Transport properties of cluster-assembled magnetic nanostructures

The aim of this thesis is to get a deeper understanding of magnetic properties of small clusters when they are embedded in a host material. The study was motivated by the observation of the very peculiar properties that clusters present in the gas phase or when deposited on a substrate. The intriguing perspective is that those properties would be preserved in a solid state sample that could be used for technological applications. Preformed and well characterized clusters are used as building blocks to produce nanostructured films in which they are embedded in a host material in order to preserve them from chemical and thermal degradation when the sample is exposed to ambient conditions. For this purpose an innovative set-up has been constructed that allows the independent control of cluster size, concentration and chemical composition. In particular samples containing small magnetic clusters embedded in a non magnetic matrix have been produced in order to study the evolution of the magnetic behaviour of such cluster assembled materials. These samples have been subsequently studied by means of magnetotransport, in particular by measuring their magneto-resistance and Hall voltage as a function of temperature in magnetic fields of up to 5T. Additionally, a novel measurement protocol detecting the derivative of the resistance with respect to temperature as a function of magnetic field has been used to characterize the samples. Magnetotransport is known to be an important tool for the investigation of the magnetic behaviour as the conduction electrons can be considered as microscopic probes of the state of the sample. However, even though these properties have been studied extensively, a complete and universally accepted theory on the mechanisms giving rise to the observed magneto-resistance in cluster assembled materials is still to be found. In this work, the high quality of the produced samples allows an unequivocal identification of all the aspects that are not correctly described by the theories in use. In particular it is observed that, in the case of small clusters, the effect of magnetic interactions cannot be neglected even at very low concentrations. The consequence is twofold since both the superparamagnetic model of magnetization and the well known (1 – m2) expression for the magneto-resistance fail if a correlation between magnetic moment exists. Furthermore we claim that the Mott hypothesis of two parallel currents adopted for electrical conduction is not sufficient to describe the details of transport in granular systems. In fact the particular differential resistance measurement introduced in this work allows an underlining of the importance of the spin-channel mixing mechanisms that are not taken into account in the Mott picture.

Related material