M.D. Cooke

Domain wall propagation in magnetic nanowires by spin-polarized current injection

We demonstrate movement of a head-to-head domain wall through a magnetic nanowire of permalloy (Ni81Fe19) simply by passing an electrical current through the domain wall and without any external magnetic field applied. The effect depends on the sense and magnitude of the electrical current and allows direct propagation of domain walls through complex nanowire shapes, contrary to the case of magnetic-field–induced propagation. The efficiency of this mechanism has been evaluated and the effective force acting on the wall has been found equal to 0.44 × 10−9 N A−1.

Magneto-optical Kerr effect analysis of magnetic nanostructures

We present a treatment of polarized light analysis and noise sources relevant to magneto-optical Kerr effect (MOKE) measurements of magnetic nanostructures and give performance details of our state-of-the-art MOKE magnetometer in these terms. By considering the various signal contributions, we are able to determine experimental conditions for optimal MOKE signal dynamic range and signal-to-noise ratio. A description of our MOKE instrument includes a novel facility to locate magnetic nanostructures by producing a susceptibility map of the sample surface. The performance of the magnetometer is demonstrated by hysteresis loops of single 100–200 nm wide magnetic nanowires, including a loop obtained with no time averaging during a single magnetic field cycle.

Domain wall injection and propagation in planar Permalloy nanowires

We have used high-sensitivity magneto-optics to study the magnetic switching of a single 20 μm long, 100 nm wide, 5 nm thick planar Permalloy wire made by focused ion beam milling. It is found that the switching field of the wire can be dramatically reduced by the addition of a large end pad to the wire, which serves as a domain wall injector. We show that once injected, the domain wall is able to move very freely along the wire. No measurable pinning of the domain wall was observed at a shallow kink in the middle of the length of the wire. We thus show that such wires can be considered as domain wall conduits, with many applications in magnetoelectronic devices.

Magnetic nanoelements for magnetoelectronics made by focused-ion-beam milling

Focused-ion-beam (FIB) milling has been used to structure magnetic nanoelements from 5 nm thick films of permalloy. We have used focused 30-keV Ga+ ions to define small arrays (6 μm×6 μm) of wires, circles, and elongated hexagons in the size range 100–500 nm. High-sensitivity magneto-optical measurements combined with atomic force microscopy show that very high quality magnetic nanostructures can be fabricated by FIB milling even in thin films of soft magnetic materials. This finding could be significant for the future commercialization of certain aspects of magnetic nanotechnology and magnetoelectronics.

Artificial domain wall nanotraps in Ni81Fe19 wires

We report on the controlled pinning and depinning of head-to-head domain walls with individual artificial nanotraps in rounded L-shaped Ni81Fe19 wires. Domain walls were nucleated and injected into one arm of an L-shaped planar wire structure with a wire width of 200 nm and a thickness of 5 nm. The domain walls were propagated through a rounded corner into an orthogonal output wire by a 27 Hz anticlockwise rotating field. A highly sensitive magneto-optical Kerr magnetometer system was used to detect magnetization reversals around single wedge shaped nanotraps in the output wire of different samples. Domain wall propagation occurred at a mean measured x-field value of 6.8 Oe in the output wire arm when not interacting with a trap. Domain wall nanotraps with dimensions as small as depth Dt=35 nm and width Wt=55 nm were found to effectively pin domain walls. In general, the depinning field of a domain wall from a trap increased with trap size. Hysteresis loops and plots of domain walls depinning fields as a ...

Shifted hysteresis loops from magnetic nanowires

We demonstrate that positively and negatively field-shifted magnetic hysteresis loops can be obtained from a single continuous L-shaped magnetic nanostructure. This is achieved by controlling the coercivity of one arm of the L-shape structure with the magnetization direction of the orthogonal arm. Furthermore, a memory effect is demonstrated by reversing the magnetization direction of one arm while leaving the other unchanged. Good discrimination between the different switching field magnitudes and the ease of fabrication make these continuous magnetic structures more suitable than chains of discrete magnetic dots for performing logical operations.

Controlled switching of ferromagnetic wire junctions by domain wall injection

The switching of submicrometer ferromagnetic wire junctions is investigated by controlled injection of domain walls (DWs). A three-terminal continuous Ni/sub 80/Fe/sub 20/ structure is described, consisting of two input wires and one output wire. Separate structures were fabricated by focused ion beam (FIB) milling, to inject either zero, one, or two DWs to the junction inputs. Introduction of DWs to the junction was performed using one or two DW injection pads with low switching fields. Hysteresis loops measured by magnetooptical Kerr effect (MOKE) magnetometry on the device output wires showed the coercivity of the output is strongly dependent on the number of DWs incident at the junction. By injecting either one or two DWs into the junction, it is possible to switch the output wire at two distinct field values, each markedly lower than the nucleation field of the junction. Results presented are relevant to the future development of spintronics DW logic systems, and for fundamental studies into DW resistance and interaction between DWs and wire morphology.

Magnetic domain wall dynamics in a permalloy nanowire

Experimental results of domain wall propagation velocity as a function of applied field in a 200-nm-wide permalloy nanostructure are presented. The structure was fabricated by focused-ion beam milling and designed to separate the domain wall nucleation from propagation. The measured velocities are very high, up to 1500 m/s at 49 Oe and the field dependence of the velocity is interpreted in terms of thermally activated wall motion through a series of pinning sites and a wall geometry that changes as a function of field.

Characterization of submicrometer ferromagnetic NOT gates

We present operation phase diagrams of all-metallic submicrometer ferromagnetic NOT–gate devices. The phase diagrams summarize four different types of behavior, in which devices can operate correctly with either one or three domain walls propagating through them, nucleate domain walls, or pin a domain wall, leading to its annihilation with a succeeding domain wall. We use these phase diagrams to investigate the influence of junction dimensions on domain wall nucleation and pinning, and determine optimized junction parameters for NOT–gate operation. Furthermore, we demonstrate how changing the NOT–gate orientation to an applied field affects the operating phase diagram and may assist the integration of NOT-gates with other types of junction in the near future to realize a full magnetic logic scheme. By fabricating the NOT–gate junctions within a magnetic feedback loop, the direction of domain wall propagation is shown to be reversible and the input and output wires therefore interchangeable.

Nanosecond pulsed field magnetization reversal in thin-film NiFe studied by Kerr effect magnetometry

Magnetization reversal has been studied in a 5 nm thick Ni80Fe20 film under the action of a pulsed magnetic field, for pulse durations between 2-102 ns. Fast pulses were applied to the sample from a thin-film microstrip transmission line and magnetization changes were measured using a continuous wave laser magneto-optic Kerr effect magnetometer. The switching field was found to increase as the pulse duration decreased, in qualitative agreement with simple models. More unusually, the switching field was also found to increase as a function of a quasi-static bias field in excess of the value of that bias. This effect is stronger for the shorter pulses. Evidence is presented which suggests that this is due to rapid relaxation of the magnetization after the termination of the pulsed field.