M Scheinfein

Scanning electron microscopy with polarization analysis (SEMPA)

The high spatial resolution imaging of magnetic microstructure has important ramifications for both fundamental studies of magnetism and the technology surrounding the magnetic recording industry. One technique for imaging surface magnetic microstructure on the 10‐nm‐length scale is scanning electron microscopy with polarization analysis (SEMPA). This technique employs a scanning electron microscope (SEM) electron optical column to form a medium energy (10–50 keV), small probe ( 1 nA) on a ferromagnetic specimen. Secondary electrons excited in the ferromagnet by the high spatial resolution probe retain their spin‐polarization orientation as they leave the sample surface. The spin polarization of the emitted secondary electrons can be related directly to the local magnetization orientation. A surface magnetization map is generated when the spin polarization of the secondary electrons is analyzed as the electron beam is rastered point‐by‐point across the ferromagnet’s surface. In th...

Laser focusing of atoms: a particle-optics approach

The use of a TEM01*-mode laser beam has been proposed as a means of focusing an atomic beam to nanometer- scale spot diameters. We have analyzed the classical trajectories of atoms through a TEM01*-mode laser beam, using methods developed for particle optics. The differential equation that describes the properties of the: first- order paraxial lens hi exactly the same form as the bell-shaped magnetic Newtonian lens that was first analyzed by Glaser for the focusing of electrons in an electron-microscope objective. We calculate the first-order properties of the lens, obtaining cardinal elements that are valid over the entire operating range of the lens including the thick and the immersion regimes. Contributions to the spot size are discussed, including four aberrations plus diffraction and atomic-beam-collimation effects. Explicit expressions for spherical chromatic, spontaneous-emission, and dipole-fluctuation aberrations are obtained. Examples are discussed for a sodium atomic beam, showing that subnanometer-diameter spots may be achieved with reasonable laser and atomic- beam parameters. Optimization of the lens is also discussed.

High spatial resolution quantitative micromagnetics (invited)

Magnetization profiles at surfaces are observed with scanning electron microscopy with polarization analysis (SEMPA). This technique allows for quantitative analysis of the vector magnetization profile with 70 nm spatial resolution. Magnetization profiles in surface Neel walls which terminate bulk 180° Bloch walls at surfaces have been calculated by solving the micromagnetic equations using energy minimization. The micromagnetic calculations show that the surface Neel wall penetrates a distance from the surface comparable to a Bloch wall width and that the surface Neel wall width is at least twice the bulk Bloch wall width. The dependence of the domain wall magnetization on sample thickness is calculated for Fe, and model predictions of the wall widths that would be determined by transmission Lorentz microscopy are compared with the experimental results. The magnetic field outside of the sample, which gives rise to contrast with the Bitter technique and magnetic force microscopy (MFM), is a complicated superposition of contributions from both bulk and surface walls. Moreover, a strong mutual interaction between the sample and the MFM tip may alter the sample magnetization.Magnetization profiles at surfaces are observed with scanning electron microscopy with polarization analysis (SEMPA). This technique allows for quantitative analysis of the vector magnetization profile with 70 nm spatial resolution. Magnetization profiles in surface Neel walls which terminate bulk 180° Bloch walls at surfaces have been calculated by solving the micromagnetic equations using energy minimization. The micromagnetic calculations show that the surface Neel wall penetrates a distance from the surface comparable to a Bloch wall width and that the surface Neel wall width is at least twice the bulk Bloch wall width. The dependence of the domain wall magnetization on sample thickness is calculated for Fe, and model predictions of the wall widths that would be determined by transmission Lorentz microscopy are compared with the experimental results. The magnetic field outside of the sample, which gives rise to contrast with the Bitter technique and magnetic force microscopy (MFM), is a complicated su...

Improved low‐energy diffuse scattering electron‐spin polarization analyzer

An improved low‐energy diffuse scattering electron‐spin polarization analyzer is described. It is based on the low‐energy (150 eV) diffuse scattering of polarized electrons from polycrystalline evaporated Au targets. By collecting large solid angles and efficiently energy filtering the scattered electrons, a maximum figure of merit, FOM=S2I/I0=2.3×10−4 is achieved. Maximum measured values of the Sherman function were S=0.15. Further, the instrumental (false) asymmetry due to changes in the trajectory of the incident electron beam has been minimized by balancing the angular and displacement asymmetries. A total residual scan asymmetry as low as 0.0035/mm has been measured over 4‐mm scan fields at the Au target in the detector. This instrumental asymmetry would produce a maximum error in the polarization in a SEMPA experiment of less than 0.3% for a 100‐μm full‐field scan. Details of the design and performance of the new detector are given.

Spin‐polarized photoemission spectroscopy of magnetic surfaces using undulator radiation

A beamline has been established at the National Synchrotron Light Source to perform angle‐resolved photoemission experiments on magnetic surfaces with spin sensitivity. The system has two novel features: it uses a miniature electron‐spin polarization analyzer and it also uses synchrotron radiation from an undulator rather than a bending magnet.

Magnetic microstructure of the (0001) surface of hcp cobalt

The magnetic domain structure of the (0001) surface of a hcp cobalt crystal was investigated using scanning electron microscopy with polarization analysis (SEMPA). This is the first observation by SEMPA of both out‐of‐plane and in‐plane magnetization components. The perpendicular magnetization imaged with SEMPA showed a branched structure very similar to that previously observed by magneto‐optic Kerr microscopy. In addition, a previously unobserved in‐plane magnetic substructure was measured. The in‐plane magnetization is divided into well‐defined submicron domains that appear to reflect the sixfold symmetry of the crystal surface.

Correlations of magnetic microstructure and anisotropy with noise spectra for CoNi and CoCrTa thin film media

The authors report on two thin-film-media alloys, Co/sub 86/Cr/sub 12/Ta/sub 2/ and Co/sub 75/Ni/sub 25/, which have very different noise characteristics. The magnetic microstructure of these films was observed with SEMPA (scanning electron microscopy with polarization analysis). Anisotropy and rotational hysteresis loss measurements have been made using a torque magnetometer. The distribution of anisotropy field H/sub k/ and its width dH/sub k/ have also been measured along with its different normalized values. It is suggested that the observed magnetic microstructure can be directly correlated with measured readback noise and anisotropy differences. >

Use of thorium as a target in electron‐spin analyzers

Measurements of the effective Sherman function have been carried out for 10–100‐keV spin‐polarized electrons scattering from a thick thorium target in a retarding Mott analyzer. At 20 and 100 keV the dependence on the maximum energy loss accepted by the detector has been measured. Comparison is made with scattering from a 1250‐A gold film. Thorium is seen to have a Seff up to 30% higher than gold. This higher Seff can not only improve the figure of merit of a spin detector, but also lessen its sensitivity to instrumental asymmetries. Comparison is also made with preliminary theoretical results. Good agreement between theory and experiment is seen in the thorium Sherman function relative to that of gold.

Correlations of modulation noise with magnetic microstructure and intergranular interactions for CoCrTa and CoNi thin‐film media

This paper reports on two thin‐film media alloys Co86Cr12Ta2 and Co75Ni25 which have very different noise characteristics. The magnetic microstructure of these films was observed with scanning electron microscopy with polarization analysis (SEMPA). The variance (σ 2t) of the magnetization across the transition region was calculated. The origin of anisotropy was determined by measuring the temperature coefficients of Hc and Ms. The interaction strengths between grains, δM(H), were obtained from the measurements of the reverse demagnetization remanence Md(H) and the forward magnetization remanence Mr(H). We found that the CoNi film showed a greater degree of interparticle interaction, which may explain the observed cross‐bit linkages in SEMPA image, larger rms transition variation (σt), and higher modulation noise.

180° surface domain wall magnetization profiles: Comparisons between scanning electron microscopy with polarization analysis measurements, magneto‐optic Kerr microscopy measurements and micromagnetic models

We compare measurements of magnetization profiles across a 180° surface domain wall in a 0.24‐μm‐thick of Permalloy (Ni81Fe19), obtained with scanning electron microscopy with polarization analysis (SEMPA) and longitudinal magneto‐optic (MO) Kerr microscopy with the predictions of a bulk micromagnetic theory. Both measurement techniques yield wall profiles in accordance with the predictions of micromagnetic theory. We conclude that for micromagnetic structure with relevant length scales on the order of tens of nanometers, SEMPA and MO Kerr microscopy yield equivalent quantitative micromagnetic information within the transverse spatial resolution limits of each technique. Near‐surface effects such as enhanced surface moments, weakened surface exchange, and surface anisotropy are not important in determining the surface domain wall profiles that we observe.