The magnetic proximity effect in the top and bottom Pt layers induced by Co in Ta/Pt/Co/Pt multilayers has been studied by interface-sensitive, element-specific x-ray resonant magnetic reflectivity. The asymmetry ratio for circularly polarized x-rays of left and right helicity has been measured at the Pt L3 absorption edge (11 567 eV) with an in-plane magnetic field (±158 mT) to verify its magnetic origin. The proximity-induced magnetic moment in the bottom Pt layer decreases with the thickness of the Ta buffer layer. Grazing incidence x-ray diffraction has been carried out to show that the Ta buffer layer induces the growth of Pt(011) rather than Pt(111), which in turn reduces the induced moment. A detailed density functional theory study shows that an adjacent Co layer induces more magnetic moments in Pt(111) than in Pt(011). The manipulation of the magnetism in Pt by the insertion of a Ta buffer layer provides a way to control the magnetic proximity effect, which is of huge importance in spin-transport experiments across similar kinds of interfaces.
(a) XRMR asymmetry ratio ΔI(q) for both magnetic-field directions at the resonant energy of 11 567 eV. (b) Experimental energy-dependent XAS spectrum (green curve) and the XMCD signal (red curve). All data correspond to the Ta(2 nm)/Pt(4 nm)/Co(2 nm)/Pt(2 nm) multilayer.
(a)–(e) The resonant (11 567 eV) averaged XRR signals with fits. (f)–(j) The corresponding measured and fitted XRMR asymmetry ratios ΔI(q) for the multilayers. (k)–(o) Magneto-optic depth profiles, which were used to fit the XRMR asymmetry ratios. Red histograms depict the in-depth distribution of the spin-polarized Pt atoms. The induced magnetic moment per Pt atom has been quantified from ab initio calculation. The dashed lines denote the corresponding interface position between the layers, whereas the dashed rectangles between the layers denote the corresponding interlayer roughness.
In this work, the investigation of in-plane uniaxial magnetic anisotropy induced by the morphology due to ion beam erosion of Si(100) has been done. Ion beam erosion at an oblique angle of incidence generates a well-ordered nanoripple structure on the Si surface and ripple propagates in a direction normal to ion beam erosion. Permalloy thin films grown on such periodic nanopatterns show a strong uniaxial magnetic anisotropy with an easy axis of magnetization in a direction normal to the ripple wave vector. The strength of uniaxial magnetic anisotropy is found to be high for the low value of ripple wavelength; it is decreasing with the increasing value of ripple wavelength. Similarly, the strength of uniaxial magnetic anisotropy decreases with increasing Permalloy film thickness. Grazing incidence small-angle x-ray scattering data reveals an anisotropic growth of Permalloy thin films with the preferential orientation of grains in the direction normal to the ripple wave vector. Permalloy thin film growth is highly conformal with the film surface replicating the substrate ripple morphology up to a film thickness of 50 nm has been observed. Correlation between observed uniaxial magnetic anisotropy to surface modification has been addressed.
(a) shows the schematic of the GISAXS measurements geometry. (b)–(d) represent the two-dimensional GISAXS images of Py thin film (t = 15 nm) deposited on nanorippled Si substrates having different wavelength values measured at an incident angle of 0.4°. Corresponding line-cuts made along the horizontal direction (see the white box drawn in (b)), along qy for a fixed value of qz value (around Si Yoneda peaks), are shown in (e)–(g).
(a) shows horizontal cuts along qy around Si Yoneda peak region of the 2D GISAXS data for different Py thickness values (shifted along the intensity axis) and (b) shows the vertical cut along qz direction for a chosen value of qy = 0.26 nm−1. One of the 2D GISAXS images has been selected to represents the direction of line integrals made.
Kerr microscopy measurements of Ta/Pt/CoFeB/Pt thin films, with ferromagnetic (FM) CoFeB layer grown by oblique-angle sputter deposition, indicated a quasi-perpendicular magnetic anisotropy. In such a system, the anisotropy is slightly tilted from the normal to the substrate plane. The effect of tilt was observed as a shift in out-of-plane (OP) room-temperature hysteresis, in presence of an in-plane (IP) bias field. Tilt angles were obtained by measuring the shifts in OP hysteresis for several magnitudes and relative azimuthal orientations of bias field with respect to IP projection of tilted anisotropy direction. We studied the variation of tilt angle as a function of FM layer thickness which was increased near to Spin Reorientation Transition (SRT) value for FM layer in the corresponding system. We also studied this variation for different bottom Pt layer thicknesses. The results of our experiment show the variation to follow two different profiles − one monotonic and the other non-monotonic.
(a) A schematic representation of the oblique-angle sputter deposition system. Rotation of the substrate during deposition of CoFeB gives rise to a uniform thickness. On the contrary, a thickness gradient is obtained in the layer without substrate rotation during deposition. (b) The representative XRR fitting measurements are shown for Ta(3 nm) Pt(3 nm) CoFeB(x nm _WR) Pt(1 nm) for the verification of actual CoFeB thicknesses, where x = 0.42 nm and 0.70 nm.
We study the evolution of the magnetic field-induced switching between the stable domain configurations of permalloy nanoring structures when magnetic nano-wires are attached to them. Magnetoresistance measurements were performed on such devices for two configurations of the attached nano-wires: (i) when they are at diametrically opposite ends of the nanoring, and (ii) when the nanowires are at an obtuse angle with respect to each other. During the measurements, the direction of application of the in-plane magnetic field is varied to understand the switching properties of the devices. Micromagnetic simulations were carried out in order to understand the domain configuration and reversal mechanism. We show that due to the nature of domain walls created by the presence of the nano-wires in the obtuse configuration, a vortex state can be stabilized in the nano-ring. We extended our studies to various nanoring devices with different widths while keeping a constant thickness.
SEM image of the nanoring devices: (a) Type I and (b) Type II. The corresponding magnified images are shown on the right side. The electrical contacts are labelled.
(a) The normalized MR (R35) of type I and II devices with 120 nm width at ϕ = 0°. The local hysteresis loops for the different regions of interest for (b) the type-I device and (c) the type-II device.
Magneto-optic Kerr effect (MOKE) is a widely used tool in surface physics to characterize magnetic thin films and single crystals. In most of its applications, the magneto-optic (MO) coupling is assumed to be linear on magnetization by neglecting the quadratic and higher dependences. However, recent observations have shown that the quadratic dependence cannot be always ignored, particularly in systems like ferromagnetic metal thin films, Heusler alloys, and even in half-metallic ferrimagnets. We have used a rotating field method to extract the QMOKE signal from Fe(100) film grown on the (100) surface of single-crystal iridium. Our results show that the QMOKE signal from this system is comparable to the linear MOKE signal. We report the parameters L, b, and c which relate the linear (K) and quadratic (G and ΔG) MO coupling coefficients. The real part of L, b, and c obtained from Fe(100)/Ir(100) are −9.75 ± 0.03 mdeg, 7.28 ± 0.09 mdeg, and −5.0 ± 0.1 mdeg, respectively.
Linear and quadratic MO Kerr rotation from Fe(100) film grown on Ir(100) substrate averaged over s- and p-polarized light. The angle of incidence is ~5°.
A magnetic field-free current-induced deterministic switching is demonstrated in a perpendicularly magnetized all-metallic Pt/Co/Pt thin film system with a small tilt in the anisotropy axis. We realized this in devices where the ultrathin Co layer was grown using an oblique angle sputter deposition technique that had resulted in a small tilt of magnetic anisotropy from the film normal. By performing out-of-plane magnetization hysteresis measurements under a bias magnetic field applied along with various in-plane directions the tilt angle was estimated to be around 3.3∘(±0.3∘). A deterministic current-induced magnetization switching could be achieved when the in-plane current was applied perpendicular to the anisotropy tilt axis, but the switching was stochastic when the current was applied in the direction of the tilt (in the tilt plane). By preparing Pt/Co/Pt stacks with unequal top and bottom Pt thickness, sufficient spin-orbit torque (SOT) could be applied to switch the magnetization of the Co layer at current densities as low as 1.5×107 A/cm2. The switching phase diagram (SPD) constructed by plotting the critical current density versus applied in-plane magnetic field (HIBx) confirms the spin Hall effect based SOT mechanism to be responsible for the magnetization switching. The asymmetry observed in the SPD (about HIBx=0) is in agreement with the macrospin simulations and it suggests that the tilt in the magnetic anisotropy from the film normal makes the switching deterministic even without an in-plane magnetic field bias.
(a) Schematic of sputter deposition geometry of Co where the sample rotation results in a uniform thickness and no rotation of the sample results in a thickness gradient. The oblique angle deposition was done at an angle ≈25∘ as indicated by the angle between the substrate normal and the target normal. (b) The gradient in the Co thickness over a sample length of 8 mm is indicated. The direction of the wire (device) axis indicated by the dash-dotted line exhibits an angle α with respect to the gradient direction. The direction of uniform Co thickness is indicated by the red dotted arrow and the direction of gradient in Co thickness by the blue arrow.
(a) The optical image of the device taken by Kerr microscope with the coordinate system viewed from the top. The in-plane bias field HIB is applied at an angle γ with respect to the wire axis. (b) The hysteresis of magnetization vs current for device A under different in-plane magnetic field bias applied along the current direction. (c) The domain structure in device A after the application of +J and −J pulses without in-plane bias field. (d) The Mz vs J for the device B90 without in-plane field bias (i.e., at HIBx=0 Oe). The loop path of the hysteresis can be traced by the arrows. (e) The domain structure after application of different current density values along with one of the arms of the loop (switching from +Mz to −Mz state at HIBx=0 Oe). (f) The domain structure in device B0 after the application of +J and −J pulses without in-plane bias field (HIBx=0 Oe). The region of bright and dark contrast corresponds to +Mz and −Mz states, respectively.
The implementation of magnetic domain wall (DW) based memory and logic devices critically depend on the control over DW assisted magnetization reversal processes. Here we investigate the magnetization reversal by DW injection, pinning and depinning at a geometrical constriction in permalloy nanowire (NW) driven by the external in-plane magnetic field, using local electrical probes. The observations of two distinct depinning field values are identified with the help of micromagnetic simulations, as being due to vortex DWs of different chiralities. The statistical analysis gave an estimate of chirality dependent pinning probability of DWs at this constriction. The stochastic nature of the DW based reversal driven by the magnetic field is revealed here. The asymmetry in the depinning field of the DWs to move to either side of constriction indicates the asymmetric nature of the barrier potential seen by the DWs. The results demonstrate the difficulties in achieving deterministic switching behaviour of DW assisted reversal and provide a platform to understand the main bottlenecks in the technological implementation of DWs.
Experiment schematic: AC source drives 10 μA current through the sample and voltages are measured across the junction (VJ) and notch (VN). The wire axis is parallel to X-axis. The directions of the applied magnetic field were along with angles θ=0°, ±10°, ±15°. The inset shows a scanning electron micrograph of the notch (430 nm wide).
(a). MR data of the junction and notch. The black plot is the resistance response as the field is swept from positive to negative saturation. The injection field (HINJ), pinning field (HPIN) and depinning field (HDPIN) are indicated. The schematic of the domain configuration of the junction and the notch are shown for the field scan from positive to negative saturation. (b) Magneto-Optical Kerr images. The position of the notch is indicated in the image. The dark and bright contrast corresponds to different magnetization configurations. The domain images are obtained after background subtraction. The domain configurations during the process of injection, pinning and depinning are shown.