Evolution of Novel Chiral Spin Textures in Fe/Gd Based Multilayers

Abstract

Research into magnetic skyrmions and their spintronics applications has revealed complex chiral spin textures. However, direct characterizations of the structure of skyrmion and related spin textures pose an experimental challenge. To better understand their structure, we use Lorentz transmission electron microscopy (LTEM) to track their evolution from well-studied domain walls (DWs) into skyrmion and biskyrmion phases. Magnetic skyrmions were first predicted as a result of the Dzyaloshinskii-Moriya interaction (DMI) found in materials with broken inversion symmetry1. However, topologically equivalent chiral bubbles, or dipole skyrmions, can be stabilized in achiral thin films by competing long-range dipole and short-range DW energies2.Recently, a novel skyrmion phase consisting of bound skyrmion pairs with opposite helicities was reported in achiral Fe/Gd multilayers3 (Fig. 1). However, the primary magnetic imaging techniques for thin films are LTEM, which yields projected in-plane magnetic field, and resonant X-ray techniques, which yield projected out-of-plane magnetization. This complicates identification of complex magnetic spin textures, as multiple spin textures can yield qualitatively similar projected magnetic fields, and there has been subsequent debate over the structure of the reported skyrmion phases4,5.By tracking the field-dependent evolution of magnetic spin textures from the well-studied magnetic stripes to the novel skyrmion phases, we can identify key components of these spin textures. First, by applying an out-of-plane magnetic field with a small in-plane component, stripe DWs form, alternatingly aligned and anti-aligned with the in-plane component, matching previous results2. As the field is further increased, the stripe DWs split, and the aligned DWs form the center of the novel texture, while the anti-aligned form closure loops (Fig. 2), supporting that the magnetization is a pair of contour-rotating vortices. Fully understanding these textures will require knowledge of the 3D structure, so surface sensitive techniques like SEMPA, SPLEEM, and MFM will be vital moving forward.  Fig. 1. The projected in-plane field of the novel texture.  Fig. 2. Stripe domains breaking into the novel phase.