Experimental Investigation of the Coprecipitation Method: An Approach to Obtain Magnetite and Maghemite Nanoparticles with Improved Properties

Iron oxides that exhibit magnetic properties have been widely studied not only from an academic standpoint, but also for numerous applications in different fields of knowledge, such as biomedical and technological research. In this work, magnetite andmaghemite nanoparticles were synthesized by chemical coprecipitation of FeCl 2 ⋅4H 2 O and FeCl 3 ⋅6H 2 O (proportion of 1 : 2) in three different cases using two bases (sodium hydroxide and hydroxide ammonium) as precipitants. The chemical coprecipitation method was selected for its simplicity, convenience, reproducibility, and low cost in the use of glassware. The nanostructured materials were characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD) and magnetometry (VSM). The objective of this work is to study the variation in the morphological characteristics and physical properties of nanoparticles magnetic as a function of the different production processes. As observed by TEM, the materials obtained from the precipitating agent NH 4 OH are more uniform than those obtained with NaOH. From XRD pattern analysis, it appears that the obtained materials correspond to magnetite and maghemite and, from magnetometry VSM analysis, show high magnetization as a function of the magnetic field at room temperature, indicating that these materials are superparamagnetic.


Introduction
In recent years, there has been great interest in the study of composites such as ferrite, which can be represented by the general formula [M 2+ ][Fe 2 3+ ]O 4 , where M 2+ is a divalent cation of a metal (Fe, Co, Ni, Mn, Cu, Zn, and Cd).The main iron oxides used in the preparation of magnetic fluids are magnetite and maghemite.These iron oxides show a spinel-like structure formed by a cell unit face-centered cubic (FCC) network of oxygen anions, with sites filled by cations, differing in tetrahedral and octahedral coordination [1,2].The maghemite spinel-inverted type structure presents unit cells with a vacancy of 2.67 cationic iron atoms located in octahedral sites [3].The chemical representation of magnetic iron oxides may be indicated as follows: where t is tetrahedral site, o is octahedral site, and ◻ is vacancy.
Iron oxides such as magnetite and maghemite, in the form of nanoparticles dispersed in certain mediums, present great scientific interest in many technological applications.These include biosensors [4], magnetic fluids [5], contrast agents for magnetic resonance imaging [6], vectorization of drugs [7], and magnetic fluid hyperthermia (MFH), due to their chemical stability, biocompatibility, and ability to be heated in aqueous suspension in the presence of an alternating field [8,9].The use of ferromagnetic nanoparticles in the biomedical field needs this type of materials to present sizes with average diameter smaller than 20 nm and stability in physiological conditions.The properties of these materials are strongly dependent on particle size, particle interactions/solution matrix, and the arrangement of nanoparticles in the matrix [10,11].
There are several methods to obtain ferromagnetic nanomaterials, for instance, chemical coprecipitation, microemulsion, and high temperature decomposition of organometallic precursors [12,13].Once these ferromagnetic nanomaterials are obtained, it is important to conduct an evaluation of their physical characteristics.It is known that crystals are periodic arrangements of atoms with size in the order of Å (angstroms), equal to 10 −10 m, and X-rays show the wavelength () in the same order.The X-ray emitted by a cobalt atom is in the range of (Co K 1 = 1,7889 Å) and thus the crystals serve as an X-ray diffraction grating.Therefore, the periodicity of atoms in a crystal can be measured by diffraction, which also permits the identification of the crystalline structure.X-ray diffraction (XRD) is a technique based on the interaction of electromagnetic radiation from the crystal structure, where the characteristic dimensions of the solid are comparable with the wavelengths of radiation ().The relationship among the diffraction angle 2, the wavelength of the radiation (), and the basal distances of the structure ( ℎ ) is presented by Bragg's Law [14]: where  is the order of interference.This method allows the indexing of the peaks characteristic of groups linked to plane (ℎ), that is, the Miller index of the crystal lattice compared to the reference data from the Joint Committee on Powder Diffraction Standards (JCPDS).
The behavior of magnetic materials can be evaluated by graphical representation, which measures the magnetic induction and magnetization as a function of a magnetizing or demagnetizing field [15].When magnetization reaches its maximum and becomes constant, the saturation magnetization has been reached.At this time, the increase in magnetizing force does not influence the material.Thus, the magnetic behavior, indicated by the magnetization curve, changes according to the material being analyzed, since each type of materials presents different permeability.
The purpose of this study is to synthesize magnetite and maghemite nanoparticles that present morphological characteristics and a quality of dispersion with improved properties for different applications.This is to be done by the exploration of new conditions applied to the chemical coprecipitation method.Here we have chosen to characterize the synthesized nanoparticles by transmission electron microscopy (TEM), and morphology and physical properties are assessed by XRD and magnetization as a function of magnetic field.Nanoparticles were synthesized according to the method of alkaline coprecipitation of iron salts in aqueous solutions [13,16,17].

Synthesis of Magnetite.
The sample magnetite nanoparticles were prepared by the aqueous mixture of ferrous and ferric solution in the form of hydrated chloride salts in alkaline medium, in the proportion of Fe 2+ /Fe 3+ = 1/2.Initially, 10 mL of the HCl 2 mol⋅L −1 was added to prevent Fe 2+ oxidation and precipitation Fe 3+ in the form of hydroxide.A volume of 125 mL of the precursors with molar concentration of aqueous solutions 0.30 mol⋅L −1 Fe 3+ and 0.15 mol⋅L −1 Fe 2+ was mixed and agitated for three minutes.Afterwards, the binary solution of Fe 3+ and Fe 2+ was added into a solution of NaOH 1.5 mol⋅L −1 , and NH 4 OH 25% solution was added in the binary solution of the Fe salts, now under constant stirring at 2000 rpm, according to each process.The resulting black precipitate was isolated by a magnetic field and washed with distilled water, reaching a value of pH equal to 9. Half of the resulting product was separated by a magnet and dried at 40 ∘ C for 24 hours for characterization.This process is described by the chemical equation below: The suspensions were obtained using NaOH 1.5 mol⋅L −1 and/or NH 4 OH 25% as precipitant agent for each procedure, under vigorous stirring for about 30 min at room temperature.It is important to note that the addition sequence had an influence on the phase formation and on the size of the magnetite nanoparticles.More precisely, when sodium hydroxide was employed in the process of coprecipitation, the solution containing the mixture of Fe 2+ /Fe 3+ was added over the base (NaOH).When ammonium hydroxide was employed, the base (NH 4 OH) was added to the solution containing the mixture of Fe 2+ /Fe 3+ .According to the different conditions used to prepare the magnetite samples, three procedures are described below.Similarly, the solution precipitate into NH 4 OH was labeled MgP11.After preparation, the suspension was kept still for 24 h in the presence of a magnetic field, at room temperature.

Oxidation of Magnetite.
The black precipitates obtained by the preparation of magnetite (described in Section 2.2) were diluted with distilled and deionized water to a volume of 300 mL.With the exception of Procedure 3, the preparation of maghemite was accomplished observing the following general conditions: the pH of the suspension was adjusted to 3.5 with HCl 0.2 mol⋅L −1 under low agitation until stabilization, and heated at 95 ∘ C for one hour.According to the different conditions used to prepare the maghemite samples, the three procedures are described below.
For Procedure 1, half of the MgP1 and MgP3 suspensions were oxidized and agitated under aeration (open system in air) for one hour by boiling at about 95 ∘ C [13].The suspension oxidized from MgP1 was labeled MgHP2.Similarly, suspension oxidized from MgP3 was labeled MgHP4.The color of the suspension slowly changed from black to brown.Finally, the product was washed with distilled water and separated four times by magnetic field and dried at 40 ∘ C for 24 h.
For Procedure 2, half of the MgP5 and MgP7 suspensions were oxidized and agitated under flux of 100 mL/min O 2 special gas for 1 h by boiling at about 95 ∘ C. The suspension oxidized from MgP5 was labeled MgHP6.Similarly, suspension oxidized from MgP7 was labeled MgHP8.The color of the suspension slowly changed from black to brown.Finally, the product was washed with distilled water and separated four times by magnetic field and dried at 40 ∘ C for 24 h.
For Procedure 3, half of the dry precipitate obtained from Procedure 3 (Section 2.2) was submitted to oxidation at 250 ∘ C ± 1 ∘ C for three hours.The solid oxidized from MgP9 was labeled MgHP10.Similarly, the solid oxidized from MgP11 was labeled MgHP12.The color of the solid slowly changed from black to brown.

Transmission Electron Microscopy.
Nanoparticle morphology was determined by transmission electron microscopy (TEM) carried out in a JEOL transmission electron microscope Model 1011 (Jeol, Tokyo, Japan) operated at 110 KV.Samples were prepared from the particle suspensions in deionized water.Samples were homogenized by Homo Mix Homogenizer D-130 (series 175959).A drop of well dispersed supernatant was placed on a carbon-coated 200 mesh copper grid and then dried at room temperature before it was attached to the microscope sample holder.The evaluation of the average diameter of the magnetite and maghemite nanoparticles synthesized by Procedures 1, 2, and 3 was performed by the analysis of electron micrographies with the aid of computer-based software (Image-Pro Plus 6.0; Media Cybernetics, Silver Spring, MD).For each nanoparticle sample produced, 350 nanoparticles were randomly observed.

X-Ray Diffractometry.
The crystallographic analysis of the samples was performed by XRD powder method.
Diffraction patterns (vs.2) were recorded with a Shimadzu XRD 6000 diffractometer, equipped with cobalt cathode (Co K 1 1,788965 Å) and Ni filter, operating at 40 kV and current of 30 mA.A continuous scan of 2 deg/min mode was used to collect 2 data from 10 to 70 degrees, to determine the crystal structure of the samples.X-ray spectra were plotted with the aid of Microcal Origin 6.0 software (Microcal Software Inc., Northampton, MA, USA).

Magnetic Properties.
Magnetic properties of the samples were measured in the solid state at room temperature using a superconducting quantum interference device (Lake Shore Model 7300 vibrating sample magnetometer VSM Controller).The vibration that occurs in the presence of an external magnetic field can be applied to both transverse direction and longitudinally in the direction of vibration.By the Faraday-Lenz principle of induction, an electromotive force (FEM) is induced in the detection coils, related with magnetic moment of the sample.The magnetic moment of each dried sample was measured over a range of applied magnetic fields between −5 × 10 4 and 5 × 10 4 Oe with a sensitivity of 0.1 emu.

Polydispersion of Samples.
The polydispersity index (PDI) is a mathematical definition that accounts for the relative error between curve fit and experimental values [18][19][20][21].PDI indicates the quality of the dispersion.Values of ≤0.1 for the nanoparticle suspensions reflect that the quality of the dispersion is monodisperse.Absolute and relative errors show approximations between the real number and the experimental number.The absolute error (EA  ) indicates the difference between the exact value of a number () and its approximate value ( ); that is, EA  = | − |.In general, is the only known value, making it impossible to obtain the exact value of the absolute error.In order to evaluate the error precision, it is necessary to consider the magnitude of .For this, the relative error (ER  ) is used.(ER  ) is defined as the ratio of the absolute error and the approximate value; that is, ER  = ER  / .The relative error considers the magnitude of the numbers involved.For this reason, the relative error is mostly used to evaluate the precision of the calculus done.
Procedures 1 and 2 show that the maghemite nanoparticles obtained by open system oxidation of the initially synthesized magnetite present average diameter values well below their precursors.A possible explanation is that the oxidation process with the adjusted 3.5 pH, under stirring and heating at 95 ∘ C for one hour, caused dissolution followed by crystallization of the seeds, known as Ostwald ripening, thus changing the size of the new materials.
As was observed in Figure 4, magnetite electron micrographs showed that the quality of dispersion for the particles is monodisperse, which is unusual for particles prepared by coprecipitation method.Results revealed that MgP1 nanoparticles present a pseudospherical shape and MgP3 nanoparticles show a more uniform shape with an average diameter of 9.82 nm and 11.22 nm, respectively.
Polydispersity indexes, calculated from the average of the nanoparticle diameters, according to the histograms, and fitted by the Gaussian law (Figures 1, 2, and 3) are in the range of 0.0256 to 0.2669 (Table 1).These values indicate relative homogeneity in the system, with good distribution, and smaller size, consequently describing a monodisperse condition [19].
The differences in size between the nanomaterials obtained by coprecipitation with NaOH and NH 4 OH are due to the differences in the process of nucleation and grain growth during synthesis.Taking into account the fact that NH 4 OH is a weak base, it produces a smaller number of magnetite cores, which favors crystal growth and Ostwald ripening, resulting in an amount of particles with larger sizes.
When NaOH is used, considering that it is a strong base, the formation of a greater number of cores occurs, which jeopardizes their growth, as also observed by Vayssieres [22] and Gnanaprakas et al. [23,24].
XRD results (Figure 5) are in accordance with the main characteristic peaks for magnetite (JCPDS 19-629) and maghemite (JCPDS 39-1346) reference patterns.In a comparison study, the peaks of the materials were observed as having the predominant phase of the powder-type cubic crystal structure corresponding to magnetite spinel.Maghemite samples presented similar diffraction peaks when compared to magnetite peaks, making their identification difficult.Thus, to facilitate the differentiation of maghemite species, two secondary peaks with low intensity were observed at 23.7 and 26.3 2 degree angle (using copper radiation), 27.4 and 30.55 2 degree angle (using cobalt radiation), characteristic of maghemite species.Therefore, the dark brown powder obtained under the established conditions for this study showed the presence of maghemite, which corroborates the observed X-ray diffraction data obtained.
Magnetite and maghemite spinel-type nanoparticles showed reflections peaks at 111, 220, 311, 400, 440, and 511.The most intense peaks were observed at 220, 311, and 400, which  were compared with JCPDS values.Thus, with this comparison, it was possible to identify if the sample corresponded to magnetite or maghemite.Using Scherrer's equation, it was possible to identify the more intense reflection (311) as a preliminary estimate of the average diameter of the nanoparticles (Table 2).
Once the most intense peak was determined at 311, Scherrer's equation [25] (shown below) was used to calculate the average nanoparticle diameter: The equation uses the reference peak width at angle , corresponding to Bragg's angle, where  is the X-ray wavelength (Co K 1 = 1.7889Å),  is the width of the XRD peak (311) at half height, and  is a shape factor of approximately 0.93 for magnetite and maghemite [25,26]: )) TEM and XRD were used to determine morphometric aspects in magnetite and maghemite nanoparticles synthesized.In relation to size, TEM results showed that nanoparticle morphometric average diameter (Figures 1, 2, and 3)   presented 60% higher values when compared to results obtained by XRD using Scherrer's equation.Considering this, it is important to note that TEM is the first method used to determine the size and size distribution of nanoparticle samples [27].XRD method, along with the application of Scherrer's equation, determines the average structural diameter of nanoparticles [12][13][14].
The size increase of maghemite nanoparticles compared with that of magnetite probably involves the well-known Ostwald ripening process of dissolution-crystallization [28,29].Magnetite nanoparticles with a relatively smaller size are unstable in acidified suspension during oxidation in Procedures 1 and 2. Therefore, they tend to dissolve in the suspension and crystallize on the seeds of the larger ones.
As also observed by Kang and collaborators [13], due to the incomplete ripening process, in the present study there still remained maghemite nanoparticles with diameters much smaller than the average size.
Maghemite nanoparticles obtained from Procedure 3 presented the same size of magnetite nanoparticles, due to the direct oxidation process not involving the Ostwald ripening process of dissolution-crystallization [28,29].
Figure 6 shows magnetization curves for Procedures 1-3, where magnetization curves of dried powders of both (a) magnetite and (b) maghemite nanoparticles are determined by vibrating sample magnetometry.
The analysis of the magnetic behavior of these materials exhibits high values of magnetization saturation.Curves without hysteresis were also observed by Cullity and Graham [30,31].These curves do not present retentivity or coercivity, characteristic of materials with the absence of residual magnetization after removal of the magnetic field.This behavior is known as superparamagnetism, which occurs only at the nanoscale dimension [32].
Comparing the magnetization curves, there was a relative decrease in the saturation magnetization value for the magnetic materials produced with NaOH in relation to materials treated with NH 4 OH.The decrease can be attributed to the effects of grain size, when using NH 4 OH, which was seen to be more effective than NaOH, for both magnetite and maghemite.It can be seen that the size is directly related to the coercive force and magnetic permeability; in other words, the greater the grain, the greater the magnetic permeability and the lower the coercive force.
The decrease in magnetization versus magnetic field (× ) observed in the produced nanoparticles is attributed to several factors, such as surface contribution, spin canting, surface disorder, stoichiometry deviation, cation distribution, and adsorbed species [33,34].
As can be noted in Figure 6, the magnetic characterization of the dried samples measured showed that the saturation magnetization of the magnetite produced is approximately 60 emu/g for NaOH and 78 emu/g for samples obtained from NH 4 OH in the same conditions.This result indicates that the chemical coprecipitation process with NH 4 OH can increase the growth rate and the crystallization progress for the magnetite nanoparticles.Magnetic characterization can also be influenced by the size of particles produced [35].Those values are lower than the theoretical value of the bulk magnetite (92 emu/g) [36], as already observed by Panda and colleagues (67.8 emu/g for 12 nm nanoparticles) [33] and Goya and collaborators (75.6 emu/g for 150 nm nanoparticles) [34].
In further studies it will be important to examine the morphology and magnetic properties, in specific conditions, of the iron oxides magnetite and maghemite using TEM, DRX, and magnetometry.Here, we saw an explanation of the structure of maghemite, which is derived from magnetite by replacing eight Fe 2+ in octahedral sites with the equivalent charge of 5 1/3 Fe 3+ and 2 2/3 cation gaps [37].To obtain more detailed information about the hyperfine structures of the two iron oxides in question, Mössbauer spectroscopy measurements at low temperatures need to be carried out.These measurements would show that all the tetrahedral sites are filled and that gaps occur in the octahedral sites.This technique could not be carried out at the time of the experiment and so is not part of the current study.

Conclusions
As observed by TEM and in Scherrer's equation, the materials obtained from the precipitating agent NH 4 OH are more uniform than those obtained with NaOH.TEM showed that small magnetite particles are possibly embedded into larger particles when treated with NaOH.In contrast, when treated with NH 4 OH, the particles formed were more uniform and monodisperse.Nanoparticles synthesized by both Procedures 2 and 3 (only for NH 4 OH treatment) showed materials with similar sizes, suggesting that this process can be adopted when one wishes to obtain similar sized nanoparticles for biomedical use.
From XRD pattern analysis, it appears that the obtained materials correspond to magnetite and maghemite, showing high magnetization saturation reachable at low applied field and superparamagnetic behavior at room temperature.This paper showed that the specific conditions, applied during coprecipitation, allow for the synthesis of magnetic nanoparticles with morphological characteristics and a quality of dispersion which is more appropriate for environmental and biomedical applications.Victoria Monge Fuentes was involved in revising the paper critically for important intellectual content and edited the paper.Renata Carvalho Silva and Carolina Madeira Lucci carried out the magnetic nanoparticle morphological characterization by transmission electron microscopy.Jaqueline Rodrigues da Silva and Maria Luiza Fascineli contributed with technical assistance.Ricardo Bentes de Azevedo supervised the whole work.All authors read and approved the final paper.

Procedure 1 .
An open system (aeration) was used to prepare solutions.The solution precipitate into NaOH was labeled MgP1.Similarly, the solution precipitate into NH 4 OH was labeled MgP3.Procedure 2. A closed system was used to prepare solutions under nitrogen bubbling at a flow rate of 100 mL/min.The solution precipitate into NaOH was labeled MgP5.Similarly, solution precipitate into NH 4 OH was labeled MgP7.Procedure 3.An open system was used to prepare solutions.The solution precipitate into NaOH was labeled MgP9.

Figure 1 :
Figure 1: Size distributions for the magnetite and maghemite nanoparticles prepared by Procedure 1.

Figure 3 :
Figure 3: Size distributions for the magnetite and maghemite nanoparticles prepared by Procedure 3.

Figure 5 :
Figure 5: X-ray diffraction (XRD) patterns with cobalt radiation of sample were obtained by chemical coprecipitation method using NaOH 1.5 mol⋅L −1 (a) and its oxidation (b) and with NH 4 OH 25% (c) and its oxidation (d).

Figure 6 :
Figure 6: Magnetization-magnetic field curves obtained for (a) magnetite nanoparticles and (b) maghemite nanoparticles at room temperature.