The dependence of magnetic transition on the treatment solution used in the preparation of magnetic nanoparticles was investigated using as-prepared products from paramagnetic FeOOH/Mg(OH)2 via a chemically induced transition. Treatment using FeCl3 and CuCl solutions led to a product that showed no magnetic transition, whereas the product after treatment with FeSO4 or FeCl2 solutions showed ferromagnetism. Experiments revealed that the magnetism was caused by the ferrimagnetic γ-Fe2O3 phase in the nanoparticles, which had a coating of ferric compound. This observation suggests that Fe2+ in the treatment solution underwent oxidation to Fe3+, thereby inducing the magnetic transition. The magnetic nanoparticles prepared via treatment with an FeSO4 solution contained a larger amount of the nonmagnetic phase. This resulted in weaker magnetization even though these nanoparticles were larger than those prepared by treatment with an FeCl2 solution. The magnetic transition of the precursor (FeOOH/Mg(OH)2) was dependent upon treatment solutions and was essentially induced by the oxidation of Fe2+ and simultaneous dehydration of FeOOH phase. The transition was independent of the acid radical ions in the treatment solution, but the coating on the magnetic crystallites varied with changes in the acid radical ion.

1. Introduction

Nanotechnology involves the understanding and control of matter with dimensions of roughly 1–100 nm [1]. Magnetic nanoparticles have attracted increasing interest because their size range enables the investigation of the fundamental aspects of magnetic-ordering phenomena in materials with reduced dimensions, and this has the potential to lead to new technological applications [2, 3]. Studies on magnetic nanoparticles have focused on the development of simple and effective methods for the fabrication of nanoparticles with controlled size, morphology, and properties [4, 5]. Iron oxide nanoparticles including magnetite (Fe3O4), hematite (α-Fe2O3), maghemite (γ-Fe2O3), goethite (α-FeOOH), and akaganeite (β-FeOOH) nanoparticles have attracted enormous attention due to their interesting properties [610]. Many methods have been developed for the preparation of γ-Fe2O3 magnetic nanoparticles, including coprecipitation [11], a gas-phase reaction technique [12], direct thermal decomposition [13], thermal decomposition-oxidation [14], sonochemical synthesis [15], a microemulsion technique [16], hydrothermal synthesis [17], a vaporization-condensation method [18], and a sol-gel approach [19].

Reactions for the synthesis of oxide nanoparticles using the coprecipitation method can be grouped into two categories. In the first category, the oxide is produced directly; in the second category, a precursor is produced initially and then subjected to further processing (drying, calcination, and subsequent steps) [20]. During the chemical reaction, which is followed by calcination or annealing, a new phase is formed. In addition to the transition from amorphous to crystalline, the particle size increases, and the crystallites aggregate as the calcination temperature increases [21]. The conventional aqueous synthesis of γ-Fe2O3 particles involves three or more steps [22], but we recently found a new route for the production of γ-Fe2O3 magnetic nanoparticles that requires only two steps. This method involves the preparation of the paramagnetic FeOOH/Mg(OH)2 precursor via a coprecipitation method, followed by treatment of the hydroxide precursor in the liquid phase using a ferrous chloride (FeCl2) solution [23]. During this treatment, Mg(OH)2 dissolves, and the paramagnetic FeOOH precursor transforms into ferrimagnetic γ-Fe2O3 nanoparticles via dehydration:

This method is referred to as a chemically induced transition [24, 25]. However, it remains unclear whether such a magnetic transition can be produced using other solutions. In the present work, we used aqueous solutions of ferric chloride (FeCl3), cuprous chloride (CuCl), ferrous sulfate (FeSO4), and ferrous chloride (FeCl2) as treatment solutions to investigate the dependence of the magnetic transition on the treatment solution used and to attempt to explain the mechanism of the transition in the liquid phase.

2. Materials and Methods

2.1. Chemicals

FeCl3, Mg(NO3)2, NaOH, CuCl, FeSO4, and FeCl2 of analytical grade and all other chemicals were used as received without further purification. Distilled water was used throughout the experiments.

2.2. Preparation

The precursor was synthesized via coprecipitation. An aqueous mixture of FeCl3 (1 M, 40 mL) and Mg(NO3)2 (2 M, with 0.05 mol HCl; 10 mL) was added to an aqueous NaOH solution (0.7 M, 500 mL). The resulting solution was then heated to boiling for 5 min under stirring. The red-brown precursor precipitated gradually during this process. The precursor was collected and washed with dilute HNO3 solution (0.01 M) until the pH of the supernatant liquid was 7-8.

The as-prepared precursor was mixed with the treatment solution (0.25 M, 400 mL), and the resulting mixture was allowed to boil for 30 min. The products were then dehydrated with acetone and allowed to air-dry. Treatment with FeCl3, CuCl, and FeSO4 solutions produced samples (), (), and (), respectively. For comparison, treatment was also performed in an FeCl2 solution to produce sample ().

2.3. Characterization

Crystal structures and specific magnetization curves were obtained for samples ()–() using X-ray diffraction (XRD; XRD-7000, Shimadzu, Japan) and vibrating-sample magnetometry (VSM; HH-15, Nju-yq, China) at room temperature, respectively. The bulk chemical species, surface chemical composition, and morphology of samples () and () were determined using energy-dispersive X-ray spectroscopy (EDS; Quanta-200, Genesis, USA), X-ray photoelectron spectroscopy (XPS; XSAM800, Krator, UK), and transmission electron microscopy (TEM; Tecnai G20 ST, FEI, USA), respectively.

3. Results and Discussion

3.1. Results

XRD spectra are shown in Figure 1 for all of the samples. The crystal structure of samples () and () was different from that of sample (), whereas the main structure of sample () was identical to that of sample (), which corresponded to ferrite-like γ-Fe2O3. The results showed that sample () contained MgSO4·5H2O and Fe2(SO4)3·11H2O. The crystal phase of samples () and () is difficult to be distinguished from the XRD spectra, but it can be determined that both did not contain ferrite-like phase. Specific magnetization curves measured at room temperature are shown for all samples in Figure 2. The curves for samples () and () exhibited similar, apparently paramagnetic, behavior. Similar to sample (), sample () appeared to be ferromagnetic having coercivity (see the windows in Figure 2), but its magnetization was weaker than that of sample ().

According to the XRD and VSM results, samples () and () showed no magnetic transition; in contrast, sample () exhibited a transition similar to that shown by sample (). The specific saturation magnetization values () for samples () and () were determined from a plot of versus (where is the strength of magnetic field) in the high-field region, which was linear for strong magnetic fields [26]. versus () plots were represented as inserts of Figure 2, and the values were determined by extrapolating plots to -axis to be 59.20 and 43.36 emu/g for samples () and (), respectively.

Figure 3 shows EDS spectra for samples () and (). Sample () contained Fe, O, and Cl, and sample () contained S and Mg, in addition to Fe and O. The quantitative results are listed in Table 1. XPS measurements revealed the same chemical species that were identified by the EDS analysis in samples () and (). The XPS spectra are shown in Figure 4. The XRD and EDS results suggested that the γ-Fe2O3 phase was dominant in samples () and (). There was also a Cl-containing phase in sample (), and two S-containing phases in sample (). It is possible that the Cl-containing phase in sample () was FeCl3·6H2O [25], but that the content of FeCl3·6H2O may be so low that it did not produce clear diffraction peaks in the XRD spectrum (see Figure 1). The XRD results also implied that the S-containing phases in sample () were MgSO4·5H2O and Fe2(SO4)3·11H2O. Accordingly, the Fe2p3/2 lines and O1s lines observed for the two samples and the S2p3/2 line observed for sample () may be associated with two or more species. Detailed data on binding energies are listed in Table 2. We deduced from the experimental results that the binding energy of Mg1s in MgSO4·5H2O was approximately 1304.9 eV. In addition, it is noticed that the ferrite-like spinel structure, γ-Fe2O3 and Fe3O4, is difficult to discriminate by XRD due to peak broadening [27] and by XPS because the data are very close (see Table 2). However, Fe3O4 is not very stable and is sensitive to oxidation [28]. It was found that Fe3O4 nanocrystallites transformed into γ-Fe2O3 nanocrystallites using ferric nitrate treatment [29]. Therefore, it is judged that the magnetic phase for samples () and () is γ-Fe2O3, rather than Fe3O4 or mixed phase of both γ-Fe2O3 and Fe3O4.

TEM observations revealed that sample () was identical to sample () and that it consisted of nearly spherical nanoparticles. Typical TEM images for both samples are shown in Figure 5. Statistical analysis [30] indicated that the diameter of the nanoparticles fitted a lognormal distribution. The median diameters () for samples () and () were 10.55 and 13.16 nm, respectively, and the associated standard deviation values (, where is the geometry deviation) were 0.37 and 0.31, respectively.

3.2. Discussion

Experimental XRD and magnetization curve measurements indicated that the products after treatment with an FeCl3 solution were not ferromagnetic. The product after treatment with an FeSO4 solution exhibited a magnetic transition, similar to that observed for the products generated after treatment with an FeCl2 solution. These products mainly consisted of a ferrimagnetic γ-Fe2O3 phase. These results show that the magnetic transition was induced by Fe2+, rather than by the acid radical ions. Additionally, samples () and () contained FeCl3·6H2O and Fe2(SO4)3·11H2O, which, respectively, correspond to the ferrous salts FeCl2 and FeSO4 treatment solution. This observation suggests that the ferrous salts induced the transformation of FeOOH in the precursor into γ-Fe2O3 via dehydration, and this resulted in the simultaneous oxidation of Fe2+ to Fe3+. The results for sample () show that the ferric salt did not induce a transition, suggesting that the oxidization of the ferrous ions in the liquid phase was necessary for the dehydration of amorphous FeOOH to form magnetic phase. The experimental results for sample () show that the cuprous ions (Cu+), which may not have had the tendency to oxidize to cupric ions (Cu2+) in the liquid phase [31], did not cause FeOOH to form magnetic phase. This implies that the dehydrating action of the cuprous ions was weaker than that of the ferrous ions in the chemically induced transition in the liquid phase.

The experimental results also indicated that the products after treatment with an FeSO4 solution contained MgSO4, but the products after treatment with an FeCl2 solution had no corresponding Mg-containing constituent. This result suggests that FeSO4 in the treatment solution not only induced the transformation of FeOOH into γ-Fe2O3 via dehydration and then underwent oxidation into Fe2(SO4)3, but also reacted with dissolved Mg(OH)2, producing MgSO4: According to (3), MgSO4 first adsorbed onto the γ-Fe2O3 crystallites, forming MgSO4·5H2O. Some Fe3+ and ions also adsorbed, forming Fe2(SO4)3·11H2O on the outermost layer. Thus, the solutions of ferrous salts with different acid radicals (e.g., FeCl2 and FeSO4) could induce the transformation of the FeOOH/Mg(OH)2 precursor into a γ-Fe2O3 magnetic phase via dehydration. However, the coating on the γ-Fe2O3 crystallites varied when the treatment solution was changed. The results revealed that sample () contained γ-Fe2O3 and FeCl3·6H2O, and sample () contained γ-Fe2O3, MgSO4·5H2O, and Fe2(SO4)3·11H2O. A schematic diagram illustrating the formation of the nanoparticles using FeCl2 and FeSO4 solutions is shown in Figure 6. As the magnetic nanoparticles have surface heterolayers that are different from the magnetic core, the magnetic core spins close to surface can be pinned by the layer, and the pinning would cause an unusually large coercivity [32, 33]. So, such γ-Fe2O3 based magnetic nanoparticles appeared to be apparently ferromagnetic due to surface pinning rather than shape effects [34].

Because the metal salts were paramagnetic, the difference in the apparent magnetization of samples () and () depended on the γ-Fe2O3 content. For sample (), the molar percentages of γ-Fe2O3 and FeCl3·6H2O ( and , resp.) could be described as follows:where and are the atomic percentages of Fe and Cl in sample (), respectively. For sample (), the molar percentages of γ-Fe2O3, MgSO4·5H2O, and Fe2(SO4)3·11H2O (, , and , resp.) could be described as follows:where , , and are the atomic percentages of Fe, Mg, and S in sample (), respectively. Thus, the molar percentages of each phase () in samples () and () could be calculated from the values, which were measured using EDS, and are listed in Table 3. The mass percentages of each phase () in a sample were deduced using the following expression: where is the molar weight of the th phase. Accordingly, the mass percentage of each phase was calculated for samples () and () from the molar percentages () and molar weights of γ-Fe2O3, FeCl3·6H2O, MgSO4·5H2O, and Fe2(SO4)3·11H2O. The mass percentage values are listed in Table 3. Table 3 indicates that the main species in samples () and () was γ-Fe2O3. The values for samples () and () could therefore be expressed as where is the mass percentage of the γ-Fe2O3 phase and is the specific saturation magnetization of the γ-Fe2O3 phase. For samples () and (), their can be regarded as same since the γ-Fe2O3 phase resulted from the same FeOOH phase. Thus, the ratio of the values for samples () and () was proportional to the mass fraction of γ-Fe2O3. From the experimental results for the magnetization, we found that the ratio of the values for samples () and () was 0.73, which agreed with the ratio of the mass percentages of the γ-Fe2O3 phase (0.79). As a consequence, the mass percentage was available.

The volume percentage of each phase of the samples () could be obtained from the value: where is the density of the phase. The values for γ-Fe2O3, FeCl3·6H2O, MgSO4·5H2O, and Fe2(SO4)3·11H2O were 4.90, 1.844, 1.896, and 2.14 g/cm3, respectively. The values calculated for are listed in Table 4.

Because the particles in samples () and () were nearly spherical, the ratio of the average particle volume in sample () () to the average particle volume in sample () () could be described as follows: where and are the diameters corresponding to the average volumes of the particles in samples () and (), respectively. could be calculated directly from the size distribution parameters obtained using TEM by using the equation [30]. Accordingly, a value of 0.62 was obtained. Additionally, the average volume of the γ-Fe2O3 based nanoparticles, which contained many phases, (), can be expressed as , where is the average volume of the γ-Fe2O3 phase and is the volume percentage of the γ-Fe2O3 phase. Because the γ-Fe2O3 phase was produced from FeOOH, the values for samples () and () were the same. Therefore, the ratio of to could be described by ; that is,where and are the volume percentages of the γ-Fe2O3 phase in samples () and (), respectively. Thus, the value obtained from the volume percentages was 0.63, which agreed with the value of (0.62) and confirmed the validity of both the coating model of the nanoparticle structure and the volume percentage.

4. Conclusions

Metal ions in the treatment solution play a key role in the magnetic transition during the preparation of magnetic nanoparticles via the chemically induced transition of FeOOH/Mg(OH)2. No magnetic transition occurred when FeCl3 and CuCl were used as treatment solutions; in contrast, the product obtained using treatment with an FeSO4 solution, which is similar to that obtained using an FeCl2 solution, showed ferromagnetic behavior. The magnetic phase of the two products was γ-Fe2O3. Experimental results revealed FeCl3 and Fe2(SO4)3 in the products after treatment with FeCl2 and FeSO4 solutions, respectively. Thus, Fe2+, which likely has a stronger dehydrating ability than the cuprous ion Cu+, might have undergone oxidation to Fe3+, thereby inducing a magnetic transition via dehydration. Treatment with an FeSO4 solution resulted in the formation of MgSO4 on the γ-Fe2O3 crystallites (the binding energy of Mg1s is ~1304.9 eV), whereas no Mg-containing compound formed in the product of the treatment using an FeCl2 solution. This shows that the nonmagnetic coating on the magnetic γ-Fe2O3 crystallites varied with changes in the acid radical ions in the treatment solution. Treatment with an FeSO4 solution led to a higher content of the nonmagnetic phase in the product, and thus weaker magnetization, despite the fact that these particles were larger than those formed after treatment with an FeCl2 solution. Due to the surface pinned effect, such γ-Fe2O3 based magnetic nanoparticles, which were produced via the chemically induced transition, exhibited apparently ferromagnetism.

Here, we demonstrated the use of an FeOOH/Mg(OH)2 precursor and a ferrous salt treatment solution (as FeCl2 solution or FeSO4 solution) as a simple and effective method for the preparation of γ-Fe2O3-based magnetic nanoparticles. Obviously, the alkali-oxide FeOOH dehydrating into γ-Fe2O3 was induced with Fe2+ in ferrous salt solution transforming into Fe3+. For the preparation of magnetic nanoparticles via chemically induced transition (CIT) method, the relation between acting energy of the alkali-oxide dehydrating and oxidation of the ferrous ions is interesting, and this mechanism will be further investigated in future work.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


Financial support for this work was provided by the Innovation Foundation Project of Southwest University, China (no. 1318001), and the National Science Foundation of China (no. 11074205).