Magneto-optical Kerr Effect Loops Research Articles

Magnetization reversal in Fe(001) films grown by magnetic field assisted molecular beam epitaxy

We studied the influence of a magnetic field (MF) on epitaxial growth and magnetic properties of Fe(001) films deposited on MgO(001). Thanks to modular sample holders and a specialized manipulator in our multi-chamber ultrahigh vacuum system, the films could be deposited and annealed in an in-plane MF of 100 mT. In situ scanning tunnelling microscopy showed that MF had a strong influence on the film morphology, and, in particular, on the structure of surface steps. The magnetic properties were studied ex situ using magneto-optic Kerr effect (MOKE) magnetometry and microscopy. We showed that the moderate in-plane magnetic field applied during growth has the visible impact on the magnetic properties. The observed angular dependence of the MOKE loops and domain structures were discussed based on a magnetization reversal model. In particular we found that magnetization reversal occurs via 90° domains and the reversal differs for the no-field and in-field grown samples, in correlation with the film morphology.

MOKE study of magnetic compensation in polycrystalline gadolinium iron garnet thin film

Abstract The present work reports the preparation and magneto-optical Kerr effect (MOKE) microscopy study of gadolinium iron garnet Gd 3 Fe 5 O 12 (GdIG) polycrystalline thin film. The GdIG thin film is grown on Si(0 0 1) by Pulsed laser deposition technique (PLD). Grazing angle X-ray diffraction confirms the formation of GdIG after ex-situ annealing at 750 ° C under ambient conditions. MOKE measurement done in both longitudinal and polar geometries shows oblique magnetic anisotropy axis at room temperature rather than either completely in-plane or out of plane anisotropy axis. Signatures of magnetic compensation are seen from the increase of coercive field obtained from temperature dependent MOKE measurements. The observation of magnetic compensation is also manifested by the inversion of MOKE loops. The magnetic domain structure shows the formation of bubble or maze-like domains with a considerable dependence on temperature. The observed domain size variation as a function of temperature is fitted considering the model for thermally activated behavior of domain size variation in rare-earth garnets.

Advanced MOKE magnetometry in wide-field Kerr-microscopy

The measurement of MOKE (Magneto-Optical Kerr Effect) magnetization loops in a wide-field Kerr microscope offers the advantage that the relevant domain images along the loop can be readily recorded. As the microscope's objective lens is exposed to the magnetic field, the loops are usually strongly distorted by non-linear Faraday rotations of the polarized light that occur in the objective lens and that are superimposed to the MOKE signal. In this paper, an experimental method, based on a motorized analyzer, is introduced which allows to compensate the Faraday contributions, thus leading to pure MOKE loops. A wide field Kerr microscope, equipped with this technology, works well as a laser-based MOKE magnetometer, additionally offering domain images and thus providing the basis for loop interpretation.

Magnetization switching behavior of exchange-coupled bilayer nanodots characterized by magneto-optical Kerr effect

We investigated the magnetization switching behavior of exchange-coupled bilayer nanodots consisting of a soft magnetic Ni81Fe19 (permalloy; Py) layer and a perpendicularly magnetized L10-FePt layer by measuring the polar magneto-optical Kerr effect (MOKE). A MOKE system having a 2-μm-diameter laser spot was used, which detected the signal from around 20 nanodots of 200 nm in diameter. The measured polar MOKE loops exhibited a two-step behavior attributable to each magnetization reversal process of the Py and L10-FePt layers, allowing us to evaluate the switching field (Hsw) of L10-FePt from the MOKE loop. Under the application of both a static magnetic field and an rf magnetic field (hrf), the nanodots showed a clear Hsw reduction with characteristic hrf frequency dependence. This Hsw reduction is well interpreted as a behavior of spin wave-assisted magnetization switching. We also discuss the influence of the Py layer thickness on spin wave-assisted Hsw reduction.

Correlation of structure and magnetism of ultrathin Co films on Pd(001) prepared by thermal and pulsed laser deposition

We present a combined experimental and theoretical analysis of the atomic structure and the magnetic properties of 1- and 2-monolayer (ML)-thick Co films on Pd(001) grown by thermal deposition (TD) and pulsed laser deposition (PLD). While surface x-ray diffraction measurements show that the geometric structures of the as-deposited samples differ depending on the deposition method (alloy formation for PLD versus epitaxial growth for TD), magneto-optic Kerr-effect loops indicate an in-plane easy magnetization axis independent of the preparation method. Annealing at $600\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ induces a reorientation of the easy magnetization axis from in-plane to out of plane. This goes in parallel with substantial structural reorganization, leading to a $\mathrm{Co}∕\mathrm{Pd}$ multilayer structure with a top Pd layer. In agreement with experiments, fully relativistic Kohn-Korringa-Rostoker calculations including experimentally derived structural relaxations and disorder predict in-plane magnetization for the as-grown samples and out-of-plane magnetization for the annealed 2 ML sample characterized by an (incomplete) $\mathrm{Pd}∕\mathrm{Co}∕\mathrm{Pd}∕\mathrm{Co}∕\mathrm{Pd}(001)$ layer sequence. However, in-plane magnetization for the 1 ML $\mathrm{Pd}∕\mathrm{Co}∕\mathrm{Pd}(001)$ sandwich is theoretically predicted. Our study emphasizes the decisive importance of structural order, relaxation, and interface contribution to the magnetic anisotropy energy.

Partial magnetization reversal in a perpendicular exchange-biased [Pd∕Co]∕FeMn film through laser annealing

Partial magnetization reversal on the perpendicular exchange-bias characteristics in the [Pd∕Co]×5∕FeMn bilayered and [Pd∕Co]×2∕Cu∕Co∕[Pd∕Co]×2∕FeMn spin valve thin films was investigated using laser. The hysteresis loops were measured using the extraordinary Hall effect and magneto-optical Kerr effect (MOKE). By increasing the output intensity of the laser, the exchange-bias fields in the [Pd∕Co]×5∕FeMn bilayered thin film were gradually changed from negative to positive. The distribution of partial magnetization reversal in the vicinity of the laser spot was measured by using MOKE. Although the magnetic moment of the pinned layer was gradually reversed by using laser annealing in the MOKE loops of the [Pd∕Co]×2∕Cu∕Co∕[Pd∕Co]×2∕FeMn spin valves’ structure, that of the free layer was not changed. When the pattern size was decreased to 5×5μm2, the direction of exchange biasing was fully reversed through laser annealing without distorting the hysteresis loop. All the patterned samples below 100×100μm2 have the same hysteresis loop, and the reversed loops were not distorted and reduced when laser annealing was used.

Asymmetry of magneto-optical Kerr effect loops of Co nano-columns grown by oblique incident angle deposition

Magneto-optical Kerr effect (MOKE) in the longitudinal mode was used to measure the in-plane magnetic anisotropy of about 300nm thick Co nano-column films that were grown by e-beam evaporation at different oblique incident angles (0–85°) on native SiO2 substrates at room temperature. The symmetry of MOKE loops measured at azimuthal angles that are 180° apart is obviously broken for Co nano-column films grown at high deposition angles >70°. The plot of coercivity Hc of loops versus azimuthal angle shows that Hc values for azimuthal angles between −90° and 90° are much larger than the values for azimuthal angles between 90° and 270°. The asymmetry of coercivity is due to MOKE measurement that combines both longitudinal and polar Kerr effects. This combination is caused by the oblique magnetic anisotropy associated with the tilted Co nano-columns. A method is introduced to separate the longitudinal and polar Kerr effects. The longitudinal Kerr effect is obtained by adding hysteresis loops measured at azimuthal angles 180° apart while the polar Kerr effect is obtained by subtracting these two loops. By comparing these two orthogonal magnetization components represented by longitudinal and polar Kerr effects, we show that magnetization rotation exists in the magnetic reversal process even at azimuthal angle φ=0° that is the closest in-plane direction to the easy axis. After separating these two Kerr effects, the two-fold symmetry of coervicity associated with tilted nano-columns measured by longitudinal Kerr effect is restored.

Magnetic properties of ultrathin Fe3O4 on GaAs(100)

The evolution of magnetic phase and anisotropy of epitaxial half-metallic Fe3O4 ultrathin films on GaAs(100) has been studied using magneto-optical Kerr effect (MOKE) measurements. While no magnetic signal was detected below 1 nm, the film shows a superparamagnetic response at a thickness of around 1.6 nm. Ferromagnetic MOKE loops with strong in-plane anisotropies were observed above 2.0 nm. The films show a magnetic uniaxial anisotropy in a thickness range of about 2.0–6.0 nm with the easy axis along the [01̄1] direction.

Measurement of MOKE Hysteresis Loop Using Non-uniform Sampling

There are two parts in MOKE hysteresis loops, typical for materials with perpendicular magnetic anisotropy. One is due to a steeply falling edge or rising edge accompanied by the sign changes in magneto-optical Kerr effect (MOKE) signal. The other corresponds to a slow change of non-transition segment. A satisfying tradeoff between response to rapid change and noise rejection in non-transition segment of the loop is hardly obtained. The problem may be solved using a method based on non-uniform sampling. Its goals are to reduce distortion of the transitional waveform and noise, to obtain high resolution of transitional waveform and to decrease the number of sampling points. This paper describes the principle of non-uniform sampling and its implementation, and we designed a comparative experiment using a constant sampling frequency and multi-sampling frequency. The result shows that the measurement system with non-uniform sampling can reduce distortion of the MOKE loops, improve the accuracy of the coercivity, and decrease the number of sampling points.

Magneto-optical properties of Fe/Cr/Fe/MgO/Fe structures epitaxially grown on GaAs(001)

Fe/Cr/Fe trilayers were epitaxially grown on atomically flat GaAs(001). For the thickness of Cr spacer layer corresponding to antiferromagnetic coupling, “reversed” minor hysteresis loops were measured with longitudinal magneto-optical Kerr effect (MOKE), i.e., a negative “magnetization” signal was detected when the thicker bottom Fe layer was saturated along the applied field. This behavior is interpreted by depth variations of the MOKE sensitivity. Magnetization reversal shows that both antiferromagnetic switching and spin–flop transition fields depend on the ratio of both Fe film thicknesses. The shape of the MOKE loops becomes more complex with further deposition of MgO and Fe layers on the top of the Fe/Cr/F/GaAs(001) stack. Superconducting quantum interference device measurements confirm the interpretation of the MOKE loops and demonstrate homogeneity and sharpness of the interfaces in the structures.

Spin-polarized electron tunneling across a Si delta-doped GaMnAs/n-GaAs interface

We study the spin-polarized tunneling of electrons from the valence band of GaMnAs into the conduction band of n-type GaAs with Si delta-doping at the interface. The injection of spin-polarized electrons is detected as circular polarized emission from a GaInAs/GaAs quantum well light emitting diode, corresponding to magneto-optical Kerr effect loops. The angular momentum selection rules are simplified by the strain-induced heavy-hole/light-hole splitting, allowing a direct relation between circular polarization and spin-polarization. Comparison with the influence of Zeeman splitting allow us to conclude a spin-injection from the majority spin-band.

Angular dependence of magnetization reversal process in patterned Co thin films

We measured longitudinal and transverse magnetooptic Kerr effect loops to understand the angular dependence of magnetization reversal in an array of 5 /spl mu/m /spl times/ 25 /spl mu/m rectangular thin Co bars with uniaxial anisotropy. When the magnetic field is applied close to the hard axis, the reversal is dominated by coherent rotation; when the field is applied close to the easy axis, the reversal is caused by domain-wall unpinning and the coercive fields observe the relation H/sub c/ = H/sub 0//cos /spl theta//sub 0/. For the other orientations, the magnetization first rotates smoothly at larger fields and then switches irreversibly by domain-wall unpinning.

Antidot density-dependent reversal dynamics in ultrathin epitaxial Fe/GaAs(001)

Easy axis dynamic magneto-optic Kerr effect loops have been obtained from ultrathin (20 Å) epitaxial Fe/GaAs(001) patterned with antidot arrays of different densities (antidot spacings s=10 and 50 μm). The external field was driven sinusoidally in time with frequency in the range 0.01 Hz–2.3 kHz. In the low-frequency regime (f<20 Hz) coercivity Hc increases with 1/s, in agreement with existing phenomenological laws of geometric coercivity scaling in quasistatic fields, e.g., Hc=ζt/x, where t is the film thickness, x is a length parameter of the magnetic structure, and ζ is a constant of proportionality. However, in the transitional region between the low- and high-frequency regimes (20 Hz<f<1000 Hz), we find that scaling parameter ζ is no longer constant and increases with frequency: at low antidot density (s=50 μm) the coercivity in the transitional region falls to a minimum, while at high antidot density (s=10μm) it remains almost a constant. The dip in coercivity for s=50 μm is attributed to a resonance of the sweeping applied field with the domain-wall propagation in the film. Meanwhile, the suppression of the dip for s=10 μm is ascribed to a change in the relative importance of wall propagation and domain nucleation mechanisms in dynamic magnetization reversal.

Magnetic anisotropy of Ni/Cu sputtered multilayers

Ni/Cu multilayers were grown by RF sputtering. The crystal orientation, the thickness of the elemental layers and the interface quality were analysed by X-ray diffraction in the θ–2 θ configuration. Magnetic anisotropy was investigated by both a static and a dynamic technique, namely magneto–optical Kerr effect (MOKE) and Brillouin light scattering (BLS). Longitudinal MOKE loops accounted for the in-plane orientation of the magnetization, while polar loops were used to determine the out-of-plane anisotropy fields, showing the existence of a minor second-order contribution in addition to the first-order one. BLS was then exploited to detect thermal excited spin waves through inelastic scattering of light. The out-of-plane anisotropy fields evaluated by this high-frequency dynamic technique compare well with the first-order values obtained by analysis of polar MOKE hysteresis cycles.

Biquadratic interlayer exchange coupling in epitaxial Fe/Si/Fe

We have studied the biquadratic exchange coupling in epitaxially grown Fe/Si/Fe. The temperature and thickness dependence of the biquadratic coupling strength were determined unambiguously by fitting the easy- and hard-axis magneto-optical Kerr effect loops. The origin of the biquadratic coupling can be fully understood in terms of Slonczewski’s loose spins mechanism.

Competition between direct exchange and indirect RKKY coupling in Fe/V[001] superlattices

Fe 10 Å /V t <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</sub> (001) superlattices are prepared by magnetron sputter deposition and characterized using x-ray diffraction (XRD), magneto-optical Kerr effect (MOKE) magnetometry, and x-ray magnetic circular dichroism (XMCD). MOKE loops show three peaks in saturation field as a function of t <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</sub> , indicating RKKY induced AF coupling at t <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</sub> =22 Å, 32 Å, and 42 Å. The expected AF coupling peak at 12 Å, however, is missing. XMCD measurements show that the V layers carry a relatively large magentic moment which extends surprisingly far from the Fe/V interface. Such 'transient ferromagentism' suggests a strong V-V direct exchange coupling within the V layer. We argue that the V-V direct exchange coupling suppresses the 12 Å AF coupling peak, but since direct exchange coupling decays more quickly than RKKY coupling, the AF peaks at 22, 32, and 42 Å are observed.

Study of magnetization reversal in ultrathin magnetic multilayers using in situ MOKE

The magnetization reversal mechanism in Co/Pt and CoCrTa/Pt thin film multilayers was investigated in situ using a magneto-optical Kerr effect (MOKE) apparatus. Multilayers Nx(Co/Pt) and Nx(CoCrTa/Pt) with different amounts of doped Cr17Ta5 were sputtered in high vacuum. The growth was interrupted at intervals of two bilayers to perform perpendicular MOKE loops in the range from N=2 to N=20. The Draaisma and de Jonge model was adapted to analyze the hysteresis loops with the domain wall energy density as a fitting parameter. The broad consistence of the experimental data and theory shows that the incorporation of Cr and Ta in the magnetic layer lowers the domain wall energy density. The model successfully predicts the thickness dependence of the loop squareness, which decreases with increasing N. According to the model, the domain repetition length d is in the range of 900–2100 Å, and the domain wall energy density σw is in the range of 1.68–11.1 erg/cm2.

Determining magnetic anisotropies from hysteresis loops

A novel method for determining magnetic anisotropies from hysteresis loops is presented. While sweeping the loop, a magnetic field is applied perpendicularly to the sweep-field axis. This causes the magnetization to rotate reversibly in a wide field range and still reach saturation at finite fields. An example is given whereby surface and volume anisotropies are determined from magneto-optical Kerr effect loops in Co films grown on stepped Cu(001).

Interlayer exchange coupling in epitaxial Fe/Cr/Fe/Ag/GaAs(100) structures

The interlayer exchange coupling has been investigated in epitaxial Fe(20 Å)/Cr/Fe(20 Å)/Ag/GaAs(100) structures that contain a wedge-shaped (0–40 Å) Cr layer. Longitudinal and polar magneto-optical Kerr-effect (MOKE) and Brillouin light-scattering measurements have been combined to determine values for the relevant anisotropy constants and both the bilinear and biquadratic coupling strengths. The phase and period of the oscillations in the interlayer coupling are found to agree well with those reported by other researchers while the total coupling strength is found to be reduced. This reduction is presumably due to the presence of structural imperfections in our samples, and our results may therefore be of use in testing some of the recently proposed extrinsic biquadratic coupling mechanisms. Specifically, we find that for the Cr thicknesses studied the biquadratic coupling strength in our samples varies as d−1.4Cr where dCr is the thickness of the Cr layer. We also present results that show how the ultrathin Cr limit may be investigated. We show that the coercivity of the easy axis MOKE loops is sensitive to submonolayer coverages of Cr and that polar MOKE is sensitive to the strong ferromagnetic coupling found in the 0–4 Å Cr thickness range.

Anisotropy and orientational dependence of magnetization reversal processes in epitaxial ferromagnetic thin films.

We have undertaken a detailed study of the macroscopic and microscopic magnetization reversal processes in epitaxial ferromagnetic thin films with varying cubic and uniaxial magnetocrystalline anisotropy strengths. The macroscopic magnetization reversal processes were observed with in-plane magneto-optic Kerr effect (MOKE) vector magnetometry as a function of the relative anisotropy strengths and on the orientation of the applied field with respect to the anisotropy directions. Measurements of the component of magnetization in the plane of the sample and perpendicular to the applied field allow a precise determination of the relative orientation of the hard and easy in-plane anisotropy axes. This can be used to accurately determine the ratio of uniaxial to cubic anisotropy constants, when the ratio is less than one. The ratios obtained from MOKE agree well with those obtained by Brillouin light scattering (BLS). MOKE vector magnetometry reveals loop features that can be associated with either one or two irreversible jumps in the direction of the magnetization, depending sensitively on the anisotropy ratio and the orientation of the applied field. Minimum-energy calculations predict that the reversal process should proceed by a continuous rotation of the magnetization vector with either one or two irreversible jumps between single-domain states, depending on the applied field orientation and the nature of the anisotropy of the film. The calculations provide a good qualitative description of the observed reversal process, although the magnetic microstructure influences the exact values of the switching fields.Sequences of Lorentz microscopy images were made in the vicinity of the two switching fields to reveal the microscopic reversal mechanism. These images provide direct evidence that the reversal proceeds by domain-wall motion between single domain states. In the case where the applied field is oriented away from the easy axis direction, a single domain intermediate state is observed during magnetization reversal, as suggested by the MOKE measurements and predicted by the calculations. In addition, the reversal mechanism with the field applied close to an easy axis reveals that the single jump observed in the MOKE loops actually corresponds to a two-step process on a microscopic scale.