Iron oxide nanoparticles (NP) are of great interest in nanomedicine. They are used in hyperthermia, tumor targeting, drug delivery therapy and in Magnetic Resonance Imaging (MRI) as negative contrast agent. Because NP’ size and magnetization play a key role in their behavior as MRI contrast agents, these parameters need optimal characterization methods.
In this work, we explored the characterization by magnetometry, in particular the effects of temperature, size distribution and anisotropy on magnetization of different-sized magnetite NP (Fe3O4). With a Vibrating Sample Magnetometer (VSM), we carried out magnetization as function of the magnetic field (MH curves) at 100, 200, 275, 300 and 315 K. Transmission electron microscopy (TEM) was also performed to measure directly the size distribution and agglomeration of the NP.
By fitting the MH curves with Langevin law, we observed that the NP’ size parameter was decreasing with the decrease of temperature in both simple Langevin and size distribution model. We attempted to explain this effect : The first hypothesis was that, at low magnetic fields, Néel relaxation intensity varies with the temperature. This would explain the differences in curves’ slope and so the size parameter. Nevertheless, the size decrease also occurs above 273 K where Brown relaxation should totally erase Néel relaxation effect. To test the second hypothesis we verified that the MH curves follow well the Langevin law by plotting the normalized magnetization (M/Msat) as function of the magnetic field normalized by the temperature (B/T). If they follow well the Langevin law, all the curves have to be superimposed, which is the case. Except for the 100 K MH curve which is the only one to diverge slightly from the other curves. Finally, we tried to fit the MH curves with a non-zero uniaxial anisotropy model used by Respaud and al but this model didn’t succeeded to fit our MH curves.
In conclusion, we have to test others hypotheses like the stability of the fit itself and/or to test others anisotropy model like cubic anisotropy to understand completely this phenomena. It is difficult to characterize nanoparticles by magnetometry. Because the size obtained is sensitive to the temperature, it is touchy to arbitrary choose one specific temperature. However, MH results at room temperature are in good agreement with TEM measurements, especially for the size distribution width.
 M. Respaud and al., Surface effects on the magnetic properties of ultrafine cobalt particles, Physical Review B, 57, 5: 2925-2935 (1998)