Patterned nanomagnets - magnetic metamaterials and vortex dynamics
MetadataShow full item record
Sustaining the rapid development of information technology will eventually require new types of devices that are smaller, faster and more power efficient. Although the transistor is, and will continue to be, tremendously successful, its operation is inherently linked to resistive losses. However, new device and computational concepts based on magnetism are promising since these may operate without movement of electrical charges. Realizing these concepts benefit from advances in the understanding and control of magnetic phe nomena and materials. This work has been concerned with fabrication and characterization of micro- and nanoscale magnetic systems and is divided on two main topics. The first part is on dipolar magnetic metamaterials, a class of materials where the magnetic properties are determined through nanofabrication. The ability to directly tailor the magnetic properties of these materials make them interesting from both a fundamental viewpoint as well as for applications. The second part is on the dynamics of magnetic vortices, which have been suggested as possible magnetic memory elements. In a dipolar magnetic metamaterial long-range magnetic order is stabilized by the dipolar interaction. In contrast, the exchange interaction is the main stabilizing factor in conventional magnets. These magnetic metamaterials are built up from nanomagnets. Each nanomagnet is, due to the small size, in a single domain state with the magnetization of all atoms aligned. This results in a net magnetization and a significant stray field. The stray field of each magnet can to first order be approximated to that of a dipole, hence the name dipolar magnetic metamaterial. The dipolar field results in coupling of neighboring nanomagnets, and if they are placed in a regular pattern, the magnetization may order across the entire array. The exper imental realization of such dipolar magnetic metamaterials is challenging since the nanomagnets must be small to be in a monodomain state and be densely packed. Dense packing is required for the dipolar interaction to be sufficiently strong to overcome disordering by thermal energy or disorder from defects. A significant part of the experimental work in this thesis has therefore been to optimize the fabrication process to find the parameters that result in metamaterials with the desired properties. The dipolar magnetic metamaterials were made using state-of-the-art elec tron beam lithography (EBL) for patterning and electron beam evaporation for thin-film deposition at NTNU NanoLab. EBL is an ideal technique for prototyping magnetic metamaterials since the geometric parameters can be readily controlled with precision of a few nanometers. In this work, we realize systems in which the magnetic order is determined by the lattice symmetry. In line with theoretical predictions we find long-range ferromagnetic order for ferromagnetic nanodisks on a hexagonal lattice and antiferromagnetic ordering on the square lattice. We observe that the mag netization of the disks selects certain magnetization axes from a degenerate manifold. By analyzing the shape of the individual disks using scanning elec tron microscopy the anisotropy was attributed to minute deviations from circularity in the disk shape. The strong coupling of the magnetic proper ties to the geometric properties in the dipolar magnetic metamaterials is a double edged sword. On the one hand it enables tailoring of the magnetic properties when the fabrication process is well controlled. On the other hand it results in the arrays being sensitive to minute process variations. We show how magnetic anisotropy can be purposely engineered in these magnetic metamaterials through changing the distance between the nano magnets in one direction. The square lattice was stretched to a rectangular lattice, resulting in the antiferromagnetic stripes to align to the direction with closest packing. Furthermore, by stretching the hexagonal lattice to face-centered rectangular we control both the size and direction of the fer romagnetic domains. The resulting domain pattern is found to follow a modified Kittel’s law, where the increase of disk spacing result in a change in all contributions relying on the magnetostatic energy. This involves both the demagnetization energy of each domain and the domain wall energy cost. Our work demonstrates that dipolar magnetic metamaterials with tailored properties can be prepared with state-of-the-art EBL. This straight forward approach to tailor macroscopic magnetic properties through nanopatterning offers a novel approach to material by design. A magnetic vortex is a domain pattern typically observed in nano- and mi cromagnets. The particular type of vortex studied in this work is a Landau flux closure. As the name implies, a magnetic vortex is characterized by the magnetization curling around the center of the magnet. In the very middle the magnetization turns out of plane, forming the vortex core. The mag netic vortex has two binary variables; the handedness of the vortex (clock wise and anti-clockwise) and the polarization of the core (up or down). The magnetic vortex may thus be the basis of a two-state or four-state magnetic memory. To realize devices based on magnetic vortices benefits from an increased understanding of the switching process. In this regard, epitaxial material systems are interesting because they give high quality thin-films and enable control of magnetic properties through strain. We have used square microplatelets of the epitaxial ferromagnet La0.7Sr0.3MnO3 (LSMO) as a model system for studying the tempereature dependence of magnetic vortex core dynamics. Using scanning transmission x-ray microscopy (STXM) we determine the temperature and field dependence of the vortex core dynamics in the epi taxial LSMO micromagnet. The temperature is varied from 150 K and up to TC of LSMO. In this temperature range the magnetic parameters of LSMO strongly vary. The experimental data is supplemented by micromagnetic simulations taking into account the temperature dependence of the mate rial parameters. Our results show that the vortex core switches when the vortex core velocity exceeds a critical value. This critical velocity is primar ily determined by the exchange interaction, and is lowered by the presence of cubic magnetocrystalline anisotropy. A lower critical velocity facilitates vortex core switching by lowering the switching field. These results can thus be used when searching for materials that give optimal device performance at room temperature. In summary, we demonstrate long-range ordering and vortex core dynamics in patterned magnets. These phenomenas are interesting from a fundamen tal viewpoint and toward realizing new magnetic materials and spintronic devices.