Direct Observation of Current-Induced Motion of a. 3D Vortex Domain Wall in Cylindrical Nanowires

Transcription

1 Supporting Information Direct Observation of Current-Induced Motion of a 3D Vortex Domain Wall in Cylindrical Nanowires Yurii P. Ivanov,,, *, Andrey Chuvilin ǁ,, Sergey Lopatin, Hanan Mohammed, Jurgen Kosel King Abdullah University of Science and Technology, Thuwal, , Saudi Arabia Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Jahnstrasse 12, A-8700, Leoben, Austria School of Natural Sciences, Far Eastern Federal University, , Vladivostok, Russia ǁ CIC nanogune Consolider, Av. de Tolosa 76, 20018, San Sebastian, Spain IKERBASQUE, Basque Foundation for Science, Maria Diaz de Haro 3, 48013, Bilbao, Spain * ivanov.yup@gmail.com S-1

2 Methods Growth of multisegmented nanowires. The experimental details of the nanowire deposition have been reported elsewhere. 1 In the present study, the nanowire diameter was 80 nm, the lengths of each Co and Ni segment was 700 nm, and total number of segments was 30. In-situ electrical TEM measurement. We used commercially available TEM chips from the Protochip Company. The sample was supported on a 0.5 mm x 0.5 mm freestanding membrane, centered on a silicon frame. The Electrical E-chips (AEW series) have an amorphous silicon nitride membrane that is transparent to electron beams. The chips are ~50 nm thick with lithographically patterned Pt electrodes with an interdigital structure extending directly onto the membrane. A drop of ethanol containing the dispersed nanowires (~ nanowires per 1 microliter) was placed on the silicon nitride membrane while applying an external magnetic field to align the nanowires parallel to the Pt electrodes (see details in Supporting Information). A dual beam (ion beam and electron beam) system (FEI Helios 450) was employed to select the appropriate nanowire and to create electrical contacts between the nanowire and the two Pt electrodes by ion beam-assisted deposition of Pt. A Dual-Channel System Source Meter (SMU, 2600B) Instrument with digital I/O Interfaces (200V/10A) connected to the TEM holder was used to apply constant DC current during in-situ experiments. Magnetic imaging. Electron microscopy studies were carried out with a transmission electron microscope (TEM), Titan G (FEI, Netherlands), equipped with a high-brightness field-emission gun (X-FEG) and a Cs image corrector. The Lorentz mode corrected by special Cs achieved a spatial resolution of about 1 nm at 300 kv. Images were captured with a Gatan US1000 CCD camera. The external magnetic field was applied to the nanowire in-situ by tilting the sample to 30 degrees and adjusting the strength of the TEM OL. Simulations. The magnetization process and DW dynamics were simulated using the OOMMF package. 2 The parameters chosen for the simulations have been reported elsewhere. 3 The cell size was 5 nm and the nanowire diameter was 80 nm. A nanowire with 2 segments of S-2

3 700 nm each was modeled. The direction of the magnetocrystalline anisotropy was considered to be in agreement with the TEM data (at 85 with respect to the NW axis for the Co segment and {220} texture for Ni segments). In the simulations of the current-driven DW motion, we used the value 0.42 for the spin polarization, as previously reported for planar structures. 4 Simulations were performed for 0 K and for finite temperatures (extracted from the experiments) using the corresponding extensions of the OOMMF code. Experimental Setup. We used commercially available TEM chips from Protochip company. The sample is supported on a freestanding membrane with an area of 0.5 mm x 0.5 mm centered on the silicon frame. The Electrical E-chips (AEW series) have an amorphous silicon nitride membrane that is transparent to the electron beam. It is ~50 nm thick with lithographically patterned metal electrodes extending directly onto the membrane. The chips have Pt electrodes with a two-finger structure. A 2 µl drop of ethanol containing dispersed nanowires was placed on the silicon nitride membrane while applying an external magnetic field to align the nanowires parallel to the Pt electrodes as shown on the Fig. S1. A dual beam (ion beam and electron beam) system (FEI Helios 450) was employed to localize a nanowire and to create electrical contacts between the nanowire and two Pt electrodes by ion beam assisted deposition of Pt (see Fig. S1) A Dual-Channel System Source Meter (SMU, 2600B) Instrument with digital I/O Interfaces (200V/10A) connected to the TEM holder was used to apply constant DC current during in-situ experiments. During the in-situ experiments a DC current of ua is applied. With a nanowire diameter of 80 nm, the calculated current densities of A/m 2 are obtained. S-3

4 Figure S1. The setup for in-situ TEM current-driven DW motion experiments in a cylindrical multisegmented Co/Ni nanowire (shown between Pt electrodes). The SiN membrane is yellow. The Pt electrodes and contacts are orange. Original over-focused Lorentz images. Fig. S2 shows the full Lorentz images of the nanowire discussed in Fig. 1 of the main manuscript. The nanowire part shown in Fig. 1 of the manuscript is marked by dashed lines. Figure S2. a STEM image of the device consisting of one multisegmented Co/Ni nanowire that is connected to Pt electrodes. b EELS elemental mapping of the multisegmented Co/Ni nanowire. c-g Overfocused Lorentz images of the nanowire: c and g are L state and R state, e initial state with the domain wall pinned at the interface - DW state, d and f the difference images of DW state L state, and DW state - R state, respectively. S-4

5 Movie S1. Field-driven 3D vortex domain wall motion. Movie S1 demonstrates the controlled motion of a 3D vortex domain wall in a cylindrical multisegmented Co/Ni nanowire under the effect of the external magnetic field. As shown in the Movie, the DW reproducibly moves from one pinning site (Co/Ni interface) to another, once the direction of the external magnetic field changes is reversed. Ex-situ magnetoresistance measurements. Magnetization reversal of a Co/Ni nanowire was also studied ex-situ by using magnetoresistance measurements (MR). Nanowires were dispersed onto a SiO 2 /Si substrate and electrodes [Cr/Au (20/180 nm)] were patterned onto a single nanowire by electron lithography protocol. Four-point probe measurements were performed by passing a current through the outer electrodes and the voltage between the inner electrodes was measured (Fig. S3a). An external magnetic field was applied parallel to the nanowire axis. The nanowire is firstly saturated in one direction, at this stage, the magnetic moments are aligned parallel to the direction of current and this is depicted by a state of high resistance in the MR curve. The magnetic field is then gradually reversed, leading to a rotation of the magnetic moments away from the direction of current, as a result, a decrease of resistance in the MR curve is observed. The next expected stage of the magnetization reversal would be a distinct jump in the resistance indicating an abrupt reversal of the magnetic moments. This is not the case in the multisegmented Co/Ni nanowires. Here, the magnetization reversal occurs through the pinning and eventual propagation of a domain wall, which is observed as an interruption in the resistance jump of the curve as observed in Fig. S3b. S-5

6 Figure S3. a Scanning electron microscopy image of a multisegmented Co / Ni nanowire with measurement schematics, b Magnetoresistance measurement of an individual multisegmented Co / Ni nanowire. On-line data of the current-driven 3D vortex DW motion experiment. Fig. S4a shows the graph of the on-line recorded time dependence of current and resistance during an in-situ experiment. The first step corresponds to the positive x direction current which de-pins the domain wall. Small variations of the resistance marked by the green arrow correspond to the restoration of the initial DW state. The last two peaks correspond to the negative x direction of the current with increasing amplitude. As shown in the resistance curve, the current needed to de-pin the DW causes a rise of the resistance due to NW heating. We estimated the temperature rise from the R(T)=R 0 [1+α(T-T 0 )] dependence for bulk material with α= K As shown in the Fig. S4b, at the applied current densities the temperature rise is within 50 K. Note, the second negative current value in Fig. S4a causes a much higher heating effect, but it still did not move the DW. This fact unambiguously rules out Joule heating as a reason for DW de-pinning. Only the current of a certain direction moves the DW. S-6

7 Figure S4. a Time dependence of the resistance and the electric current density during current induced domain wall motion. b Temperature vs resistance of the nanowire for different current density values used for the experiments. Movie S2. Micromagnetic simulation of the 3D vortex domain wall depinning from Co/Ni interface under electric current. Movie S2 shows typical micromagnetic simulation results of the 3D vortex DW current induced dynamics. The case corresponds to the triangle in Fig. 3b of the main manuscript. References (1) Ivanov, Yu. P.; Chuvilin, A.; Lopatin, S.; Kosel, J. Modulated Magnetic Nanowires for Controlling domain Wall Motion: Towards 3D Magnetic Memories. ACS Nano 2016, 10, (2) Donahue, M.; Porter, D. Object Oriented Micromagnetic Framework (OOMMF) The National Institute of Standards and Technology (NIST) Home Page. (accessed Aug 28, 2006). S-7

8 (3) Ivanov, Yu. P.; Vázquez, M; Chubykalo-Fesenko, O. Magnetic Reversal Modes in Cylindrical Nanowires. J. Phys. D: Appl. Phys. 2013, 46, (4) Boulle, O.; Malinowski, G.; Kläui, M. Current-Induced Domain Wall Motion in Nanoscale Ferromagnetic Elements. Mater. Sci. Eng., R 2011, 72, (5) Marick, L. Variation of Resistance and Structure of Cobalt with Temperature and a Discussion of Its Photoelectric Emission. Phys. Rev. 1936, 49, S-8