Dynamic cantilever magnetometry of reversal processes and phase transitions in individual nanomagnets

Magnetic materials constitute a field of research covered by continuous and large interest. Their broad use in various areas and their potential adaption in novel and promising applications motivates the search for new materials and investigation on their fundamental properties. Magnetic samples are in fact candidates for applications in various fields. Especially their potential adaption in high density magnetic storage media increased in the recent years the interest in investigating different types and shapes of magnetic materials. Potential carriers of information could be for instance domain walls or magnetic skyrmions. A fundamental knowledge about the stability, the size and the controllability is needed before any kind of magnetic material can be implemented into a device.
The motivation of this thesis is to give a contribution to the understanding of the growing, interesting and versatile field of magnetic materials by using dynamic cantilever magnetometry (DCM), which is sensitive, non-invasive and compatible to a broad range of samples with different properties. Using DCM we provide both complimentary and new information about the properties of novel or well-known magnetic materials, taking advantage of the high sensitivity of nanomechanical resonators.
In fact, we demonstrate the use of DCM, in combination with numerical simulations, to study the process of magnetic reversal of short ferromagnetic (FM) nanotubes (NTs). This reversal is driven by the nucleation of vortices at the end of the tubes and shows a strong aspect-ratio dependence. For instance, short FM NTs go through a vortex state and tubes above a certain length through a mixed state composed of a vortex at each end of the tube and an axial domain in the center. Apart from the aspect ratio, we find that the shape of the ends and impurities have a significant influence on the starting of the reversal process and how it evolves. A precise knowledge of the magnetic reversal is important, since a reliable process is needed for applications like magnetic storage.
We further investigate the stability of the skyrmion lattice phase of two different materials carrying the two different types of skyrmions. We measure MnSi nanowires (NWs) hosting Bloch-type skyrmions and a GaV4S8 single crystal carrying Néel skyrmions. During the studies of MnSi NWs we observe an unexpectedly stable skyrmion phase, extending from 29 K down to at least 0.4 K. The stability of the skyrmion lattice is strongly dependent on the applied magnetic field direction. An extended skyrmion lattice is present only if the long axis of the NW is parallel to the magnetic field. This stabilization occurs despite of the fact that the dimensions of the NW are too large to spatially confine the skyrmion lattice. In fact, an important difference between a NW and a single-crystal bulk sample is that a NW has an especially large surface-to-volume ratio for surfaces perpendicular to the long axis of the wire. The demagnetization influence of theses surfaces may produce an effective magnetic anisotropy which suppresses the conical phase, and therefore, in combination with a parallel applied magnetic field, stabilize and extend the skyrmion lattice.
The investigation of a single-crystal of GaV4S8 also results in a surprising stability of the skyrmion lattice phase. In this case, we study the occurrence of the skyrmion phase depending on the orientation of the applied magnetic field. We find the skyrmion lattice still formed with a misalignment of 77:1˚ ± 2:3̊ between the [111] crystal axis, which is the axis along which the skyrmions form in this material, and the applied magnetic field direction. The stability of the skyrmion phase is correlated with the strength of the uniaxial anisotropy present in this material. Since the observed stability is higher than what the theoretical model predicts, for the moment we cannot conclude on the strength of the anisotropy, but the findings could help to refine the model, so that it may include all relevant interactions and fit the experimental observations.