Structural, compositional and magnetic study of the transformation of a hematite thin film into a magnetite/hematite epitaxial bilayer

Oxides, as a wide class of materials with a huge range of functionalities, give rise to novel phenomena when different types are combined in heterostructures, whose production is an everlasting topic in the development of new materials. Thus the exploration of new fabrication methods is mandatory. Low energy ion bombardment has been used on single-crystalline oxide thin films to produce epitaxial layers of the corresponding suboxides [1,2]. We have carried out an investigation of the magnetite/hematite epitaxial bilayer (Fe₃O₄(111)/α-Fe₂O₃(0001)) on different substrates, which is an ideal oxide system to study magnetic exchange effects at the interface. The characterization of both systems has included structural, chemical and magnetic properties. We have used synchrotron-based experimental techniques such as Surface X-ray Diffraction (SXRD), Hard X-ray Photoemission Spectroscopy (HAXPES), Low Energy Electron Microscopy (LEEM), Photoemission Electron Microscopy (PEEM) or X-ray Circular Magnetic Dichroism (XCMD). The aim of this work has been to obtain microscopic information on the as-grown material and on the evolution of the materials during the ion bombardment, as well as during high temperature annealing of the bilayers. Ion bombardment gradually reduces the initial hematite into a defective sub-oxide where magnetite nuclei start to grow with increasing ion doses and coalesce into a defective epitaxial magnetite upper layer. Two rotational domains emerge, with small grains which are typically a few tens of nanometres large. They do not grow significantly with temperature annealings. This robustness may be explained by the incoherent growth of the grains which may favour the formation of twin or antiphase boundaries. The distribution of magnetic domains, as visualized with XCMD, shows no apparent correlation with the structural domains. The epitaxial magnetite/hematite bilayer shows a strong exchange bias. However, a very different interfacial exchange interaction appears when Ni film is directly grown on the hematite thin film, which is explained in terms of the interaction between an antiferromagnetic (hematite) and a ferromagnetic (Ni).



[1] B.M. Pabón et al., Nat. Comm. 6, 6147, 2015.

[2] S. Ruiz-Gomez et al., Appl. Phys. Lett. 110, 093103, 2017.