. Noirclerc and G. Busca, Chem. Eng. Commun., 1993, 123, 1.
7. M. M. Hessien, Z. I. Zaki and Q. Mohsen, Int. J. Mater. Sci., 2011, 1, 30.
8. N. Sivakumar, A. Narayanasamy, J. M. Greneche, R. Murugaraj and Y. S. Leed, J. Alloys Compd., 2010, 504, 395.
9. H. Aono, H. Hirazawa, T. Naohara and T. Maehara, Appl. Surf. Sci., 2008, 254, 2319.
10. Y. J. Huang, Y. Tang, J. Wang and Q. W. Chen, Mater. Chem. Phys., 2006, 97, 394.
11. D. Gherca, A. Pui, V. Nica, O. Caltun and N. Cornei, Ceram. Int., 2014, 40, 9599.
12. A. X. Shan, X. Wu, J. Lu, C. P. Chen and R. M. Wang, CrystEngComm, 2015, 17, 1603.
13. H. L. Andersen and M. Christensen, Nanoscale, 2015, 7, 3481.
14. Y. Shen, Y. B. Wu, X. Y. Li, Q. Zhao and Y. Hou, Mater. Lett., 2013, 96, 85.
15. T. Sasaki, S. Ohara, T. Naka and J. Vejpravova, J. Supercrit. Fluids, 2010, 53, 92.
16. B. Harris, C. White, M. Jansen and D. Pursell, ALTA 2006 Nickel/Cobalt 11, ALTA Metallurgical Services, Melbourne, 2006, p. 20.
17. D. Neudorf, ALTA 2007 Nickel/Cobalt 12, ALTA Metallurgical Services, Melbourne, 2007, p. 5.
18. S. F. Zhang, Mineralogy, Commercial Press, Shanghai, 1954, vol. 5, ch. 7, pp. 720–721.
19. Y. J. Liang and Y. C. Che, Inorganic Thermodynamic Data Sheet, Northeastern University Press, Shenyang, 1993, vol. 2, ch. 3, p. 458.
20. Z. W. Wang, P. Lazor, S. K. Saxena and H. S. C. O'Neill, Mater. Res. Bull., 2002, 37, 1589.
21. R. Gupta, A. K. Sood, P. Metcalf and J. M. Honig, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 65, 104430.
22. K. Ishikawa, N. Fujima and H. Komura, J. Appl. Phys., 1985, 57, 973.
23. F. Y. Ran, Y. Tsunemaru, T. Hasegawa, Y. Takeichi, A. Harasawa, K. Yaji, S. Kim and A. Kakizaki, J. Appl. Phys., 2011, 109, 123919.
24. Z. Zhou, Y. Zhang, Z. Wang, W. Wei, W. Tang, J. Shi and R. Xiong, Appl. Surf. Sci., 2008, 254, 6972.
25. F. Tudorache, P. D. Popa, M. Dobromir and F. Iacomi, Mater. Sci. Eng., B, 2013, 178, 1334.
26. F. Li, X. F. Liu, Q. Z. Yang, J. J. Liu, D. G. Evans and X. Duan, Mater. Res. Bull., 2005, 40, 1244.
27. A. Miyakoshi, A. Ueno and M. Ichikawa, Appl. Catal., A, 2001, 219, 249.
28. C. Liu, B. S. Zou, A. J. Rondinone and Z. J. Zhang, J. Am. Chem. Soc., 2000, 122, 6263.
29. S. Ardizzone, C. L. Bianchi, M. Fadoni and B. Vercelli, Appl. Surf. Sci., 1997, 119, 253.
30. H. Zhang, R. Qi, D. G. Evans and X. Duan, J. Solid State Chem., 2004, 177, 772.
31. L. Zhang, J. Zhu, X. Jian, D. G. Evans and F. Li, J. Phys. Chem. Solids, 2006, 67, 1678.
32. P. Laokul, V. Amornkitbamrung, S. Seraphin and S. Maensiri, Curr. Appl. Phys., 2011, 11, 101.
33. A. Pradeep, P. Priyadharsini and G. Chandrasekaran, J. Magn. Magn. Mater., 2008, 320, 2774.
34. V. Sepelak, I. Bergmann, D. Menzel, A. Feldhoff, P. Heitjans, F. J. Litterst and K. D. Becker, J. Magn. Magn. Mater., 2007, 316, 764.
35. S. Kanagesan, M. Hashim, S. Tamilselvan, N. B. Alitheen, I. Ismail and G. Bahmanrokh, J. Nanomater., 2013, 2013, 865024.
36. S. Phumying, S. Labuayai, E. Swatsitang, V. Amornkitbamrung and S. Maensiri, Mater. Res. Bull., 2013, 48, 2060.
37. B. Antic, N. Jovic, M. B. Pavlovic, A. Kremenovic, D. Manojlović, M. Vucinic-Vasic and A. S. Nikolić, J. Appl. Phys., 2010, 107, 043525.
38. D. Chen, D. Y. Li, Y. Z. Zhang and Z. T. Kang, Ultrason. Sonochem., 2013, 20, 1337.
39. B. Y. Yu and S. Y. Kwak, J. Mater. Chem., 2010, 20, 8320.
40. S. J. Iyengar, M. Joy, C. K. Ghosh, S. Dey, R. K. Kotnala and S. Ghosh, RSC Adv., 2014, 4, 64919.
41. J. Wang, J. Sun, Q. Sun and Q. Chen, Mater. Res. Bull., 2003, 38, 1113.
42. S. Maensiri, C. Masingboon, B. Banjong and S. Seraphin, Scr. Mater., 2007, 56, 797.
43. M. G. Ferreirada Silva and M. A. Valente, Mater. Chem. Phys., 2012, 132, 264.
44. W. S. Tang, Y. Su, Q. Li, S. A. Gao and J. K. Shang, RSC Adv., 2013, 3, 13961.
45. Z. Gao, J. Zhou, F. M. Cui, Y. Zhu, Z. L. Hua and J. L. Shi, Dalton Trans., 2010, 39, 11132.

---