1. Introduction
Cancer is a disease, which is become one of the major public health concerns in the present world. Magnetic hyperthermia treatment is one of the promising non-invasive approaches for therapy on cancerous tumors upon exposure to external alternating magnetic fields (AMF). Magnetic nanoparticles (in the range of 10 to 100 nm) can generate heat through the oscillation of their magnetic moments. Magnetic hyperthermia can then kill completely cancer cells by elevating their temperature between 42˚C and 46˚C with minimal injury to normal cells. This is an exciting potential therapy as it could replace or reduce the need for more destructive therapies such as chemotherapy [1] [2] . The most used materials for magnetic hyperthermia are ferrite nanoparticles in particular magnetite (Fe3O4) or maghemite (γ-Fe2O3). When exposed to an alternating magnetic field the magnetic nanoparticles produce heat via two main mechanisms, one is hysteresis loss and the other is relaxational loss (Neel and Brownian relaxations). The contribution of each process depends strongly on the nanoparticle size. For example, in small nanoparticles (20 nm), heat dissipation is due mainly to the Neel and Brownian relaxations with negligible contribution of hysteresis loss. For larger single-domain particles, hysteresis losses are significant.
In general, the heat generated per unit gram of magnetic material and per unit time under alternating magnetic field is known as specific absorption rate (SAR). Previous reports showed that the SAR values of the nanoparticles could be affected by several parameters such as sample preparation method, size of nanoparticles, magnetic properties, the amplitudes and frequency of the applied field, coating, etc. [3] [4] [5] [6] [7] .
Most of the nanomaterials dedicated for magnetic hyperthermia are synthesized by chemical methods, due to the uniformity and the possibility to tune the size of nanoparticles. However, such methods use organic solvent and reducing agent in the synthesis process in addition to the difficulty to fabricate huge quantities. Others methods which are low-cost and friendly to the environment attracted an increasing interest during the last decades. Among these preparation methods, mechanical alloying is widely used to prepare nanocrystalline of a variety of materials as result of its simplicity, low cost and its capability to produce large volumes [8] . Lemine et al. [9] fabricated Co1-xFexO4 Nanocrystalline for magnetic hyperthermia. They showed that the nanocrystalline dissipated heat under an AMF and this heat is able to kill cancerous cells. Ban et al. [10] used ball milled nickel-copper nanoalloys for magnetic hyperthermia. They investigated the effect of milling time for the optimization of the best sample, which gives the highest heating ability. Nanoalloys Cu27.5Ni72.5 (at%) alloy with a crystallite size of around 10 nm showed the best heating ability under an AMF.
The aim of this study is to prepare Gd2O3/γ-Fe2O3 nanocomposite by using ball milling technique for possible use in magnetic hyperthermia application. The effect of the milling time on the structure, morphology, magnetism and heating efficiency under an AMF of obtained nanocomposite will be investigated.
2. Materials and Methods
Commercially powders of Gd2O3 and maghemite (γ-Fe2O3) were introduced into a stainless-steel vial with balls in Fritsch machine (P7) for the preparation of Gd0.05Fe1.95O3 nanocomposite. Two different milling times were considered (5 h and 30 h). The balls to powder ratio was fixed to 20:1. The rotation speed was 300 rpm and the milling process was interrupted after every hour for ten minutes to prevent overheating. As-milled powder samples were characterized using X-ray diffraction (XRD) and vibratory sample magnetometry (VSM).
X-ray powder diffraction (XRD) measurements were performed using Bruker D8 Discover diffractometer (θ-2θ) equipped with Cu-Kα radiation (λ = 1.5406 Å). The average crystallite size, D, of all the samples was calculated from Scherer’s equation [9] without considering the effect of lattice strains
(1)
where λ is the incident wavelength (1.5406 Å), θ is the Bragg angle (in degree), K is a constant whose value is approximately 0.9 and β (rad) is the full width at half maximum (FWHM) of a diffraction peak or integral breadth [11] [12] .
The morphology was investigated through Field Scanning Electron Microscopy (FE-SEM, JEOL-6700F).
Magnetic measurements were measured at room temperature using Lake Shore 7404 model vibrating sample magnetometer (VSM) having a 1.8 T magnet. The magnetic parameters (i.e., saturation magnetization (Ms), remanence magnetization (Mr), and coercivity (Hc)) were determined from the M-H curves.
The heating efficiency of the samples was carried out with a commercial system “Nanotherics Magnetherm” under alternating current (AC) magnetic field of 170 Oe and 332 kHz. The samples were dissolved in distilled water and sonicated during 10 min. The temperature of particle suspension is measured with a fiber optical thermometer and registered with data acquisition system. The temperature increase in the samples is measured during 20 min.
The amount of heat generated by magnetic nanoparticles is normally quantified in terms of the SAR, which can be calculated by [13] :
(2)
where Cw is the specific heat capacity of water (4.185 J/g∙k), ρ is the density of the colloid, mNP is the concentration of the magnetic nanoparticles in the suspension and ΔT/Δt is the initial slope of the time-dependent temperature curve. This gives SAR the units W/g [14] . The fundamental problem with this parameter is that SAR is extrinsic and varies depending on both H (field strength) and f (frequency). Therefore, measurements are only comparable if made on the same machine that has the same H and f [15] . We have considered the rise in temperature for the initial 30 s to calculate the slope.
3. Results and Discussions
3.1. Structure, Morphology and Magnetic Properties
The X-ray diffraction patterns for Fe2O3 and γ-Fe2O3/Gd2O3 powders as received and milled for several times are shown in Figure 1. Practically, the patterns of as-received γ-Fe2O3 powder shows a series of five clear diffraction peaks with the strongest one (311) plane occurring at 2θ = 35.78˚. The other clear peaks have occurred at 2θ = 30.52˚, 43.46˚, 57.56˚, and 63.12˚ which assigned to (220), (400), (511), and (440) planes respectively. With increasing milling time, the diffraction peaks became broader and their relative intensity decreases. This effect is a typical behavior of materials after milling and attributed usually to the presence of crystallites with small sizes and due to the strain induced by milling. The most dominant crystalline phase is γ-Fe2O3 but it is important to notice the presence of two clear peaks due to Gadolinium oxide crystal phases which have occurred at 2θ = 28.62˚ and 30.32˚ which corresponding to (222) and (400) planes respectively. For longer milling time, progressive disappearance of some diffraction peak of γ-Fe2O3 are observed. For increasing milling time, the shift of peak position towards lower angle (2θ) is observed. An example of this angle shift is highlighted by considering one of characteristic peak corresponding to (311) plane (Figure 1) This shifting of peaks toward the lower angle indicates an increase of lattice parameters with increasing milling time, so, Gd is incorporated with Fe lattice and produced γ-Fe2O3/Gd2O3 as a nanocomposite.
The crystallite size changes monotonically with the increase of milling time (Table 1). When the milling time is increased from 5 h to 30 h, an increase of crystallite size is observed. This behavior might be due to the agglomeration of crystallite size for longer milling.
![]()
Figure 1. XRD patterns of (a) γ-Fe2O3, as received (b) γ-Fe2O3/Gd2O3 nanostructure for 5 h milling time and (c) γ-Fe2O3/Gd2O3 nanostructure for 30 h milling time.
![]()
Table 1. Average crystallite size of γ-Fe2O3 and γ-Fe2O3/Gd2O3 nanostructures for different milling times.
The SEM images of the samples milled for 5 h and 30 are depicted in Figure 2. It can be seen that morphology of nanocomposites milled for 5 h shows that the shape is quasi-spherical, the size is in the nanometer scale and the particles are relatively agglomerated (Figure 2(a)). After milling time of 30 h (Figure 2(b)), the particles are agglomerated compared to the sample milled for 5 h.
The vibrating sample magnetometer (VSM) was used to determine the magnetic properties of samples as received and ball milled for different milling times. Figure 3 shows the magnetization-field (M-H) curves recorded at room temperature for the as-received Fe2O3 and milled powders. It can be seen that γ-Fe2O3 is completely ferromagnetic. It can be seen also that presents hysteresis with small coercivity and remanence. The magnetic parameters deduced from M-H curves are summarized in Table 2.
![]()
![]()
(a) (b)
Figure 2. SEM images of (a) sample milled for 5 h, and (b) milled for 30 h.
![]()
Figure 3. Magnetization (M) versus magnetic field (H) curves of (a) γ-Fe2O3, (b) γ-Fe2O3/Gd2O3 nanostructure for 5 h milling time and (c) γ-Fe2O3/Gd2O3 nanostructure for 30 h milling time.
![]()
Table 2. Ms, Mr and Hc for γ-Fe2O3 and γ-Fe2O3/Gd2O3 nanostructures for different milling times.
Table 2 indicates the increases of coercivity with increasing milling time, while the saturation decreases. The increase in Hc for the nanocomposite with increasing milling time can be explained by magnetic anisotropy induced by ball milling. Regarding the decreases of the magnetic saturation (Ms), that might be attributed to the site preference of Gd ions. Increasing milling time might affect Fe-Gd interactions and that will affect the magnetic behavior of the nanocomposite. Zhang et al. reported similar behavior of Gd doped magnetite nanoparticles [16] .
3.2. SAR Measurements
The heating efficiency is studied for the as-prepared nanostructures with concentration of 10 mg/mL at applied field of 170 Oe and frequency of 332 kHz. Figure 3 shows the temperature increase due to the nanocomposite when submitted to an alternating magnetic field. It can be seen from Figure 4 that Fe2O3/Gd2O3 5 h milling time has best heating compared to that of 30 h milling time. Furthermore, the value of SAR calculated by Equation (2) for 5 h milling time nanocomposite (68 W/g) is higher than that obtained for 30 h milling time nanocomposite (28.7 W/g). 5 h milling time nanocomposite reached 42˚C (hyperthermia temperature) in just 5 min 46 s compared to that of 30 h milling time nanocomposite which reached this temperature in 12 min 22 s.
Comparison of SAR values with similar works showed that both samples show relatively a good heating ability (Table 3). SAR obtained for both samples are
![]()
Figure 4. Temperature vs. time of (a) γ-Fe2O3/Gd2O3 nanostructure for 5 h milling time and (b) γ-Fe2O3/Gd2O3 nanostructure for 30 h milling time.
![]()
Table 3. Comparison of SAR values for different magnetic ferrofluids.
higher than that reported for ball milled chitosan coated Gd2O3-Fe2O3 nanocomposite (18.5 W/g) [17] . The heating efficiency is also better than PAA coated magnetite and Fe2O3-TiO2 nancomposite [19] [20] .
4. Conclusion
Ball milling technique was used to prepare γ-Fe2O3/Gd2O3 nanocomposite for possible use in magnetic hyperthermia application. In order to investigate the effect of milling time on this sample, two samples were prepared for different milling time. Milling for longer time induces an increase of the coercivity, a decrease of magnetic saturation, an increase of the crystallite size and a decrease of heating efficiency of nanocomposites. It was also observed that the two samples present hysteresis with small coercivity and remanence. The results obtained in this study show that the obtained nanocomposite for 5 h milling time is a promising candidate for magnetic hyperthermia due to its properties which show high heating efficiency and interesting magnetic behavior.