Complex magnetic distribution of diameter‐modulated FeCoCu nanowires resolved by Electron Holography

Abstract

In the last years, diameter‐modulated (D‐M) ferromagnetic nanowires (NWs) have been intensively studied to evaluate their efficiency to control the motion of domain walls (DWs) along these one‐dimensional nanostructures by the application of magnetic field or the injection of electrical current, which is essential for spintronic applications in the field of information storage, sensors and logical operations [1]. Preliminary studies in D‐M NWs have been performed using theoretical and experimental procedures [2‐4] in individual and isolated NWs and they have provided a first approach of the spin configuration in these systems, obtaining a non‐trivial interpretation. In this work, we have exploited the potential of electron holography technique (high spatial resolution, high sensitivity and quantitative capability in volume) for achieving a full picture of the magnetic distribution in cylindrical D‐M FeCoCu NWs. These NWs were prepared by filling self‐assembled cylindrical D‐M nanochannels of anodic aluminum oxide templates. The D‐M geometry of the polycrystalline NWs consist of alternating segments of small (100 nm) and large (144 nm) diameters, with segment lengths ranging between 1000 to 300 nm. At remanence, the high‐shape anisotropy of the NWs induces a single‐domain state where the spins are mainly oriented along the NW axis, with the possibility to create a small closure domain in large‐diameter tips. The transition zones where the diameter is varied induce a complex demagnetizing field where the stray field follows a flux‐closure configuration around the large‐diameter segments and a magnetic coupling between them around small‐diameter segments (See Fig 1). The complex configuration of the demagnetizing field can be understood if we treat the D‐M transition zones as magnetic charges. The interpretation of the magnetic distribution by EH experiments was compared with micromagnetic simulation finding a very good agreement (see Fig 2). In addition, In‐situ Lorentz microscopy experiments of the magnetization reversal process allowed evaluating the DW nucleation and propagate process by the application of magnetic fields. We found that a field‐driven manipulation of DWs is not possible for the NWs in study. This “unsuccessful” result however helps us to take a step forward for the optimization of the geometry to reach the desired DW manipulation.