High saturation magnetization of γ-Fe2O3 nano-particles by a facile one-step synthesis approach

We have demonstrated the synthesis of γ-Fe2O3 nano-particles through a facile and novel calcination process in the air. There is no pH regulation, gas atmosphere, additive, centrifugation or other complicated procedures during the preparing process. A detailed formation process of the nano-particles is proposed, and DMF as a polar solvent may slower the reaction process of calcination. The structures, morphologies, and magnetic properties of γ-Fe2O3 nano-particles were investigated systematically, and the pure γ-Fe2O3 nano-particles obtained at 200 °C display uniform morphology good magnetic property. The saturation magnetization of obtained pure γ-Fe2O3 is about 74 emu/g, which is comparable with bulk material (76 emu/g) and larger than other results. In addition, the photocatalytic activity for degradation of methylene blue is also studied, which shows proper photocatalytic activity.

Magnetic nanomaterials have attracted much interest gradually, since these materials have many potential applications such as information storage, color imaging, magnetic refrigeration, gas sensors, ferrofluids, and photocatalysis etc. [1][2][3][4][5][6][7] Recently, researches of magnetic nanomaterials are fascinating due to its powerful usefulness for a variety of biomedical 8,9 and chemical engineering applications 10 . Among the magnetic nanomaterials, maghemite (γ-Fe 2 O 3 ) is considered as one of the most desirable materials for various applications due to its inherent biocompatible nature and stability of oxidation as well as its good magnetic properties 1,[11][12][13] . γ-Fe 2 O 3 also exhibits modest photocatalytic activity and separability 7,14 . It can be used associated with ZnO or TiO 2 to enhance the visible light adsorption and increase the electron/hole separation 15,16 . High magnetization of γ-Fe 2 O 3 has potential applications for cleaning polluted water with the help of magnetic separation. As a result, magnetic properties, as an important symbolic characteristic of γ-Fe 2 O 3 nano-particles, are noticeable for study. Therefore, new approaches for the synthesis of γ-Fe 2 O 3 particles as well as the investigation on their properties are of fundamental importance for the development of science and technology. The basic and conventional route of these methods or processes for the synthesis of γ-Fe 2 O 3 nano-particles are controlling the oxidation of Fe 3 O 4 [17][18][19][20] , and the total preparation process is shown as follows:  33 , ultrasonic decomposition 34 , high-temperature solution reaction 1 , chemical reaction 6 , and other chemical process [35][36][37][38][39] . These synthesis processes or methods are the important routes for the synthesis of γ-Fe 2 O 3 nano-particles and its composite materials. However, suitable pH value, long reaction time, and definite additives or surfactants are indispensable for obtaining pureγ-Fe 2 O 3 nano-particles with controllable morphology during those synthesis processes. Furthermore, centrifugation and purification are the vital factor for nano-particles with single products and good dispersity. Those preparation processes were totally complicated and cumbersome.
Significantly, the saturation magnetization of γ-Fe 2 O 3 nano-particles of the most methods mentioned above is still dissatisfied. Herein, we report a unified approach for the synthesis of γ-Fe 2 O 3 nano-particles in the air via a facile and novel calcination process. The method is different from the earlier approaches, and there is no pH regulation, gas atmosphere, centrifugation and other supplementary reagents during the preparing process. Various characterizations were measured to perform the obtained pure γ-Fe 2 O 3 nano-particles, and γ-Fe 2 O 3 nano-particles show a high saturation magnetization. In addition, the photocatalytic activity of γ-Fe 2 O 3 nano-particles was also studied.

Methods
A unified method was provided using a simple and convenient route to assemble γ -Fe 2 O 3 nano-particles. Ferric nitrate was dissolved in Dimethyl Formamide (DMF), the precursor was 0.6 mol/L, and calcined at different temperature (100 °C~400 °C, the interval is 50 °C) for 2 hours in the air. The heating rate was 1 °C/min. The schematic diagram of experiment is shown in Fig. 1.
The crystal structure of samples were measured by X-ray diffraction (XRD, PANalytical X'Pert) equipped with Cu-Kα radiation (λ = 1.5406 Å). The morphology of all samples was observed by using field emission scanning electron microscopy (FESEM, Hitachi S-4800) and transmission electron microscopy (TEM, Tecnai TM G 2 F30, FEI) equipped with an energy-dispersive spectrometer (EDS). The X-ray photoelectron spectroscopy (XPS, PHI-5702, Physical Electronics) were performed using a monochromatic Al-Kα irradiation and a charge neutralizer. All binding energies were referred to the C1 s peak at 284.6 eV of the surface adventitious carbon. The magnetic properties of the samples were measured by a vibrating sample magnetometer (VSM, Lakeshore 7304). The measurement process of surface areas and photocatalytic activity of the sample were shown in the Supporting Information (SI).

Results and Discussion
On the basis of the below experiments and results, a formation mechanism of the nano-particles in this work is proposed, which is outlined in Fig. 2. It is suggested that the following reactions occur during the calcination process: As shown in Fig. 2(a), the precursor is composed of DMF and iron nitrate. The solvents begin to volatilize at the beginning of the heating process (Fig. 2b). DMF in precursors plays a role of solvent, which helps the diffusion and contaction of the reactant molecules in the course of volatilization. When calcination temperature (CT) is increased (Fig. 2c), DMF is exhausted gradually, and the iron nitrite nonahydrate (Fe(NO 3 ) 3 ·9H 2 O) loses its water of hydration. The processes of Fig. 2(a-c) reveal that the surface morphology of sample seems not to be changed  from their nature. The corresponding SEM picture and XRD pattern at 100 °C show bulk grains, suggesting an amorphous structure. The non-magnetic results (VSM loop) also confirm the amorphous structure. When the specimen is calcined at a moderate temperature (Fig. 2d), Fe(NO 3 ) 3 decomposes into γ-Fe 2 O 3 , and the nucleation process is observed (See SEM image and XRD spectrum at 150 °C). When CT then increases, a large area of γ-Fe 2 O 3 nucleation is crystallized to the uniform nano-particles, which can be demonstrated by SEM image and XRD spectrum of 200 °C. Thereafter, when CT is about 220~400 °C (Fig. 2e), γ-Fe 2 O 3 nano-particles begin to engulf its surrounded crystals, and various sizes of α-Fe 2 O 3 particles are formed. This can be supported by  SEM images and XRD spectra of 250~300 °C. As CT is further improved, more γ-Fe 2 O 3 particles are transformed to α-Fe 2 O 3 , and the morphologies and structures of samples are mainly composed of α-Fe 2 O 3 particles (see SEM images and XRD spectra of 350~400 °C). Finally, when CT exceeds 500 °C, all γ-Fe 2 O 3 is disappeared, and plenty of α-Fe 2 O 3 nano-particles are achieved. The result can be confirmed by the previous phase transformation studies 2,38,40 .
To sum up, combining all results of γ -Fe 2 O 3 nano-granules, the detailed formation processes are discussed below. DMF is a commendable solvent when compared with water, and DMF is associated well with the cation 41,42 . As a solvent, DMF could disperse the ions, and coats each ion during the calcination process. As shown in Figure S1-S2 of SI, α -Fe 2 O 3 particles are produced in the water but it cannot generate in DMF under the same experiment condition (1 °C/min, 200 °C). This indicates DMF may slower the reaction process of calcination, which restrains the transformation process of γ -Fe 2 O 3 to α -Fe 2 O 3 . When CT or heating rates is increased, this restraint will be weakened. Figure 3 shows SEM images of the samples of different CT. It can be seen that the morphologies change obviously with the increased CT. The sample of 100 °C (Fig. 3a) presents a number of disorderly bulk particles, and the shape of particles changes to compact and well-regulated nano-particles at 150 °C (Fig. 3b). Especially when CT is 200 °C (Fig. 3c), a large quantity of uniform nano-particles with the size about 60 nm are achieved. However, when CT is further improved (250~400 °C of Fig. 3d-g), the particles display an increased grain growth (Fig. 3d,e). The particle size becomes nonuniform, and strong piece of reunited particles are observed with the improvement of CT (Fig. 3f,g). These results are detailedly studied by the following techniques.
XRD data were used to determine the structural parameters of all the samples. Figure 4 shows XRD patterns of all the samples at different CT. It can be observed that when CT is 100 °C, the sample is not crystallized due to the slow evaporation of DMF. Thus, there are no peaks could be seen in XRD pattern. Afterwards, the diffraction spectra of samples reveal the good single γ-Fe 2 O 3 phases (150 °C and 200 °C) with the cubic crystal system (JCPDS#39-1346), and all the diffraction peaks can be well indexed. It can be confirmed that the particles are γ-Fe 2 O 3 rather than α-Fe 2 O 3 , due to the different XRD pattern of α-Fe 2 O 3 (JCPDS#80-2377) and γ-Fe 2 O 3 17,19 . The average crystalline size estimated from Scherrer analysis is about 36 nm (150 °C) and 32 nm (200 °C) for γ -Fe 2 O 3 . In addition, the peak intensity of γ-Fe 2 O 3 calcined at 200 °C is stronger than that of 150 °C, which indicates that the nano-particles present higher crystallization at 200 °C. However, the samples display α-Fe 2 O 3 phases when CT exceeds to 250 °C. The emergence of additional impurity phase, i.e., α-Fe 2 O 3 , is more obvious with the enhanced temperature (300~400 °C). These XRD results indicate that the oxide of iron cannot be formed when CT is 100 °C. When CT is increased to 150~200 °C, γ-Fe 2 O 3 nano-particles can be achieved, and α-form is observed as CT exceeds 250 °C. It is well-known that γ-Fe 2 O 3 can be further transformed into α-Fe 2 O 3 at higher temperature 33,40 , and the phase transformation temperature in our research (250 °C) is similar to the previous literature 2, 40,43 .
In order to distinguish the chemical composition of γ-Fe 2 O 3 as opposed to Fe 3 O 4 , XPS measurement of pure γ-Fe 2 O 3 nano-particles is further performed, which is displayed in Fig. 5. The full scanned XPS spectra of γ-Fe 2 O 3 sample of 200 °C in the range of 0-1200 eV were shown in Fig. 5(a). Except for Fe 2p, O 1s, and C 1s peaks in the spectra, no redundant peaks appear together in γ-Fe 2 O 3 nano-particles, and C element belongs to the carbon contaminants absorbed on the surface of the tested samples. In particular, Fe 2p 3/2 spectra (Fig. 5b) exhibit two peaks at 710.6 and 724.1 eV, which are the characteristic peaks of the 3+ ion of γ-Fe 2 O 3 3,19 , and there is no signal or shoulder at smaller binding energies as would be expected for the presence of the Fe 2+ ion (~708 eV) 44,45 . Furthermore, an additional peak at about 718.7 eV is the shakeup satellite peak, which also indicates the absence of the Fe 2+ ion 3,44 . XPS results are consistent with the judgment of XRD spectra.
As a representative, the morphology and structure of the pure γ-Fe 2 O 3 nano-particles (200 °C) are further characterized by TEM. As shown in the Fig. 6(a,b), the results indicate large and black areas of near-spherical γ-Fe 2 O 3 nano-particles, and the granules present a low dispersity, which may be due to the reuniting of the nano-particles. HRTEM characterizations show the lattice fringes of the obtained ferrites, and the interfringe distance shown in Fig. 6(d,e) are 0.252 nm and 0.295 nm, which are correspond well to {311} and {220} planar spaces of γ-Fe 2 O 3 nano-particles, respectively. Both the lattice fringes and SAED (Fig. 6c) clearly presents a group of atomic planes within each particle, revealing the highly crystalline nature of these nano-particles. Meanwhile, Fig. 6(f) gives EDX data of γ-Fe 2 O 3 nano-particles, and the appearance of Cu peaks results from copper net used in the experiment. The element ratio of Fe:O is calculated to be 17.9:30.8, which is very close to the stoichiometry of γ-Fe 2 O 3 , which further confirms that the composition and structure are coincident with the chemical formulation of γ-Fe 2 O 3 . The particle size distributions, obtained from TEM micrographs, are shown in Fig. 6(g), and the histograms show that the samples display uniform particle distributions. The mean particle sizes obtained from the Gaussian fit of the histograms are 27 ± 2 nm, which is comparable with XRD line width results. However, size distributions obtained from SEM are larger than that of TEM, which may be due to the superimposed crystal or the compact arrangement of particles, and the shadow or astigmatism of the nano-particles could also cause the measurement error 46 . Some reports 47 also show a relatively large particle size distribution.

Centrifugation or wash Others
Particle size (nm) M s (emu/g) solvothermal 30 water, H 2 O 2 , D-(+ )-C 6  On the base of good understanding of the microstructure and chemical phase of nano-particles, the room temperature magnetic performance of products is discussed below. As shown in Fig. 7(a), the non-crystalline sample (100 °C) has no magnetism. When CT is 150 °C, i.e. γ-Fe 2 O 3 appears, the nano-particles emerge strong magnetism immediately, and the saturation magnetization (M s ) is about 61 emu/g. Particularly, when CT reaches to 200 °C, M s increases to 74 emu/g, which is comparable with bulk γ-Fe 2 O 3 sample (M s = 76 emu/g) 34,48,49 but larger than other γ-Fe 2 O 3 nanoparticles 11,30,33,50 . The higher values of M s are due to the better crystals of the nanoparticles, consistent with both the XRD and HRTEM data. A recent work also shows that the higher crystallinity is benefited to enhanced M s of the sample 51 . Typical comparative results of variable quantity of M s are shown in Fig. 7(d). It can be seen M s of this work is higher than other nanoparticles, but less when compared with nanoplate and nanocluster. However, when CT exceeds 250 °C, M s is decreased monotonously. That is because that the presence of non-magnetic α -Fe 2 O 3 leads to the relative reduction of magnetic γ-Fe 2 O 3 , and the magnetic moments total quality drop. When CT is further increased to 250~400 °C, the impurity α -Fe 2 O 3 appears more obviously (which can be confirmed by XRD results), and M s is reduced gradually. Furthermore, low temperature hysteresis loops of γ-Fe 2 O 3 (200 °C) have also been carried out at 80 K, 180 K, and 300 K, which are shown in Fig. 7(c). The low temperature is realized by the liquid nitrogen. As expected, M s and coercivity are also enhanced at low temperature. It is well-known that the coercivity and M s will increase when the temperature decreases 34,50 .
The degradation of methylene blue (MB) was performed as a model reaction to investigate the photocatalytic activity of the sample, which was shown in Figure S3. The results reveal that γ-Fe 2 O 3 has a little adsorption ability of MB with proper photocatalytic activity, which can degrade 16% MB dye in 60 min under UV irradiation. The photocatalytic activity of this work is comparable with the previous report 52 . As a result, the results provide the fabrication of Fe-based nanocomposites as proper performance photocatalysts, and high magnetization of γ-Fe 2 O 3 also has potential in addressing environmental protection issues. The Brunauer-Emmett-Teller (BET) surface areas of γ-Fe 2 O 3 nano-particles were measured to be 24.8 m 2 /g, which is smaller than the mesoporous nano-particles 24,29,30 . This suggests that the content of mesopores in the sample is considerably low. BJH average pore diameters calculated from the adsorption branch of the isotherms is 4.7 nm for the γ-Fe 2 O 3 nano-particles, and the corresponding total pore volume is 0.04 cm 3 /g.
Various above investigations have demonstrated that a number of uniform and smooth γ-Fe 2 O 3 nano-particles are obtained using calcination process in the air. Significantly, some typical methods or processes as the comparative results are discussed, and the comparative data are presented in Table 1. Except for the iron source of the preparing process, Table 1 detailedly shows the experiment parameters of solvent, additive or surfactant, pH, reaction time, centrifugation, and other parts of various methods. These methods or processes are not limited to the literatures we provided. As a result, although the dispersity and size of nano-particles in our research are not good with some of the previous reports, this technique only needs one solvent during the preparing process, and other additional processes are omitted. The mehtod realizes a simple, rapid and convenient route for assembling γ-Fe 2 O 3 nano-particles when compared with others.