Research Area A: Metal-based NANO-SPINTRONIC
X-ray holography has been established in Hamburg in recent years in a successful cooperation between the University and DESY (Oepen, Grübel). In the framework of the proposed Cluster, the goal is to take images of the domain structure in magnetic nanowires via X-ray holography while measuring magnetoresistance and Anomalous Hall Effect in order to determine the domain wall resistance. In a second step, we will try to push domain walls via current pulses and image the time evolution of the domain structure. Perpendicularly magnetized systems, like Co/Pt, are in the focus of present research as their domain walls are very small. The ballistic transport through narrow walls causes a larger momentum transfer which enables switching at lower current densities.
Thermoelectric effects in magnetic or semi-metallic nanowires under the influence of magnetic fields have recently attracted considerable interest. For instance, the observation of the spin-dependent Seebeck and Peltier effects in nanowires has been reported. This emerging field is called ”Spin Caloritronics”and bridges the gap between spintronics, magnetism, and thermoelectricity. The Nielsch group will study the effect of the Giant Magneto Resistance (GMR) on the thermal conductivity in nanowires. In general the thermal conductivity in metallic systems is proportional to the electrical conductivity (Wiedemann-Franz law). The limits of the Wiedemann-Franz law for nanostructured GMR systems will be investigated. Co/Cu multilayered nanowires will be synthesized via pulsed electrodeposition in self-ordered Al2O3 membranes. Nanowires released from the Al2O3 matrix will be electrically contacted. Thermal conductivity measurements will be performed based on the hot wire approach and the 3-omega technique. In a second step the project will be extended to thermopower and thermo-galvanic measurements of GMR nanostructures and semi-metallic systems, e.g. bismuth, in magnetic fields.
Logic devices based on macro-spins are in the scope of a project that is focused on the fabrication of nanomagnets and the investigation of their interaction in an ordered matrix (Nielsch, Oepen). Imprint lithography and electrochemical oxidation processes will be applied to prepare perfectly ordered Al2O3 membranes consisting of hexagonally arranged pore channels. Individual pore channels will be functionalized via focused ion beam (FIB) and filled selectively with ferromagnetic material, enabling the fabrication of cylindrical nanomagnets with high-aspect ratio, so called macro-spins, which exhibit a bistable behavior caused by the shape anisotropy. The ultimate goal is the development of macro-spin arrays for applications as logic elements. The dipolar interaction will be used to mediate the information along lines of macro-spins. Logic devices will be achieved via branching and crossing of lines of macro-spins and their specific two-dimensional geometrical arrangements. The experimental results will be compared with theoretical studies (Vedmedenko).
Since many spintronic devices require insulating elements as building blocks we also plan to study the properties of magnetic nanostructures in direct contact with insulators. For this purpose we will utilize our recently developed technique of magnetic exchange force microscopy (MExFM), which is sensitive to the force between atomic magnetic moments and hence is not limited to conducting samples (Schwarz). MExFM has a short interaction range of Angstroms which enables the imaging of the magnetic structure with atomic resolution, including antiferromagnetic and complex spin structures. For instance, exchange-bias systems including metallic-magnetic/insulating-magnetic interfaces will be addressed. By recording the energy dissipation in MExFM experiments, we will additionally study spin excitations, i.e. magnons, in laterally confined systems.
Using spin-polarized scanning tunneling microscopy (SP-STM) it is feasible to resolve atomic scale magnetic structures with non-collinear and complex magnetic order (Wiesendanger group). In systems lacking inversion symmetry, like at interfaces and surfaces, spin-orbit coupling gives rise to the Dzyaloshinskii-Moriya interaction (DMI) which induces a canting between magnetic moments of adjacent atoms with specific sense of rotation. DMI can compete with exchange interaction and magnetic anisotropy in metallic nanostructures. The competition of these parameters generates a huge variety of magnetic order such as helical, cycloidal, and skyrmion spin structures. We want to explore this complex magnetic phase space using a combination of density functional theory (Heinze), Monte-Carlo simulations (Vedmedenko), and SP-STM experiments (von Bergmann, Wiesendanger). The first goal is to understand the underlying principles and the microscopic origin of competing magnetic interactions leading to uniaxial spirals or nanoscale two-dimensional magnetic order. In a second step, we will investigate the dynamics of such structures due to excitations by tunneling electrons. Moreover, we will use spin-polarized current pulses from an STM tip to study the stability of non-collinear spin states in nanoscale magnetic systems (Krause). We will also study systematically the thermal stability of such systems by variable-temperature SP-STM (Pietzsch, Krause, Wiesendanger).
Current-induced manipulation of magnetization will facilitate further down-scaling of magnetic random access memories (MRAMs), since magnetization reversal by means of an electric current is much easier to perform than using external magnetic fields. One aim of this initiative is to understand the microscopic processes involved in interactions between spin-polarized currents and nanoscale magneticstructures with complex spin states. The transport parameters will be gained from unified ab-initio calculations (Lichtenstein, Lechermann) and be taken as an input for Monte-Carlo and spin dynamics simulations (Pfannkuche, Vedmedenko, Potthoff, Bolte) as well as an analytic formalism (Gunesch, Vedmedenko, Hinze) bridging the nano- and micro-scales. These multi-scale theoretical approaches will play a decisive role for a fundamental understanding of complex nanodevices in research areas A and C.
When the spin-torque is exploited to control magnetic states in nanostructures, spin excitations are inherently involved. While spin waves in bulk materials are well understood, excitations in nanoscale magnetic structures are in the focus of current research activities worldwide. We plan a joint effort to investigate the spin dynamics in nanostructured materials by inelastic neutron scattering (Schreyer) and inelastic electron tunneling spectroscopy (Kubetzka). Neutron scattering on one hand can access averaged properties of magnetic systems and, with its wave vector resolution, the dispersion of spin waves for the entire Brillouin zone. With inelastic scanning tunneling spectroscopy (ISTS) on the other hand, individual nanoscale elements can be addressed and their dynamic properties can be correlated to their precisely known atomic arrangement. In addition, we will investigate spin waves in superlattices made of monodisperse magnetic nanoparticles separated by polymers (Förster, Weller). These samples can be prepared in a comparatively large volume, providing maximum intensity in neutron scattering experiments. We aim at a detailed understanding of the crossover from quasi-classical spin waves at large length scales to quantum-spin excitations at the atomic limit.
3D-Animation: Current-Induced Magnetization Switching with a Spin-Polarized Scanning Tunneling Microscope
Soft X-ray microscopy with a spatial resolution of 10 nm in combination with time resolution in the picosecond regime is a powerful tool for spatio-temporal investigations. At the Advanced Light Source in Berkeley, the groups of Meier, Bolte, and Merkt will investigate, in collaboration with international partners, soft magnetic, micron- and nanosized thin-film permalloy (Ni80Fe20) structures with magnetic vortices and domain walls. A vortex could serve as bit in future magnetic memory devices. Interesting topics are the influence of the confining potential on the dynamics of the vortex and a possible blue or red shift for larger gyration amplitudes. The planned experiments will be compared to theoretical descriptions by Guslienko, Shibata, and our own analytical models (Pfannkuche) as well as micromagnetic and transport simulations (Bolte, Meier, Möller). Through electric transport measurements and time-resolved X-ray microscopy in collaboration with our partner at the IBM Almaden Research Center, San José, we will advance the understanding of the spin-transfer torque in magnetic nanostructures. We will study the magnetization dynamics of domain walls in nanowires with in-plane and perpendicular magnetic anisotropy.
In nanostructured ferromagnets with dimensions of less than 100 nm, the spin torque becomes comparable to the magnetic damping torque if the current densities exceed 1010 A/m². This gives rise to novel current-induced dynamic excitations and magnetic switching phenomena. Striking examples are current-induced reversal processes in spin valves, high-frequency magnetic precessional motion in magnetic tunnel junctions induced by a direct current, and current-driven domain wall movement. Such investigations are particularly important in view of potential applications like the Spin-Transfer Nano-Oscillator or the Racetrack Memory. In the framework of our Cluster, we want to study spin-torque phenomena with very high spatial and temporal resolution via nuclear resonant scattering (NRS) of synchrotron radiation at PETRA III (Röhlsberger). The outstanding brilliance of the X-ray beams at PETRA III allows for a very efficient illumination of nanoscale structures. The heterostructures will be laterally patterned via focused ion beams (Oepen) and the samples will be electrically contacted to apply perpendicular currents. Selective enrichment of the magnetic heterostructure with thin layers of 57Fe (replacement of natural Fe with 57Fe) will enable the analysis of the magnetic structure in specific regions of the sample with sub-nanometer depth resolution. Since NRS is intrinsically a time-resolved technique, information about the time dependence of spin-torque processes can be obtained simultaneously. The data will provide an important ingredient for simulations and ultimately a detailed understanding of the dynamics.
The genuine frequency domain of magnetization and spin dynamics is the microwave range. In the low-frequency regime, lateral all-metal spin-valve devices are workhorses in spintronics which provide reliable injection, manipulation, and detection of spin-polarized currents. Research on metal-based spintronic devices benefits enormously from microwave-measurement techniques. Our spin-valve devices (Meier) include tunnel barriers at the interface between the ferromagnetic electrodes (e.g. Ni80Fe20) and the interconnecting metal strip (Al or Cu), which enlarge the spin polarization of the injected current. Pulsed and continuous-wave investigations will provide insight into the intrinsic dynamics of charge and spin transport and will give access to important parameters like spin-relaxation times and lengths. Also the inverse effect, called spin pumping, where a pure spin current is generated at the interface between a ferromagnet with a precessing magnetization and a normal metal will be studied (Meier). The development of a highly efficient spin source by spin injection of a precessing magnetization can provide a solution to the vision of spintronic devices.
Micromagnetic modelling and simulation have proven to be invaluable tools for understanding static and dynamic properties of ferromagnetic systems. In cooperation between the Departments of Physics, Mathematics, and Informatics, we develop micromagnetic simulation codes that allow for an in-depth study of electric currents interacting with ferromagnets. We will apply concepts of constrained optimization of partial differential equations (Hinze) in order to design ferromagnetic nanostructures with prescribed magnetization behavior. Furthermore, we will improve the simulation tools by integrating temperature effects and diffusive transport, including magneto-resistance and Hall effects, into the numerical micromagnetic code. Numerical treatment of the resulting optimization problems necessitates close multi-disciplinary collaboration between the fields of physical modeling (Meier, Merkt), high-performance simulation in computer science (Bolte, Möller), and applied mathematics (Hinze).
