Reduction kinetics of hematite to magnetite under hydrothermal treatments

Abstract

The phase transformation of α-Fe₂O₃ to Fe₃O₄ was observed by the hydrothermal treatment of a ferric solution at 160–220 °C with the addition of both KOH and EDA into the reaction system. The reactions began with the formation of α-Fe₂O₃ hexagonal plates followed by the phase transformation involving the dissolution of the α-Fe₂O₃ hexagonal platelets, the reduction of Fe³⁺ to Fe²⁺, and the nucleation and growth of new Fe₃O₄ polyhedral particles. The activation energies for the phase transformation of α-Fe₂O₃ to Fe₃O₄ under hydrothermal conditions are estimated to be 96.4 ± 11, 113.1 ± 5, and 118.3 ± 11 kJ mol⁻¹ for the addition of 0.5, 1 and 1.5 ml of EDA, respectively, which are about the same for the typical phase transformation of α-Fe₂O₃ to Fe₃O₄ in a hydrogen atmosphere.

---

1. Introduction

The more stable phases of iron oxides are hematite (α-Fe₂O₃) and magnetite (Fe₃O₄). Hematite can be used in a lot of applications such as sensors,¹ water photooxidation,² drug delivery,³ lithium ion battery,⁴ pigmentation,⁵ and solar cell,⁶ and the application of magnetite can be found in biomedicine,^7–11 magnetic devices,¹² etc. Thus, studies about the nano/microstructures of iron oxides and their properties, which are related to the intrinsic structure and crystal shapes, have been intensively engaged, particularly for hematite and magnetite. Many researches have demonstrated the capability of using chemical syntheses to control the particle morphologies of the iron oxide by surfactants.^13–26 Morphologies, such as wires,¹⁶ rods,¹⁷ tubes,¹⁸ rings,¹⁹ disks,²⁰ cubes,²¹ spheres,²² hexagonal platelets,^23,24 of α-Fe₂O₃ and polyhedral particles of Fe₃O₄ (ref. 25 and 26) have been synthesized successfully.

The phase transformation between different iron oxides is technologically important due to the fact that the process provides alternative routes to produce unusual particle shapes for some phases.²⁷ The phase transformation between hematite and magnetite is also a topic of geological interest.²⁸

Several reaction routes were reported for the phase transformation between hematite and magnetite. The first route is the non-redox transformation of iron oxides, which was made possible by adding Fe²⁺ into solution to form Fe₃O₄ or removing Fe²⁺ out of the solution to reverse to α-Fe₂O₃.²⁹ The second route is the redox reaction of iron oxides under oxidizing (oxygen) or reducing (hydrogen) condition.^27,30–36 Viswanath et al. investigated the kinetics of the reduction of Fe₂O₃ to Fe₃O₄ by the isothermal differential thermal analysis method.³⁰ Lin et al. used a temperature-programmed reduction method to obtain the activation energy for the transformation of Fe₂O₃ to Fe₃O₄ in H₂ containing atmosphere, and the activation energy was reported to be 89.13 kJ mol⁻¹.³¹ Wang et al. used thermal gravimetric analysis (TGA) to obtain an activation energy of 152.44 kJ mol⁻¹ for the transformation. The relatively large activation energy was attributed to the high density of the particle structure.³² Tiernan et al. investigated the reduction of iron oxide and obtained its activation energy by constant rate thermal analysis (CRTA), and the value of the activation energy for the Fe₂O₃ to Fe₃O₄ transformation was 96 kJ mol⁻¹.³³

Yanagisawa et al. also showed that by controlling the mineralizer solutions, temperatures, and partial pressures of hydrogen in a hydrothermal system, phase transformation from α-Fe₂O₃ to Fe₃O₄ particles can be achieved, However, the activation energy was not reported.³⁷ The third route is the redox reaction of iron oxide in hydrothermal conditions with a reducing agent such as sodium sulfite and diethylene glycol.^38,39 Sapieszko et al. observed a similar phase transformation from α-Fe₂O₃ hexagonal platelets to octahedral Fe₃O₄ particles triggered by the addition of hydrazine, which was used as an anti-oxidant during the hydrothermal process.⁴⁰ However, no details of the transformation process were discussed. The fourth route is the mechanochemical reaction, which involves ball-milling α-Fe₂O₃ with the addition of other oxides that decompose and assist the transformation of α-Fe₂O₃ to Fe₃O₄.⁴¹

Most of the phase transformation kinetics of hematite and magnetite were mainly conducted with solid iron oxides in gas ambient containing H₂ molecules. In this experiment, we explored the phase transformation kinetics of hematite and magnetite in a solution environment containing ethylenediamine (EDA or en in ligand form) in alkaline solution under hydrothermal conditions. The role of different ions in the reaction system was explored, and the activation energies for the transformation in a solution environment are reported for the first time.

2. Experimental methods

Ferric nitrate (Fe(NO₃)₃·9H₂O) (SHOWA, 99%), 1 mmol, was dissolved in 10 ml of distilled water to form a transparent yellow solution. Next, three different mineralizing agents were added to the ferric solution. The first mineralizing agent was a 10.67 M potassium hydroxide (KOH) or sodium hydroxide (NaOH) aqueous solution, 5 ml, which was added dropwisely into the ferric solution. The second one was 0.5 ml of EDA (TEDIA, 99%) which was added gradually to the ferric solution. The third type of mineralizing agent was the combination of KOH and EDA. The 10.67 M KOH solution, 5 ml, was added first followed by the addition of 0.5, 1, or 1.5 ml of EDA. After adding these mineralizing agents, brown Fe(OH)₃ or FeOOH suspensions were obtained. Then, these solutions were stirred for 30 min before transferring the mixture into a Teflon-lined stainless steel autoclave of 40 ml capacity, followed by heat treatments at 160 °C to 220 °C for different times. After that, the autoclave was cooled down to room temperature in air. The precipitates were collected by centrifugation, washed with deionized water and ethanol several times to remove organic and impurities, and finally dried in air at 80 °C for 12 h.

The as-synthesized powder was characterized by X-ray diffraction (XRD, BRUKER, D2) with Cu-Kα radiation, field emission scanning electron microscopy (JOEL, 6500 FESEM), transmission electron microscopy (TEM) (JOEL, TEM-2010), and Raman spectroscopy (Horiba Jobin Yvon, LABRAM HR 800 UV).

3. Results and discussions

Fig. 1 shows the typical SEM and TEM images of the iron oxide particles that were synthesized under the hydrothermal condition of 200 °C for 9 h in the ferric solution with three different mineralizing agents, KOH, EDA, and KOH/EDA. Fig. 1(a) shows the α-Fe₂O₃ hexagonal plates, which were obtained with the addition of KOH. The plates have an average size of about 10 μm in edge-length and about 500 nm in thickness. In Fig. 1(b), the hexagonal plate has the zone axis of [0001] and the six directions normal to the edge are 〈21̄1̄0〉 and its other five equivalent directions. Fig. 1(c) shows the α-Fe₂O₃ hexagonal bipyramid particles obtained when EDA was added into the system. The average lateral size of the α-Fe₂O₃ particles is about 120 nm. In Fig. 1(d), the hexagonal bipyramid shows that the pyramid is pointed in the direction of 〈0001〉. According to previous studies, the bipyramidal structure is enclosed by {101̄1} crystal planes.⁴² Fig. 1(e) shows the Fe₃O₄ polyhedral particles obtained with the addition of both KOH and 1 ml of EDA into the reaction system. The Fe₃O₄ polyhedral particles, mainly octahedral in shape, have an average lateral size in the range of 5–25 μm. In Fig. 1(f), the Fe₃O₄ polyhedral particles are composed of a pure magnetite phase and the diffraction spots are identified to be the (202), (02̄2), (2̄2̄0) planes and their equivalent planes under an incident electron beam along [1̄11].

Fig. 1 SEM and TEM images of iron oxide particles prepared with the addition of (a and b) 5 ml of 10.67 M KOH, (c and d) 1 ml of EDA, and (e and f) both 5 ml of 10.67 M KOH and 1 ml of EDA to the ferric solutions. All samples were synthesized at 200 °C for 9 hours. The insets in TEM images are the corresponding diffraction patterns of the particles.

The crystal structures of these iron oxide particles were analyzed by XRD as shown in Fig. 2(a). The phase was identified to be α-Fe₂O₃ when either KOH or EDA alone was added to the reaction system, despite the different morphologies. The diffraction peaks match the JCPDS card no. 33-0664, which is a rhombohedral structure with the space group R3̄c. The diffraction peaks of the polyhedral particles, which were obtained with the addition of both KOH and EDA into the reaction system, corresponds to the phase of Fe₃O₄, JCPDS card no. 19-0629, which is a face-centered cubic structure with the space group Fd3̄m. The characteristic reflections in Fe₃O₄ phase and γ-Fe₂O₃ phase are about the same.⁴³ Here, the diffraction peaks of the (221), (210) and (213) planes for the γ-Fe₂O₃ phase does not exist. To further clarify the phase of the polyhedral particles, the Raman spectra of the α-Fe₂O₃ hexagonal plates and Fe₃O₄ polyhedral particles are shown in Fig. 2(b). α-Fe₂O₃ can be characterized by the peaks at around 225, 247, 299, 412, 497, 613, and 1320 cm⁻¹. The peaks at 306, 538 and 668 cm⁻¹ are attributed to Fe₃O₄. For γ-Fe₂O₃ particles, Raman peaks should be observed at around 350, 500, 700, 1400, and 1560 cm⁻¹.^44–46 The lack of Raman peaks at the high wavenumbers (>1000 cm⁻¹) for the bottom part of Fig. 2(b) clearly indicates that the particles are Fe₃O₄. The appearance of the Fe₃O₄ phase during reaction is a clear evidence that the valence change from Fe³⁺ to Fe²⁺ during the reaction must occur due to the fact that Fe²⁺ ions are required to occupy the octahedral sites of Fe₃O₄. It is interesting to know that the particles obtained from the reaction systems with the addition of KOH and EDA alone have the same phase but difference shapes. One would assume that the reaction system with the addition of both KOH and EDA would produce particles with maybe different shapes but still maintain the phase of α-Fe₂O₃. However, the results show that the particles that we obtained have a different phase, Fe₃O₄.

Fig. 2 (a) The corresponding XRD patterns for the samples shown in Fig. 1. (b) The Raman spectra of α-Fe₂O₃ micro-plates obtained with the KOH mineralizing agent and Fe₃O₄ particles obtained with both KOH and EDA mineralizing agents.

The phase transformation process of α-Fe₂O₃ to the Fe₃O₄ was further investigated by varying the heat treatment times of the reaction systems with the addition of both KOH and EDA and the hydrothermal temperature still fixed at 200 °C. Fig. 3(a) shows that after 2 h of growth, the main phase of the particles is α-Fe₂O₃ hexagonal plates. The edge of the hexagonal plate is not as straight as that obtained for the reaction system with KOH only. As the reaction times increased to 3, 5, and 7 h, as shown in Fig. 3(b)–(d), respectively, one can observe that the original hexagonal plate started to dissolve and become smaller and evolved to plates with an irregular shape. At the same time, small octahedron particles appeared and gradually became larger as the reaction time increased. At the reaction time of 8 h, as shown in Fig. 3(e), the observed particles are mainly polyhedron in shape and small amount of plate-like particles still existed. Finally, all particles were in the form of polyhedron when the reaction time was increased to 9 h, as shown in Fig. 3(f). In order to check whether the Fe₃O₄ particles could be directly transformed by the dehydration of iron hydroxide, we also examined the end product of iron hydroxides with only EDA or only KOH solution added and hydrothermally treated at the same conditions. Both cases produced the α-Fe₂O₃ phase only. Hence, the result indicated that the reaction processes involved four steps: (1) the reaction systems rapidly transformed Fe(OH)₃ or FeOOH to α-Fe₂O₃ hexagonal plates under the hydrothermal conditions, (2) the α-Fe₂O₃ hexagonal plates dissolved gradually, (3) the reduction process caused valence transition from Fe³⁺ to Fe²⁺, and (4) the Fe₃O₄ particles started to nucleate then finally grew to form polyhedral particles.

Fig. 3 Mixture of α-Fe₂O₃ and Fe₃O₄ particles obtained under hydrothermal treatment with both 5 ml of 10.67 M KOH and 1 ml of EDA added into the ferric solutions at 200 °C for (a) 2 h, (b) 3 h, (c) 5 h, (d) 7 h, (e) 8 h, and (f) 9 h.

In order to understand the role of K⁺ ions and NO₃⁻ ions on the phase transformation process, the precursor of Fe(NO₃)₃ was replaced by FeCl₃ to check the effect of NO₃⁻ on the reaction process. Moreover, the KOH mineralizing agent was substituted by NaOH with the same molarity to examine its effect on the reaction process. The hydrothermal conditions were maintained the same at 200 °C for 9 h. Four cases were investigated: the first and second cases were solutions with the FeCl₃ precursor and the addition of either KOH only or both KOH and EDA (1 ml). The third and fourth cases were solutions with the Fe(NO₃)₃ precursor and the addition of either NaOH only or both NaOH and EDA (1 ml). Fig. 4(a) shows that the α-Fe₂O₃ hexagonal plates were obtained when the reaction system consisted of FeCl₃ and KOH. The phase transformation from α-Fe₂O₃ hexagonal plates to Fe₃O₄ polyhedral particles still occurred when the reaction system consisted of FeCl₃, KOH, and EDA, as shown in Fig. 4(b). The shape of the polyhedral particles is more irregular in this case. Fig. 4(c) shows that the irregular α-Fe₂O₃ plates were formed when the reaction system constituted of Fe(NO₃)₃ and NaOH. The phase transformation from irregular α-Fe₂O₃ plates to Fe₃O₄ polyhedral particles still occurred with the addition of EDA into the reaction system, as shown in Fig. 4(d). The XRD patterns, shown in Fig. 4(e), confirmed the related phases. It can be seen that the α-Fe₂O₃ plates were not completely reduced to Fe₃O₄ particles in the case of the reaction system containing both NaOH and EDA. Thus K⁺ ions and NO₃⁻ ions are not directly involved in the reduction process of Fe³⁺ to Fe²⁺. However, the transformation process is faster with the presence of KOH in the reaction system than that with NaOH, which might be due to the pH values. The pH value is 11.02 for EDA, 13.17 for NaOH and EDA, and 15.26 for KOH and EDA in the reaction systems. The KOH has the fastest phase transformation rate followed by the NaOH. No phase transformation was observed for pure EDA within the limited reaction time used in the experiment. This phenomenon indicates that the pH value may affect the kinetics of the reduction process of Fe³⁺ to Fe²⁺. Moreover, the transformation process is faster with the presence of NO₃⁻ ions in the reaction system than that with Cl⁻ ions. The role that NO₃⁻ ions played in the phase transformation was explored by Lu et al.,⁴⁷ and they found that there exist a certain amount of HNO₃ that could induce the maximum rate for phase transformation. The slow phase transformation rate observed for small and large amounts of HNO₃ may be attributed to the limiting dissolution of α-Fe₂O₃ and the NO₃⁻ ions can be an oxidant to further suppress the reduction of iron ions in system.

Fig. 4 SEM images of iron oxide particles formed with (a) FeCl₃ + KOH, (b) FeCl₃ + KOH + EDA, (c) Fe(NO₃)₃ + NaOH, (d) Fe(NO₃)₃ + NaOH + EDA. (e) The corresponding XRD patterns for iron oxides obtained for the cases of (a)–(d). All samples were synthesized at 200 °C for 9 hours.

To investigate the role of pH values, first, different amounts of HNO₃ were added into the solution with the Fe(NO₃)₃ precursor, KOH and EDA under hydrothermal conditions at 200 °C for 9 h, as shown in Fig. 5(a). The phase transformation rates decrease gradually when the amount of HNO₃ increases from 0 to 0.3 ml and the pH values decrease from 15.26 to 15.71. Second, we used 9 mg of the pre-synthesized α-Fe₂O₃ hexagonal plates and added HNO₃, 0.19 ml, to match the concentration of NO₃⁻ provided by the Fe(NO₃)₃ precursor in the previous experiments, and 10.67 M KOH with different amounts of EDA, ranging from 0.5 to 2.0 ml, heat-treated at 200 °C for 7 h, as shown in Fig. 5(b). The pH values varied from 15.38 for 0.5 ml EDA to 15.16 for 2.0 ml. The results show that the phase transition rate is fast when the solution contains a large amount of EDA, although the pH value decreases slightly from 15.38 to 15.12. Third, we used 9 ml of the pre-synthesized α-Fe₂O₃ hexagonal plates and added 0.19 ml of HNO₃ and 1 ml of EDA with different molalities of KOH and heated to 200 °C for 7 h. As shown in Fig. 5(c), the results show that the phase transition rates are faster when the solution contains a higher molality of KOH. The pH value changes from 14.39 to 15.33 as the molality of KOH in the solution changes from 5.33 M to 10.67 M. If one examines the pH values for the cases of 0.5 HNO₃ in Fig. 5(a) and 5.33 M KOH in Fig. 5(c), the phase transformation rate became very slow, even slower than the solution with the Fe(NO₃)₃ precursor, NaOH and EDA, which has even lower pH value. Thus, this phenomenon simply explained by pH value dependence is too simple. Rather, the results showed that hydroxide ions are a necessary ingredient for the phase transformation. The reason that pure EDA does not trigger the phase transformation is probably due to the fact that there are no hydroxide ions in the reaction system. The hydroxide ions can react with the hydrogen atoms of EDA and allow the Fe ion to be chelated with the EDA ligand and the reduction occurs. Thus, in a solution with fixed ion species, the more the hydroxide ions, the faster the phase transformation is a suitable explanation. However, when the species of the ions in a solution changes (e.g. K⁺ is changed to Na⁺), the interactions between all the species are changed as well. As we know, even α-Fe₂O₃ formation using the Fe(NO₃)₃ precursor and NaOH or Fe(NO₃)₃ and KOH has different rates and shapes. Although increasing the amount of EDA decreases the pH value, which should decrease the transformation rate, more EDA provides more reaction sites for the reduction process. Therefore, the overall transformation rate still increases.

Fig. 5 XRD patterns of iron oxide particles formed with the reaction systems (a) Fe(NO₃)₃ precursor, different amounts of HNO₃, 10.67 M of KOH and 1 ml of EDA under hydrothermal treatment at 200 °C for 9 h, (b) different amounts of EDA, 0.192 ml of HNO₃, and 10.67 M of KOH added to 9 mg of pre-synthesized α-Fe₂O₃, under hydrothermal treatment at 200 °C for 7 h, and (c) different amounts of KOH, 0.192 ml of HNO₃, and 1 ml of EDA added to 9 mg of the pre-synthesized α-Fe₂O₃, under hydrothermal treatment at 200 °C for 7 h.

Fig. 6(a) shows the curves of the transformed fraction of magnetite (Fe₃O₄) as a function of reaction time. Using the Avrami equation, 1 − exp(−ktⁿ), where k is the reaction constant, t is the reaction time, and n is the exponent of the reaction, one can fit, relatively well, the corresponding experimental data of the magnetite fraction obtained by the hydrothermal treatment with 1 ml EDA at 200 °C for different times. In this case, the entire process includes the time required for the appearance of α-Fe₂O_3. The time required for the initial appearance of α-Fe₂O₃ hexagonal plates was about 3600 s, and α-Fe₂O₃ hexagonal plates were produced completely at about 7200 s under hydrothermal treatment with 1 ml of EDA at 200 °C. Thus the appearance of α-Fe₂O₃ is a much faster process than the phase transformation process that follows. The curve could be fitted by setting 7200 s as 0 s for the beginning of the phase transformation, as shown in Fig. 6(b). The fitted values of n are close to 2 and 3. The fraction of α-Fe₂O₃ and Fe₃O₄ were determined by XRD measurements in conjunction with the Rietveld method, as shown in Fig. 6(c). One can further acquire sigmoid-shaped curves at different temperatures and estimate the activation energies for the phase transformation in the reaction conditions with 0.5, 1, and 1.5 ml of EDA, which is shown in Fig. 7.

Fig. 6 The fraction of magnetite transformed as a function of reaction time (a) including and (b) excluding the time required for the completion of α-Fe₂O₃ for Fe(NO₃)₃ with KOH, and 1 ml of EDA under hydrothermal treatment at 200 °C. (c) The corresponding Rietveld refinements on XRD patterns to determine the fraction of Fe₂O₃ and Fe₃O₄.

Fig. 7 The fraction of magnetite versus reduction time under hydrothermal treatment at different temperatures with (a) 0.5 ml, (b) 1 ml, and (c) 1.5 ml of EDA. The solid lines are the fitting curves to the data using the Avrami equation.

The values of the apparent activation energy for the reduction of hematite to magnetite were estimated based on the Arrhenius equation, which is the following:

where τ is the time at which 50 percent of the phase transformation was achieved, R is the gas constant (8.37 J mol⁻¹ K⁻¹), and Q_r is the activation energy for the phase transformation. The plots of ln(1/τ) versus 1/T are shown in Fig. 8 for cases with different amounts of EDA. The slopes were then used to estimate the apparent activation energy. The values of activation energy extracted are 96.4 ± 11, 113.1 ± 5, and 118.3 ± 11 kJ mol⁻¹ under reaction conditions with 0.5, 1 and 1.5 ml of EDA, respectively. We also checked the activation energies extracted from the fittings without subtracting the time required for the completion of the formation of α-Fe₂O₃ hexagonal plates and found them to be 89.6 ± 9, 87.1 ± 5, and 101 ± 9 kJ mol⁻¹ under reaction conditions with 0.5, 1 and 1.5 ml of EDA, respectively. The values are lower than those obtained after subtracting the time required for the α-Fe₂O₃ formation. With increasing amount of EDA in the reaction system, the phase transformation rate increases significantly. At the hydrothermal temperature of 200 °C, the time required for 50% transformation is 61 450 s for 0.5 ml of EDA and 8102 s for 1.5 ml of EDA. However, the activation energies do not change significantly. This further confirms that increasing the amount of EDA ligand mainly increases the available reaction sites for the reduction of iron ions. However, detail mechanism is not affected.

Fig. 8 Arrhenius plots for 50% reduction of hematite to magnetite with (a) 0.5 ml, (b) 1 ml, and (c) 1.5 ml of EDA. (d) The plot of activation energies obtained from (a)–(c) versus the amount of EDA in the reaction system.

The values of the activation energy obtained in this work, and other results reported previously, which were based on kinetics studies within H₂ containing ambient, are listed in Table 1. Because the activation energies for the α-Fe₂O₃ to Fe₃O₄ transformation obtained in this experiment are close to those obtained for the typical phase transformation in H₂ containing ambient, one could conjecture that H₂ might be produced locally during the hydrothermal treatment, which triggers the valence change from Fe³⁺ to Fe²⁺. With all these different environments, different locations for nuclei formation and different growth dimensions, the activation energies obtained for the transformation are all roughly the same. This may indicate that the slowest kinetic process that controlled the transformation is actually the reduction process of Fe³⁺ to Fe²⁺. Tiernan et al. concluded from the shape of the CRTA profile, that nucleation and diffusion are not the rate-controlling process for the reduction step of Fe₂O₃ to Fe₃O₄.³³

Table 1 Summary of transformation models and activation energies for the α-Fe₂O₃ to Fe₃O₄ transformation reported previously and the present result

References	Nucleation model	Qr (kJ mol^−1)	Method
Lin et al.^31	Unimolecular reaction	89.13	Linear heating rate ∼ 300 °C
Wang et al.^32	Formation and growth of nuclei (2D)	152.44	Isothermal 290–310 °C
Tiernan et al.^33	Phase boundary	96	CRTA rate jump 100–600 °C
Wang et al.^34	Phase-boundary-controlled reaction (3D)	107	250–350 °C
Shimokawabe et al.^35	Formation and growth of nuclei (1D)	74–117	Linear heating rate 500–1200 °C
Wimmers et al.^36	Formation and growth of nuclei	124	Linear heating rate 287–417 °C
This work	Formation and growth of nuclei (3D)	96–118	Isothermal 160–220 °C

---

4. Conclusions

α-Fe₂O₃ nano/micro hexagonal platelets can be successfully reduced to octahedral Fe₃O₄ particles with EDA in alkaline solution under a low temperature hydrothermal process. The entire reaction consists of two stages. The first stage is the rapid formation of α-Fe₂O₃ hexagonal plate, triggered by KOH, followed by the phase transformation stage, which includes three steps: (1) the dissolution of the α-Fe₂O₃ hexagonal platelets, (2) the reduction of Fe³⁺ to Fe²⁺, and (3) the nucleation and growth of new Fe₃O₄ polyhedral particles. The Avrami equation can be used to describe the transformation kinetics and the activation energies obtained are 96.4 ± 11, 113.1 ± 5, and 118.3 ± 11 kJ mol⁻¹ for the reaction systems with the addition of 0.5, 1, and 1.5 ml of EDA, respectively. This is the first reported activation energy for the hematite to magnetite phase transformation in hydrothermal conditions, and the values are right in the range of those obtained from the typical solid reduction in H₂ containing environment.

Acknowledgements

The authors acknowledge the support from National Science Council through grant no. 101-2221-E-007-061-MY2.

Notes and references

1. Y. Wang, J. Cao, S. Wang, X. Guo, J. Zhang, H. Xia, S. Zhang and S. Wu, Facile Synthesis of Porous α-Fe₂O₃ Nanorods and Their Application in Ethanol Sensors, J. Phys. Chem. C, 2008, 112, 17804–17808.
2. F. L. Souza, K. P. Lopes, E. Longo and E. R. Leite, The Influence of the Film Thickness of Nanostructured α-Fe₂O₃ on Water Photooxidation, Phys. Chem. Chem. Phys., 2009, 11, 1215–1219.
3. P. C. Wu, W. S. Wang, Y. T. Huang, H. S. Sheu, Y. W. Lo, T. L. Tsai, D. B. Shieh and C. S. Yeh, Porous Iron Oxide Based Nanorods Developed as Delivery Nanocapsules, Chem.–Eur. J., 2007, 13, 3878–3885.
4. Y. Zou, J. Kan and Y. Wang, Fe₂O₃-Graphene Rice-on-Sheet Nanocomposite for High and Fast Lithium Ion Storage, J. Phys. Chem. C, 2011, 115, 20747–20753.
5. F. Z. Dong, D. S. Ling, J. J. Chun, G. Y. Zheng, P. Y. Li and H. Y. Chun, Hierarchical Assembly of SnO₂ Nanorod Arrays on α-Fe₂O₃ Nanotubes: A Case of Interfacial Lattice Compatibility, J. Am. Chem. Soc., 2005, 127, 13492–13493.
6. S. M. Reda, Synthesis of ZnO and Fe₂O₃ Nanoparticles by Sol–Gel Method and Their Application in Dye-Sensitized Solar Cells, Mater. Sci. Semicond. Process., 2010, 13, 417–425.
7. S. Zhang, X. Chen, C. Gu, Y. Zhang, J. Xu, Z. Bian, D. Yang and N. Gu, The Effect of Iron Oxide Magnetic Nanoparticles on Smooth Muscle Cells, Nanoscale Res. Lett., 2009, 4, 70–77.
8. M. Kallumadil, M. Tada, T. Nakagawa, M. Abe, P. Southern and Q. A. Pankhurst, Suitability of Commercial Colloids for Magnetic Hyperthermia, J. Magn. Magn. Mater., 2009, 321, 1509–1513.
9. D. Thapa, V. R. Palkar, M. B. Kurup and S. K. Malik, Properties of Magnetite Nanoparticles Synthesized Through a Novel Chemical Route, Mater. Lett., 2004, 58, 2692–2694.
10. D. Zhang, Z. Liu, S. Han, C. Li, B. Lei, M. P. Stewart, J. M. Tour and C. Zhou, Magnetite (Fe₃O₄) Core–Shell Nanowires: Synthesis and Magnetoresistance, Nano Lett., 2004, 4, 2151–2155.
11. M. K. Yu, Y. Y. Jeong, J. Park, S. Park, J. W. Kim, J. J. Min, K. Kim and S. Jon, Drug-Loaded Superparamagnetic Iron Oxide Nanoparticles for Combined Cancer Imaging and Therapy In Vivo, Angew. Chem., Int. Ed., 2008, 47, 5362–5365.
12. H. Zeng, J. Li, J. P. Liu, Z. L. Wang and S. Sun, Exchange-Coupled Nanocomposite Magnets by Nanoparticle Self-Assembly, Nature, 2002, 420, 395–398.
13. G. Y. Zhang, Y. Y. Xu, D. Z. Gao and Y. Q. Sun, α-Fe₂O₃ Nanoplates: PEG-600 Assisted Hydrothermal Synthesis and Formation Mechanism, J. Alloys Compd., 2011, 509, 885–890.
14. W. Yin, X. Chen, M. Cao, C. Hu and B. Wei, α-Fe₂O₃ Nanocrystals: Controllable SSA-Assisted Hydrothermal Synthesis, Growth Mechanism, and Magnetic Properties, J. Phys. Chem. C, 2009, 113, 15897–15903.
15. L. Liu, H. Z. Kou, W. Mo, H. Liu and Y. Wang, Surfactant-Assisted Synthesis of Alpha-Fe₂O₃ Nanotubes and Nanorods with Shape-Dependent Magnetic Properties, J. Phys. Chem. B, 2006, 110, 15218–15223.
16. A. G. Nasibulin, S. Rackauskas, H. Jiang, Y. Tian, P. R. Mudimela, S. D. Shandakov, L. I. Nasibulina, J. Sainio and E. I. Kauppinen, Simple and Rapid Synthesis of α-Fe₂O₃ Nanowires Under Ambient Conditions, Nano Res., 2009, 2, 373–379.
17. R. Ramesh, K. Ashok, G. M. Bhalero, S. Ponnusamy and C. Muthamizhchelvan, Synthesis and Properties of α-Fe₂O₃ Nanorods, Cryst. Res. Technol., 2010, 45, 965–968.
18. Z. Zhang, M. F. Hossain and T. Takahashi, Self-Assembled Hematite (α-Fe₂O₃) Nanotube Arrays for Photoelectrocatalytic Degradation of Azo dye under Simulated Solar Light Irradiation, Appl. Catal., B, 2010, 95, 423–429.
19. X. Hu, J. C. Yu, J. Gong, Q. Li and G. Li, α-Fe₂O₃ Nanorings Prepared by a Microwave-Assisted Hydrothermal Process and Their Sensing Properties, Adv. Mater., 2007, 19, 2324–2329.
20. D. Chen and L. Gao, A Facile Route for High-Throughput Formation of Single-Crystal α-Fe₂O₃ Nanodisks in Aqueous Solutions of Tween 80 and Triblock Copolymer, Chem. Phys. Lett., 2004, 395, 316–320.
21. W. Qin, C. Yang, R. Yi and G. Gao, Hydrothermal Synthesis and Characterization of Single-crystalline α-Fe₂O₃ Nanocubes, J. Nanomater., 2011, 159259.
22. G. Liu, Q. Deng, H. Wang, D. H. L. Ng, M. Kong, W. Cai and G. Wang, Micro/Nanostructured α-Fe₂O₃ Spheres: Synthesis, Characterization, and Structurally Enhanced Visible-Light Photocatalytic Activity, J. Mater. Chem., 2012, 22, 9704–9713.
23. D. Nishino, A. Nakafuji, J. M. Yang and D. Shindo, Precise Morphology Analysis on Platelet-type Hematite Particles by Transmission Electron Microscopy, ISIJ Int., 1998, 38, 1369–1374.
24. D. Peng, S. Beysen, Q. Li, Y. Sun and L. Yang, Hydrothermal Synthesis of Monodisperse α-Fe₂O₃ Hexagonal Platelets, Particuology, 2010, 8, 386–389.
25. W. Yu, T. Zhang, J. Zhang, X. Qiao, L. Yang and Y. Liu, The Synthesis of Octahedral Nanoparticles of Magnetite, Mater. Lett., 2006, 60, 2998–3001.
26. Z. Li, M. Kawasita, N. Araki, M. Mitsumori, M. Hiraoka and M. Doi, Preparation of Magnetic Iron Oxide Nanoparticles for Hyperthermia of Cancer in a FeCl₂–NaNO₃–NaOH Aqueous System, J. Biomater. Appl., 2011, 25, 643–661.
27. X. Zhang, Y. Niu, X. Meng, Y. Li and J. Zhao, Structural Evolution and Characteristics of the Phase Transformations Between α-Fe₂O₃, Fe₃O₄ and γ-Fe₂O₃ Nanoparticles under Reducing and Oxidizing Atmospheres, CrystEngComm, 2013, 15, 8166–8172.
28. T. Otake, D. J. Wesolowski, L. M. Anovitz, L. F. Allard and H. Ohmoto, Mechanisms of Iron Oxide Transformations in Hydrothermal Systems, Geochim. Cosmochim. Acta, 2010, 74, 6141–6156.
29. H. Ohmoto, Nonredox Transformations of Magnetite–Hematite in Hydrothermal Systems, Econ. Geol., 2003, 98, 157–161.
30. R. P. Viswanath, B. Viswanathan and M. V. C. Sastri, Kinetics of Reduction of Fe₂O₃ to Fe₃O₄ by the Constant Temperature Differential Thermal Analysis Method, Thermochim. Acta, 1976, 16, 240–244.
31. H. Y. Lin, Y. W. Chen and C. Li, The Mechanism of Reduction of Iron Oxide by Hydrogen, Thermochim. Acta, 2003, 400, 61–67.
32. J. H. Wang, Z. C. Tao, Y. Yang, J. Chang, H. W. Xiang and Y. W. Li, Reduction Kinetics of α-Fe₂O₃ Microspheres, Chin. J. Process Eng., 2007, 7, 288–292.
33. M. J. Tiernan, P. A. Barnes and G. M. B. Parkes, Reduction of Iron Oxide Catalysts: The Investigation of Kinetic Parameters Using Rate Perturbation and Linear Heating Thermoanalytical Techniques, J. Phys. Chem. B, 2001, 105, 220–228.
34. H. Wang, Y. Yang, B. S. Wu, J. Xu, M. Y. Ding, H. L. Wang, W. H. Fan, H. W. Xiang and Y. W. Li, Hydrogen Reduction Kinetics Modeling of a Precipitated Iron Fischer–Tropsch Catalyst, J. Mol. Catal. A: Chem., 2009, 308, 96–107.
35. M. Shimokawabe, R. Furuichi and T. Ishii, Influence of the Preparation History of α-Fe₂O₃ on Its Reactivity for Hydrogen Reduction, Thermochim. Acta, 1979, 28, 287–305.
36. O. J. Wimmers, P. Arnoldy and J. A. Moulijn, Determination of the Reduction Mechanism by Temperature-Programmed Reduction: Application to Small Fe₂O₃ Particles, J. Phys. Chem., 1986, 90, 1331–1337.
37. K. Yanagisawa and N. Yamasaki, Reduction of Haematite to Magnetite under Controlled Hydrothermal Conditions with Hydrogen gas, J. Mater. Sci., 1991, 26, 473–478.
38. J. Ge, Y. Hu, M. Biasini, W. P. Beyermann and Y. Yinty, Superparamagnetic Magnetite Colloidal Nanocrystal Clusters, Angew. Chem., Int. Ed., 2007, 46, 4342–4345.
39. S. Qu, H. Yang, D. Ren, S. Kan, G. Zou, D. Li and M. Li, Magnetite Nanoparticles Prepared by Precipitation from Partially Reduced Ferric Chloride Aqueous Solutions, J. Colloid Interface Sci., 1999, 215, 190–192.
40. R. S. Sapieszko and E. Matijewic, Preparation of Well-Defined Colloidal Particles by Thermal Decomposition of Metal Chelates, J. Colloid Interface Sci., 1980, 74, 405–422.
41. H. Wang, Q. He, B. Yao, G. Wen, F. Wang, Y. Xu, Y. Li, J. Li, C. Zhou, J. Wang, G. Li, L. Shan and J. Chen, Transformation of α-Fe₂O₃ to Fe₃O₄ Realized by Mechanochemical Reaction of α-Fe₂O₃ and SrCO₃, Metall. Mater. Trans. B, 2012, 43, 1574–1578.
42. B. Lv, Z. Liu, H. Tian, Y. Xu, D. Wu and Y. Sun, Single-Crystalline Dodecahedral and Octodecahedral α-Fe₂O₃ Particles Synthesized by a Fluoride Anion-Assisted Hydrothermal Method, Adv. Funct. Mater., 2010, 20, 3987–3996.
43. Y. B. Khollam, S. R. Dhage, H. S. Potdar, S. B. Deshpande, P. P. Bakare, S. D. Kulkarni and S. K. Date, Microwave Hydrothermal Preparation of Submicron-Sized Spherical Magnetite (Fe₃O₄) Powders, Mater. Lett., 2002, 56, 571–577.
44. L. Slavov, M. V. Abrashev, T. Merodiiska, C. Gelev, R. E. Vandenberghe, I. Markova-Deneva and I. Nedkovt, Raman Spectroscopy Investigation of Magnetite Nanoparticles in Ferrofluids, J. Magn. Magn. Mater., 2010, 322, 1904–1911.
45. D. L. A. de Faria, S. Venancio Silva and M. T. de Oliveira, Raman Microspectroscopy of Some Iron Oxides and Oxyhydroxides, J. Raman Spectrosc., 1997, 28, 873–878.
46. H. Monika, Raman Spectroscopy of Iron Oxides and (Oxy)hydroxides at Low Laser Power and Possible Applications in Environmental Magnetic Studies, Geophys. J. Int., 2009, 177, 941–948.
47. J. F. Lu and C. J. Tsai, Hydrothermal Phase Transformation of Hematite to Magnetite, Nanoscale Res. Lett., 2014, 9, 230.

---