In the thermal analyses TG-DTG/DSC in Fig. 1, it's possible to verify that the xerogel lost a total of 75% of mass, a typical behavior in the decomposition of polymeric resins.
Initially, an endothermic peak was observed at approximately 74°C, corresponding to the elimination of residual water in the sample, followed by the decomposition and oxidation of the organic phase, which occurs in two distinct stages, marked by abrupt changes in the slope of the TG curve, with clear peaks indicated by the maxima of the DTG at 150°C and 340°C. It can be observed that the width of the peak at 340°C in the DTG is relatively narrow, indicating that the evolution of decomposition/oxidation of the organic phase at this temperature is relatively rapid, initiating the crystallization process, unlike the first peak (150°C) where the decomposition and oxidation of the organic phase are slower, possibly associated with the breakage of complex polymeric chains.
Through DSC analysis, the maximum exothermic peak at 330°C and the endothermic peak at 350°C, accompanied by a mass loss of approximately 15%, possibly associated with the transformation of the crystalline phase from magnetite (Fe3O4) to hematite (α-Fe2O3) (the more stable polymorphic phase of iron oxide), followed by the complete decomposition of the xerogel, above 350°C, no thermal events or significant changes in mass loss were observed, indicating that the system is stable.
From the analysis of TG-DSC/DSC results, it was found that the thermal treatment temperature where the synthesized magnetite phase possibly retained by sol-gel is between 147°C and 330°C.
Therefore, thermal treatments were carried out between 150°C and 300°C for 4 hours without atmosphere control, in order to evaluate the onset of magnetite crystallization. In Fig. 2, the X-ray diffractograms of the xerogel thermally treated at 150°C, 200°C, and 300°C for 4 hours are presented.
The thermal treatment at 150°C revealed little crystallization of the sample, indicating that there was no onset of magnetite phase crystallization at this temperature, corroborating the TG-DTG/DSC analysis in Fig. 1, where the first exothermic peak is solely associated with the decomposition of the organic phase with a 60% mass loss. With the temperature increase to 200°C, it is possible to observe the presence of all diffraction peaks related to magnetite, namely (220), (311), (222), (400), (422), (511), and (440) according to crystallographic card PDF 75–449, indicating that the onset of magnetite crystallization occurs shortly after the elimination of the organic phase and these are not concomitant events. The peak widths suggest the presence of magnetite nanoparticles. At 300°C, the formation of magnetite phase was also observed, with approximately 25% presence of hematite phase (PDF 85–599), confirming the TG-DTG/DSC analysis in Fig. 1 regarding the temperature range between 300 and 350°C where the complete phase transformation from magnetite (Fe3O4) to hematite phase (α-Fe2O3) would take place.
Thus, 200°C was defined as the minimum temperature for the crystallization and stabilization of magnetite without atmosphere control. One parameter taken into consideration was the fact that, according to the thermal analyses in Fig. 1, at 200°C, the organic phase of the xerogel was not completely oxidized and decomposed, with 15% of organic material remaining. To overcome this situation, the material treated thermally at 200°C was subjected to a chemical washing step using acetic acid (CH3COOH) at a concentration of 0.3 M. Figure 3 presents a comparison of the diffractograms, with and without chemical washing of the NPs obtained at 200°C.
It is possible to observe that the diffraction peaks became more evident after chemical washing, revealing that the 0.3 M acetic acid used favored the elimination of residual organic phase at 200°C. Similar to Manami et al. [19], who reported the use of chemical washing to promote the precipitation of iron oxyhydroxide. Meanwhile, Ali et al. [20], to promote magnetite synthesis, prepared a nanoemulsion containing iron sources and sodium hydroxide, and finally, chemical washing with ethanol was carried out to exhibit superparamagnetic behavior with magnetization values.
Once the minimum crystallization temperature and parameters for eliminating residual organic phase were defined, different holding times were proposed: 4, 8, 16, 24, and 48 hours at 200°C, aiming to evaluate the effect of the thermal treatment time under uncontrolled atmosphere conditions on the structural, microstructural, and physical properties of magnetite.
The XRD analysis of the NPs subjected to different thermal treatment times presented in Fig. 4 (left) indicates the maintenance of the complete integrity of the magnetite phase for up to 8 hours of holding time at 200°C without atmosphere control. Above 16 hours, the formation of some traces of hematite phase (α-Fe2O3) (PDF 85–599) can be observed, as evidenced by the presence of the peak at 2θ = 33.1 corresponding to the crystallographic plane (104), present at a concentration of approximately 22.5% (Eq. 1). It is possible to observe a slight increase in the concentration of hematite phase with increasing holding time, reaching concentrations of 24.0% and 25.4% in the treatments of 24 and 48 hours, respectively, as listed in Table 1. It's noticeable the narrowing of all crystallographic peaks with increasing holding time, suggesting the enlargement of NP sizes, as observed in the analysis in Fig. 4 (right), which zooms in on the peak of highest intensity, corresponding to the crystallographic plane (311) adjusted using a Lorentz-type function. This indicates a smaller full width at half maximum (FWHM), and consequently, larger Scherrer crystallite size, as can be observed by calculating the crystallite size using the Scherrer equation (Eq. 2), as presented in Table 1.
Xu et al. [7] also observed this effect for sol-gel synthesized magnetite using iron nitrate and ethylene glycol as precursors. The authors noted that with the increase in thermal treatment temperature in the range between 200°C and 400°C, the full width at half maximum decreases and the reflection peaks become narrower, indicating larger crystallite sizes and better crystallinity of the phase.
Table 1
Percentage values of hematite phase, crystallite size according to Scherrer euqation, average particle size from SEM and TEM micrographs .
Holding time (hours) | % Magnetite Phase* | Crystallite size (nm) | SEM average particle size (nm) ** | TEM average particle size (nm) *** |
4 | 100 | 25 | 35 ± 3 | 18 ± 0,1 |
8 | 100 | 30 | 39 ± 5 | |
16 | 77.5 | 32 | 43 ± 4 | 32 ± 0,5 |
24 | 76.0 | 43 | 49 ± 3 | - |
48 | 74.6 | 52 | 54 ± 4 | 46 ± 1,1 |
*Calculated according to Eq. 1, ** Based on SEM micrographs analysis, LogNormal fitting, *** Based on TEM micrographs analysis, LogNormal fitting.
From the microstructural analysis presented in the SEM micrographs in Fig. 5, it is possible to observe a high agglomeration degree, inherent to the synthesis method. Soft, easily pulverizable clusters with a heterogeneous size distribution are observed. It was also noted that at all evaluated holding times, the nanoscale size of the particles was retained, but there is evident nanoparticle growth with increasing holding time.
It's possible to identify that the clusters consist of equiaxial primary particles with sizes between 35 ± 3 nm at 4 hours of holding time and 54 ± 4 nm at 48 hours, as presented in Table 1 and depicted in the size vs. time curve in Fig. 5(f). Furthermore, it was noticeable that with longer holding times, the particles acquired a more spherical shape due to the higher energy expended in the process, assuming the thermodynamically more stable form, as verified by Cotica et al. [18]. The crystallite size was in line with the calculated average particle size inferred from SEM micrographs analysis. In this case, the same trend of larger average particle sizes with increasing holding time of the thermal treatment was observed.
From the microstructural analysis using transmission electron microscopy (TEM) in Fig. 6, the nanoscale nature of the particles is highlighted. With a holding time of 4 hours, as shown in Table 1, the estimated average size of primary particles was 18 ± 0.1 nm, while with longer holding times, the estimated size of primary particles tripled in size, with an estimated value of 46 ± 1.1 nm, as shown in the histograms of distribution measurements in dark-field micrographs in Fig. 6 (bottom).
The SAED inset in the bright-field micrographs of 4 and 16 hours shows a characteristic pattern of continuous rings typical of nanocrystalline systems, while at 48 hours, a polycrystalline pattern composed of regular spots associated with larger crystals is observed, corroborating the difference in particle size with increasing holding time.
From the BET analysis, it is possible to observe the trend of decreasing surface area with increasing holding time, while the values of equivalent spherical diameter calculated using Eq. 2 were close to those obtained by TEM analysis. This result indicates the high dispersion of magnetite NPs obtained under all evaluated holding time conditions.
Shaker et al. [21] studied the synthesis of magnetite with ethylene glycol at different thermal treatment temperatures ranging from 200 to 400°C, where an average particle size in the range of 30 to 90 nm was observed. Temperatures above 400°C failed to retain the magnetite phase, resulting in the formation of hematite phase. On the other hand, Nkurikiyimfura et al. [22] investigated the synthesis of magnetite by co-precipitation-reduction, where the average particle diameter was around 11 nm, exhibiting superparamagnetic behavior as expected.
Table 2
Surface area and equivalent spherical diameter of the NPs obtained at different holding times.
Holding time (hours) | Surface area (BET) (m2/g) | Equivalent spherical diameter (nm) * |
4 | 106.4 | 11.0 |
8 | 114.6 | 10.0 |
16 | 105.0 | 11.0 |
24 | 104.1 | 11.0 |
48 | 67.2 | 17.0 |
* Eq. 3 |
The magnetite NPs obtained at 200°C for 8 and 48 hours were dispersed in isopropyl alcohol, distilled water, and ethylene glycol via ultrasonication for 5 minutes at room temperature. All resulting suspensions responded to an external magnetic field, as shown in the photographs of Fig. 7. From the assays, it was possible to observe the high dispersion of both samples (8 and 48 h) in the three liquid media evaluated, indicating their reduced particle size, an important characteristic, especially for suspensions in water, specifically for biomedical applications, where a high dispersion of magnetite NPs in body fluids is required. In the case of the magnetic fluid used in cancer therapy, which requires high dispersion and suspension stability for a certain period, in order to improve heating efficiency to ensure the elimination of tumor cells. The total collection of NPs obtained in 8 hours when subjected to an external magnetic field occurred at times of 5, 15, and 120 minutes in alcohol, water, and ethylene glycol media, respectively. For the NPs obtained in 48 hours, which exhibit 25.4% hematite phase (α-Fe2O3), according to XRD analysis (Fig. 4), as expected, the total collection times were doubled, being 15, 30, and 240 minutes in alcohol, water, and ethylene glycol media, respectively. These results indicate the good magnetic behavior of the obtained NPs.
The stability of the suspensions was observed through the precipitation of the samples over a period of 2 days, as shown in Fig. 8. These results indicate that the synthesized NPs offer a long suspension stability characteristic both in water and in ethylene glycol, an important feature for applications where prolonged dispersion is required, such as in magnetic fluids used in cancer therapy as mentioned earlier.
In Fig. 9, the magnetization curves (emu/g) as a function of the applied magnetic field (Oe) of the hematite NPs obtained at different holding times are presented. The magnetization curves revealed the superparamagnetic behavior of all NPs, which was expected for materials with particle sizes below 50 nm. As reported by several authors, superparamagnetic behavior is common in magnetite nanoparticles [7, 19, 25, 26]. Table 3 presents the values of saturation magnetization, remanent magnetization, and coercive field of the magnetite NPs subjected to different thermal treatment times (4–48 hours), resulting in different initial particle sizes ranging from 30 nm to 50 nm.
It is possible to observe an increase in saturation magnetization when the sample is subjected to holding times of 8 hours, possibly due to the increased crystallinity of the magnetite phase. On the other hand, for holding times above 16 hours, there is a gradual decrease in saturation magnetization, associated with the presence of hematite phase in the sample, corroborating the XRD results presented in Fig. 4.
Table 3
Magnetization properties NPs
Holding time (hours) | Inital Particle size (nm) | Saturation magnetization (emu/g) | Remanent magnetization (emu/g) | Coercitive field (kOe) |
4 | 35 | 13 | 0.82 | 22.5 x10− 3 |
8 | 39 | 57 | 0.90 | 6.1x10− 3 |
16 | 43 | 5 | 0.31 | 13.95 x10− 3 |
24 | 49 | 30 | 0.42 | 25.35 x10− 3 |
48 | 54 | 13 | 0.085 | 25.14 x10− 3 |