Experimental evidence of Bloch-point domain walls in cylindrical nanowires

In 2004, IBM proposed a concept for a three-dimensional (3D) data storage device, in which a series of magnetic domain walls (DWs) act as bits to be moved along vertical nanowires ordered in a dense matrix. Holding great promise in terms of areal density and speed, this futuristic concept stimulated an intense research activity on DW motion along one-dimensional systems. So far, physics and demonstrators have been based on flat strips made by lithography, owing to ease of fabrication and observation. During the past years, we started investigating cylindrical nanowires, synthesised through bottom-up chemical techniques. We are addressing a few of the fundamental bottlenecks that need to be solved, before such a 3D storage scheme may be implemented. A corner stone in our studies was the experimental evidence for Bloch-point domain walls, for which we combined synthesis strategies, high-resolution magnetic microscopies, and micromagnetic simulations.

Various simulations predicted since 2000 that besides the conventional transverse wall, another type of DW not present in strips exists when the wire diameter exceeds a few times the dipolar exchange length. It consists of an orthoradial curling of magnetisation at the perimeter of the wire to best close the magnetic flux, along with a Bloch point on the axis at the center of the DW. This motivated the name Bloch-point domain wall. A Bloch point is a unique object in micromagnetism, close to atomic dimensions, where the magnitude of the magnetisation vector vanishes. It is imposed due to magnetisation spanning all 4π directions around the DW. It is thus topologically protected, essentially preventing Walker breakdown and promising DW speeds above 1000 m/s, compared to values of the order of 100 m/s for DWs in strips.

Resolving the inner structure of such DWs requires a bulk-sensitive microscopy, due to its 3D nature. We used high spatial resolution Photo-Emission Electron Microscopy combined with X-ray Magnetic Circular Dichroism (XMCD-PEEM) with both surface and volume sensitivity, the latter being obtained in a projection geometry by imaging the shadow of the wire on the supporting surface. This allowed us to identify the two types of DWs predicted by simulations, the transverse wall for small diameter and the Bloch-point wall for larger diameter [1]. Due to the grazing angle of incidence of the photons on the supporting surface (16°), the shadow is inflated by a factor 3.5, resulting in a spatial resolution of about 10 nm. This is the first time that the direct environment of a Bloch point is imaged at a scale comparable to the dipolar exchange length, confirming the expected arrangement of magnetisation. A quantitative agreement is indeed found with simulations. This opens the door to experimental investigations of the dynamics of the Bloch-point wall, for which peculiar features have been predicted by simulation (absence of Walker field, and DW speeds of around 1 km/s).

Figure : Identification of the Bloch-point and transverse domain walls based on XMCD-PEEM in transmission-projection geometry. (a-b) Absorption, XMCD contrast parallel and across two wires. (c-d) Absorption, XMCD and simulations of two DWs at the Fe L3 edge. The photons arrive from the upper part of the images. (wire diameters : 95nm and 70nm, respectively) (e-f) Views of the micromagnetic states used in c-d.

Links and related pages

- [1] S. Da Col, S. Jamet, N. Rougemaille, A. Locatelli, T. O. Mentes, B. Santos Burgos, R. Afid, M. Darques, L. Cagnon, J. C. Toussaint, and O. Fruchart, Observation of Bloch-point domain walls in cylindrical magnetic nanowires, Phys. Rev. B 89, 180405(R) (2014).
- Reduction of magnetostatic interactions in self-organized arrays of nickel nanowires using atomic layer deposition

© Institut Néel 2012 l Webdesign chrisgaillard.com l Propulsé par spip l Dernière mise à jour : Wednesday 10 June 2020 l