Influence of clustering on the magnetic properties and hyperthermia performance of iron oxide nanoparticles.

Clustering of magnetic nanoparticles can drastically change their collective magnetic properties, which in turn may influence their performance in technological or biomedical applications. Here, we investigate a commercial colloidal dispersion (FeraSpin (TM) R), which contains dense clusters of iron oxide cores (mean size around 9 nm according to neutron diffraction) with varying cluster size (about 18-56 nm according to small angle x-ray diffraction), and its individual size fractions (FeraSpin (TM) XS, S, M, L, XL, XXL). The magnetic properties of the colloids were characterized by isothermal magnetization, as well as frequency-dependent optomagnetic and AC susceptibility measurements. From these measurements we derive the underlying moment and relaxation frequency distributions, respectively. Analysis of the distributions shows that the clustering of the initially superparamagnetic cores leads to remanent magnetic moments within the large clusters. At frequencies below 10(5) rad s(-1), the relaxation of the clusters is dominated by Brownian (rotation) relaxation. At higher frequencies, where Brownian relaxation is inhibited due to viscous friction, the clusters still show an appreciable magnetic relaxation due to internal moment relaxation within the clusters. As a result of the internal moment relaxation, the colloids with the large clusters (FSL, XL, XXL) excel in magnetic hyperthermia experiments.