Magnetic Characterization by Scanning Microscopy of Functionalized Iron Oxide Nanoparticles

This study aimed to systematically understand the magnetic properties of magnetite (Fe3O4) nanoparticles functionalized with different Pluronic F-127 surfactant concentrations (Fe3O4@Pluronic F-127) obtained by using an improved magnetic characterization method based on three-dimensional magnetic maps generated by scanning magnetic microscopy. Additionally, these Fe3O4 and Fe3O4@Pluronic F-127 nanoparticles, as promising systems for biomedical applications, were prepared by a wet chemical reaction. The magnetization curve was obtained through these three-dimensional maps, confirming that both Fe3O4 and Fe3O4@Pluronic F-127 nanoparticles have a superparamagnetic behavior. The as-prepared samples, stored at approximately 20 °C, showed no change in the magnetization curve even months after their generation, resulting in no nanoparticles free from oxidation, as Raman measurements have confirmed. Furthermore, by applying this magnetic technique, it was possible to estimate that the nanoparticles’ magnetic core diameter was about 5 nm. Our results were confirmed by comparison with other techniques, namely as transmission electron microscopy imaging and diffraction together with Raman spectroscopy. Finally, these results, in addition to validating scanning magnetic microscopy, also highlight its potential for a detailed magnetic characterization of nanoparticles.


Introduction
A spectrum of current studies in nanotechnology highlights the rapid advance that has been taking place in this area of science, especially in medical applications. This microscopy (TEM) and Raman results also show that the synthetized samples are composed mostly of magnetite, some of them containing small amounts of hematite. These results are crucial for most of the above-mentioned applications to avoid the agglomerations of MNPs, in addition to the reduction in oxidation.
The following production procedures were used, as illustrated in Figure 1: (1) Two solutions were prepared for the iron salt mixture, one dissolving 0.02516 mol of FeCl 3 ·6H 2 O in 25 mL of distilled water and another dissolving 0.01421 mol of FeSO 4 ·7H 2 O in 10 mL of HCl (5.49 mol L −1 ); (2) the iron salt solutions were mixed in a ratio of 4 mL Fe 3+ to 1 mL Fe 2+ , in agreement with the ratio of 2:1; (3) an aqueous solution of NH 4 OH (1.30 mol L −1 ) was preheated to 80 • C in a separate container on a heating plate for 10 min; (4) the obtained salt mixture was added to the basic solution of ammonium hydroxide (28% P.A.) under manual vigorous stirring with a glass rod; (5) a black precipitate was formed, indicating the formation of NPs; (6) the MNPs were kept in the ultrasound bath for one hour to prevent agglomeration; (7) subsequently, the MNPs were washed three times with distilled water (distiller NT 422, NovaTecnica, Brasil) and the help of a permanent magnet to hold the nanoparticles. The washing procedure was repeated several times to neutralize the solution pH. The procedures for coating Fe3O4 samples with Pluronic surfactants (Figure 2) were performed, according to the procedure, as follows: (1) solutions at different concentrations of Pluronic F-127 were prepared (Table 1); (2) a 10 mL aliquot was taken from each solution and added to the aqueous NH4OH solution (1.28 mol L −1 ); (3) the basic solution was heated to 80 °C on a heating plate for 10 min; (4) the iron salt solution was added to the  The procedures for coating Fe 3 O 4 samples with Pluronic surfactants (Figure 2) were performed, according to the procedure, as follows: (1) solutions at different concentrations of Pluronic F-127 were prepared (Table 1); (2) a 10 mL aliquot was taken from each solution and added to the aqueous NH 4 OH solution (1.28 mol L −1 ); (3) the basic solution was heated to 80 • C on a heating plate for 10 min; (4) the iron salt solution was added to the Pluronic F-127 (Sigma-Aldrich, St. Louis, MI, USA) solution under vigorous stirring, using a nonmagnetic spatula; (5) a black precipitate product was immediately observed, indicating the formation of NPs; (6) the NPs were kept in an ultrasound bath for one hour to prevent agglomeration; (7) the same procedure used for washing MNPs was performed for the samples coated with Pluronic F-127. After these procedures, we could observe the NPs (Figure 3), that were stored at approximately 20 • C.

Nanoparticles Characterization
Uncoated and functionalized MNPs were characterized using Raman spectroscopy, transmission electron microscopy (TEM) scanning magnetic microscopy (SMM) and vibrating sample magnetometer (VSM), as described below.

Raman Spectroscopy
Raman spectra were obtained from the accumulation of 10 spectra of 20 s each, using a Senterra Bruker micro-Raman spectrometer equipped with a CCD system and adjusted to a resolution of 4 cm −1 . A 785 nm solid-state laser was used as excitation source, with 20 and 50 mW power. The spectrometer uses an Olympus BX50 microscope with 50× objective to focus the laser beam onto the sample surface. To avoid local heating effects, the laser was set to 20 mW and 50 mW was used to examine local heating effects.
Both the magnetite [31] and maghemite [32], a phase transition to hematite products were was investigated by varying the laser power. The spectra shown in Figure 4 were obtained with a power of 20 mW at 0 (black lines) and 6 (red lines) months after synthesis. By using a 20 mW power, we can guarantee the early phase spectrum of materials without any interference from the laser power. The bands observed in the spectra, shown in Figure 4, at 293, 345, 500, and 685 cm −1 are characteristic of the magnetite phase [22,31,37,38], although hematite phase bands (222 and 400 cm −1 ) have been observed in some spectra [22,39], featuring a phase mixture. Furthermore, the high similarity between the spectra taken at 0 and 6 months (See Figure 4a), for each sample, is a strong indication of the particles stability and assures us that the sample has not been showing any oxidation process over time. The position and FW HM were obtained through the deconvolution of Raman spectra using the Origin 6.0 software, the result for NP0 e NP5 with 0 months are shown in Figure 4b,c.  In addition, Chandramohan et al. [40]-considering that in nano-sized systems, due to lack of long range order, scattering with 0 is allowed, which results at the appearance, widening and displacement of the peak position in the Raman spectra [40]-proposed an relation for estimating size of the magnetite particles, Equation (1) based on Raman spectra (modes A1g(1) (685 cm −1 ) and TEM micrographs for the ferrite particles synthesized by the double microemulsion and polymer combustion methods. In addition, Chandramohan et al. [40]-considering that in nano-sized systems, due to lack of long range order, scattering with q = 0 is allowed, which results at the appearance, widening and displacement of the peak position in the Raman spectra [40]-proposed an relation for estimating size of the magnetite particles, Equation (1) based on Raman spectra (modes A 1g(1) (685 cm −1 ) and TEM micrographs for the ferrite particles synthesized by the double microemulsion and polymer combustion methods.
where ∆ω is the variation between the frequency of Raman peak (ω(L)) with particle size L and the frequency at the Brillouin zone center (ω 0 ). a is the lattice constant of the crystal and, A and γ are the fit parameters that describe the phonon confinement in nano-crystallites. Hence, we used the inverse function of the relation proposed by Chandramohan et al., Equation (1), to determine an expression for estimating size of the particles (L) in function of ∆ω, in this case, ∆ω is full width at half maximum (FW HM), as follows: with a = 8.38 Å, A = 104 cm −1 , and γ = 0.8 as constants, and obtained by Chandramohan et al. The estimated value of particle sizes based on the FW HM of the 685 cm −1 peak are shown in Table 2. The Raman spectra obtained with a power of 50 mW for the same samples revealed the presence of Raman peaks at 221, 284, 400, 488, and 607 cm −1 (see Figure 5). These values are typical of α-Fe 2 O 3 and are in accordance with the literature [22,39,[41][42][43]. Peaks at 671 and 710 cm −1 were also observed and according to the literature, the appearance of these peaks in the hematite phase may be related to the nanometric size of the particles [41,42,44]; such activation shows that the disorder of the nanomaterial surface induces the phonon scattering symmetry breaking [45], which explains why this peak is not always observed in the spectra. In addition, the 400 and 607 cm −1 vibrational modes show a gradual softening with increasing percentage of Pluronic surfactant, indicating the presence of Pluronic bound to the MNPs.

Transmission Electron Microscopy
Powder samples were diluted in an aqueous medium and deposited on a 400-mesh carbon-coated copper grid. After the drying process, the samples were analyzed in a TEM (Jeol 2010 instrument), operating at 200 kV.
The six samples analyzed in this set of TEM micrographs ( Figure 6) have similar shapes, sizes, and modes of aggregation of nanoparticles. Spheroidal nanoparticles (around 5-10 nm) were predominant among the samples; however, a small amount of larger faceted shapes (30-40 nm) were observed ( Figure 6).

Transmission Electron Microscopy
Powder samples were diluted in an aqueous medium and deposited on a 400-mesh carbon-coated copper grid. After the drying process, the samples were analyzed in a TEM (Jeol 2010 instrument), operating at 200 kV.
The six samples analyzed in this set of TEM micrographs ( Figure 6) have similar shapes, sizes, and modes of aggregation of nanoparticles. Spheroidal nanoparticles (around 5-10 nm) were predominant among the samples; however, a small amount of larger faceted shapes (30-40 nm) were observed ( Figure 6). The high-resolution micrographs of the six samples (Figure 7), show the atomic planes of spheroidal nanoparticles with size of around 5 nm. Figure 8 shows a high-resolution image of the NP4 sample, highlighting the atomic plane corresponded to the (220) family, with 0.29 nm spacing, of the magnetite spinel structure indexed with ICDD PDF: 01-075-0449, observed along the zone axis (111). The high-resolution micrographs of the six samples (Figure 7), show the atomic planes of spheroidal nanoparticles with size of around 5 nm. Figure 8 shows a high-resolution image of the NP4 sample, highlighting the atomic plane corresponded to the (220) family, with 0.29 nm spacing, of the magnetite spinel structure indexed with ICDD PDF: 01-075-0449, observed along the zone axis (111). The high-resolution micrographs of the six samples (Figure 7), show the atomic planes of spheroidal nanoparticles with size of around 5 nm. Figure 8 shows a high-resolution image of the NP4 sample, highlighting the atomic plane corresponded to the (220) family, with 0.29 nm spacing, of the magnetite spinel structure indexed with ICDD PDF: 01-075-0449, observed along the zone axis (111).  The electron diffraction (ED) pattern shown in Figure 9 indicates a typical structur of magnetite with spinel structure and a very little fraction of hematite, as shown in Figur 9e. These results show that the samples consist of an inverse spinel structure which corre sponds to magnetite. The impurity phase of hematite has been found in the NP4. Th mean particle size and distribution were evaluated by measuring the largest internal d mension of at a thousand particles of each sample. The six samples displayed a simila The electron diffraction (ED) pattern shown in Figure 9 indicates a typical structure of magnetite with spinel structure and a very little fraction of hematite, as shown in Figure 9e. These results show that the samples consist of an inverse spinel structure which corresponds to magnetite. The impurity phase of hematite has been found in the NP4. The mean particle size and distribution were evaluated by measuring the largest internal dimension of at a thousand particles of each sample. The six samples displayed a similar size distribution. They had an average size on the order of 5 nm (spheroidal morphologies), with the presence of a small number of larger particles of size around 30-40 nm (faceted shapes). The statistical analysis (of population of measured particles (N), mean size (nm), standard deviation (SD), standard error (SE), minimum size (Min), maximum size (Max), and range) were obtained using the Origin 6.0 software ( Figure 10 and Table 3). The electron diffraction (ED) pattern shown in Figure 9 indicates a typical structure of magnetite with spinel structure and a very little fraction of hematite, as shown in Figure  9e. These results show that the samples consist of an inverse spinel structure which corresponds to magnetite. The impurity phase of hematite has been found in the NP4. The mean particle size and distribution were evaluated by measuring the largest internal dimension of at a thousand particles of each sample. The six samples displayed a similar size distribution. They had an average size on the order of 5 nm (spheroidal morphologies), with the presence of a small number of larger particles of size around 30-40 nm (faceted shapes). The statistical analysis (of population of measured particles (N), mean size (nm), standard deviation (SD), standard error (SE), minimum size (Min), maximum size (Max), and range) were obtained using the Origin 6.0 software ( Figure 10 and Table  3).

Magnetometer and Scanning Magnetic Microscope
The magnetic maps were generated using a home-built SMM, with a reading syst containing Hall effect sensors connected to a reading system [33,34]. The SMM techniq consists of an improved method to produce magnetic maps for each magnetic field plied to the sample. In this case, the scan is made by moving the sample in a XY st inside the magnetic field, applied generated by an electromagnet. A drawing sketch of microscope assembly can be found on references [33,34]. The magnetic field induced fr the sample is detected by two Hall effect sensors, connected to standard amplifiers, filt and A/D converters to optimize the detection of the sample field. The reading system fixed on a printed circuit board and the sample, placed in a sample holder, moves in space through a system composed of two stepper motors. The applied magnetic field be set up to 0.5 T and has a magnetic moment sensitivity of approximately 10 −11 A [33,34].
Approximately tens of micrograms of MNPs were placed into a cylindrical cav measuring 400 µm in diameter and 400 µm deep (See Figure 11a). In the previous meth

Magnetometer and Scanning Magnetic Microscope
The magnetic maps were generated using a home-built SMM, with a reading system containing Hall effect sensors connected to a reading system [33,34]. The SMM technique consists of an improved method to produce magnetic maps for each magnetic field applied to the sample. In this case, the scan is made by moving the sample in a XY stage inside the magnetic field, applied generated by an electromagnet. A drawing sketch of the microscope assembly can be found on references [33,34]. The magnetic field induced from the sample is detected by two Hall effect sensors, connected to standard amplifiers, filters and A/D converters to optimize the detection of the sample field. The reading system is fixed on a printed circuit board and the sample, placed in a sample holder, moves in XY space through a system composed of two stepper motors. The applied magnetic field can be set up to 0.5 T and has a magnetic moment sensitivity of approximately 10 −11 Am 2 [33,34].
Approximately tens of micrograms of MNPs were placed into a cylindrical cavity measuring 400 µm in diameter and 400 µm deep (See Figure 11a). In the previous method, the magnetization curve was obtained by a technique that required several points along a two-dimensional map to obtain the magnetic moment. This method was improved by obtaining the magnetic moment using only two points of the 2D map, one in the region of minimal induced field strength of the sample and another the second point exactly in the region of maximum induced magnetic field strength of the samples. The minimal field measurement must be performed in the sample holder (away from the cylindrical cavity), as shown in Figure 11b. These two regions are necessary to obtain the sample magnetic moment using a theoretical model that considers the sample holder cylindrical shape [33,34,46]. region of maximum induced magnetic field strength of the samples. The minimal field measurement must be performed in the sample holder (away from the cylindrical cavity), as shown in Figure 11b. These two regions are necessary to obtain the sample magnetic moment using a theoretical model that considers the sample holder cylindrical shape [33,34,46]. For each magnetic field applied by the electromagnet, a magnetic scanning map of the induced field from the sample, from positive values of external fields to negative fields (i.e., 510 mT to −510 mT and returning to 510 mT), were produced and a hysteresis cycle was thus completed (Figure 12). From these 3D graphs, the maximum and minimum regions necessary to obtain the sample magnetic moment could be precisely obtained for each applied magnetic field, from where we can build a hysteresis curve. Despite the limitation of the magnetic field strength applied to the sample NP0, we can see that the hysteresis curve ( Figure 13) indicates a superparamagnetic behavior material, as there is neither coercivity nor remanence [33,34]. For each magnetic field applied by the electromagnet, a magnetic scanning map of the induced field from the sample, from positive values of external fields to negative fields (i.e., 510 mT to −510 mT and returning to 510 mT), were produced and a hysteresis cycle was thus completed (Figure 12). From these 3D graphs, the maximum and minimum regions necessary to obtain the sample magnetic moment could be precisely obtained for each applied magnetic field, from where we can build a hysteresis curve. Despite the limitation of the magnetic field strength applied to the sample NP0, we can see that the hysteresis curve ( Figure 13) indicates a superparamagnetic behavior material, as there is neither coercivity nor remanence [33,34]. Nanomaterials 2021, 11, x FOR PEER REVIEW 14 of 21 Figure 12. Magnetic maps of the sample NP0 in 3D, obtained directly from the maps in XY space from 500 mT to −500 mT. Figure 12. Magnetic maps of the sample NP0 in 3D, obtained directly from the maps in XY space from 500 mT to −500 mT.  This same NP0 sample was measured in a commercial Quantum Design PPM (model VersaLab), where the measurements were made in vibrating sample magnetom ter (VSM) mode. The comparison between the two measurements is shown in Figure  The curve composed of blue circles resulted from the measurement obtained by SMM PUC-Rio and the continuous curve, in red, is the measurement made by VSM, showing good agreement with each other. The average diameter of the MNPs could also be es mated. Although there is a size distribution, as observed in the TEM images, for this es mate, all particles were considered as having the same size, disregarding any interactio In this case, the Langevin equation was used, making a first-order approximation to lo magnetic fields [47,48]. That is: where is the magnetic susceptibility per volume, µ is the magnetic permeability vacuum, Ms is the saturation magnetization, V is the volume to be estimated, t Boltzmann constant, and T is the temperature (all measurements were made at room te perature, 293 K). Equation (3) can be rewritten as a function of the diameter (D), cons ering the particles are spherical, Thus, the average diameter of NP0 could be estimated. The following values we used to estimate the MNPs diameter, based on measurements made: = 2.7, obtained using a linear low field approximation of the MxH initial curve, (taken from blue circ Figure 13. Graph representing the measurements of the sample NP0. The curve composed of blue circles is taken from data obtained from the measurement made in the magnetic microscope of PUC-Rio and the continuous curve in red color is the measurement made in the commercial magnetometer VSM. This same NP0 sample was measured in a commercial Quantum Design PPMS (model VersaLab), where the measurements were made in vibrating sample magnetometer (VSM) mode. The comparison between the two measurements is shown in Figure 13. The curve composed of blue circles resulted from the measurement obtained by SMM at PUC-Rio and the continuous curve, in red, is the measurement made by VSM, showing a good agreement with each other. The average diameter of the MNPs could also be estimated. Although there is a size distribution, as observed in the TEM images, for this estimate, all particles were considered as having the same size, disregarding any interaction. In this case, the Langevin equation was used, making a first-order approximation to low magnetic fields [47,48]. That is: where x is the magnetic susceptibility per volume, µ 0 is the magnetic permeability of vacuum, M s is the saturation magnetization, V is the volume to be estimated, K B the Boltzmann constant, and T is the temperature (all measurements were made at room temperature, 293 K). Equation (3) can be rewritten as a function of the diameter (D), considering the particles are spherical, Thus, the average diameter of NP0 could be estimated. The following values were used to estimate the MNPs diameter, based on measurements made: x = 2.7, obtained by using a linear low field approximation of the MxH initial curve, (taken from blue circles on Figure 13), and using magnetite density ρ = 5.197 × 10 3 Kg m −3 [49] and M s = 754 kA m −1 (this value was also estimated by extrapolating the magnetic magnetization curve M in A/m as a function of 1/H magnetic field applied in A m −1 ), we obtain a diameter of 4.4 nm, according to Equation (4). The values for estimating the average diameter, the standard deviation σ for the two magnetic measurements, and the estimate using the other characterization techniques are summarized in Table 4. This diameter estimation is based on the magnetic properties of the material, considering only the nanoparticle magnetic core, subtracting the PL-127 shell. Although the estimated diameters on Table 4 are in the same order of magnitude, the differences among them are due to some factors: (i) the Raman measurements were taken at low powers, since phase changes occur for high laser powers, and this leads to a low-resolution spectra; (ii) TEM measurements shows a wide distribution of particle size, ranging from 1.5 to 39 nm; and (iii) the diameter estimation made by using Equation (4) is carried out by assuming that all particles are spherical and have the same size. Additionally, this calculation does not consider the Pluronic coating of MNPs, which might affect the TEM and Raman calculations.
This same method was also applied to samples NP1, NP2, NP3, NP4, and NP5. The same superparamagnetic behavior of these materials was observed for all Pluronic coated NPs (Figure 14, showing the NP2 maps), with no remanence and coercivity as well, shown in Figure 15a (NP2) and Figure 15b (NP5).
on Figure 13), and using magnetite density = 5.197 × 10 3 Kg m −3 [49] and Ms = 754 kA m −1 (this value was also estimated by extrapolating the magnetic magnetization curve M in A/m as a function of 1/H magnetic field applied in A m −1 ), we obtain a diameter of 4.4 nm, according to Equation (4). The values for estimating the average diameter, the standard deviation σ for the two magnetic measurements, and the estimate using the other characterization techniques are summarized in Table 4. This diameter estimation is based on the magnetic properties of the material, considering only the nanoparticle magnetic core, subtracting the PL-127 shell. Although the estimated diameters on Table 4 are in the same order of magnitude, the differences among them are due to some factors: (i) the Raman measurements were taken at low powers, since phase changes occur for high laser powers, and this leads to a lowresolution spectra; (ii) TEM measurements shows a wide distribution of particle size, ranging from 1.5 to 39 nm; and (iii) the diameter estimation made by using Equation (4) is carried out by assuming that all particles are spherical and have the same size. Additionally, this calculation does not consider the Pluronic coating of MNPs, which might affect the TEM and Raman calculations.
This same method was also applied to samples NP1, NP2, NP3, NP4, and NP5. The same superparamagnetic behavior of these materials was observed for all Pluronic coated NPs (Figure 14, showing the NP2 maps), with no remanence and coercivity as well, shown in Figure 15a (NP2) and 15b (NP5).   The values for magnetization at the limit of each magnetic field applied (see Figure  16a) are shown in Table 5 and summarized in Figure 16b. These results suggest that, after coating, there was a change in the magnetization curve in most samples. An important result of this magnetic characterization is that the Pluronic coating is performed during the mixing of iron salts and not after the production of NPs [5,17]. The results of the magnetic characterization curves, typical of these processes, are represented by a pronounced decrease in magnetization, as can be seen in Figure 16a, where we can also note that, even applying 2.0 T and after coating, all samples preserve the superparamagnetic behavior at room temperature.  The values for magnetization at the limit of each magnetic field applied (see Figure 16a) are shown in Table 5 and summarized in Figure 16b. These results suggest that, after coating, there was a change in the magnetization curve in most samples. An important result of this magnetic characterization is that the Pluronic coating is performed during the mixing of iron salts and not after the production of NPs [5,17]. The results of the magnetic characterization curves, typical of these processes, are represented by a pronounced decrease in magnetization, as can be seen in Figure 16a, where we can also note that, even applying 2.0 T and after coating, all samples preserve the superparamagnetic behavior at room temperature.

Conclusions
In summary, we successfully synthesized magnetite nanoparticles using the cocipitation method, coating them with a Pluronic F-127 surfactant, during the synthesi the mixture of iron salts at different concentrations. The structural properties of the s ples were characterized by Raman spectroscopy and transmission electron microsco

Conclusions
In summary, we successfully synthesized magnetite nanoparticles using the coprecipitation method, coating them with a Pluronic F-127 surfactant, during the synthesis, in the mixture of iron salts at different concentrations. The structural properties of the samples were characterized by Raman spectroscopy and transmission electron microscopy. From the Raman results, we verified that the characteristic bands of magnetite phase of Fe 3 O 4 were observed in the spectra, at 195, 293, 545, and 685 cm −1 . However, in some samples, we also observed some bands from the hematite phase, also confirmed by ED patterns results of the diffraction patterns acquired with the transmission electron microscope. The as-prepared samples, stored at approximately 20 • C, showed no change in the Raman spectrum, even after six months, indicating no oxidation of the material. In some samples, the hematite appeared during the synthesis, and it was observed even in the 0 months Raman spectrum, indicating that there was no oxidation or phase change throughout time. Using the Raman spectrum and TEM images, it was possible to estimate the diameter of the synthetized nanoparticles, with values of 7.8 nm and 5.4 nm, respectively.
The magnetic characterization of the coated nanoparticles was carried out by using a commercial vibrating sample magnetometer and a home-built scanning magnetic microscope. The novel SMM technique uses a set of Hall sensors for measuring three-dimensional magnetic maps of a sample. By this measurement, the values of the magnetic moment of a sample, in a cylindrical cavity, for an applied magnetic field, can be extracted and it is possible to build a MxH hysteresis loop. The magnetization curves from the SMM data are compared to a VSM measurement, with very good agreement. Both SMM and VSM measurements shows that all samples exhibit a superparamagnetic behavior, suitable for biomedical applications. Furthermore, by the magnetic measurements it was also possible to estimate the nanoparticles' magnetic core diameter about 4.4 nm, which is consistent with TEM and Raman measurements.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.