Growth and characterisation of frustrated kagome ferromagnet thin films

The layered kagome ferromagnet Fe3Sn2 has massive Dirac fermions and frustrated magnetism that shows skyrmions at room temperature. Magnetic skyrmions are small (nm to 100s of nm in size), stable (protected by their non-trivial topology) and easily moved (they respond to spin torques at low current densities). As such they are appealing candidates for the representation of data in new forms of spintronic data storage and logic devices that will be non-volatile, consume little energy, and permit novel compute-in-memory architectures suitable for both Boolean and neuromorphic computation.

So far Fe3Sn2 has only been studied in bulk form, but thin films are required for spintronic devices. It has recently been shown that the related compound FeSn can be grown as an epitaxial thin film on (111) SrTiO3. Since both compounds have the same kagome places of Fe with a very close lattice constant, we expect that we could grow Fe3Sn2 on SrTiO3. In this project we will develop the means to grow thin films of Fe3Sn2 in the state-of-the-art Royce Institute multi-chamber deposition system at Leeds, characterise the films using world-leading electron microscopy in the Bragg Centre, and study their electron transport and magnetic properties, especially those related to skyrmions. A particular feature of this project is that we can grow the magnetic metal layer epitaxially on the surface of an oxide heterostructure, each grown in a specialised chamber of the growth system and transferred under UHV. This should ensure the highest quality growth of the critical interfaces in the stack, which we shall confirm using high resolution TEM of cross-section specimens. The epitaxial oxide growth means that it will also be possible to include other oxide materials in the heterostructure such as water-soluble Sr3Al2O6, which can be used to form a sacrificial layer to allow the thin film to be floated off for study by means of transmission experiments. As well as providing plan-view samples for TEM characterisation within this project, such samples will also be useful for magnetic imaging by means of Lorentz or soft x-ray microscopy through our network of collaborators, e.g. at the University of Glasgow and Paul Scherrer Institute.

The combination of an exotic electronic structure with spin textures with non-trivial real-space topology is expected to lead to new insights into magnetotransport phenomena governed by the Berry phase, such as the topological Hall effect. Applying gate voltages will allow us to probe transport below, at, and above the Dirac point. Meanwhile, the fact that skyrmions are stabilised by frustration, rather than the more usual means of a chiral Dzyaloshinskii-Moriya interaction means that the skyrmion chirality is in principle switchable, providing a bistable state variable for the representation of digital data.