Nondestructive evaluation of steels is both an important engineering need and a scientific challenge. Recent advances include a SQUID imaging technique that is capable of detecting small changes in stray fields at reasonable spatial resolution. The technique is motivated by the fact that the stray fields, determined by the underlying domain configurations in the material, would provide insight into the microstructure including defects, provided the interactions between the microstructure and the domain walls can be well understood.Initial work is being carried out on 1018 low-carbon steels, plastically deformed in a controlled fashion, with the goal of correlating SQUID images with magnetic domain images measured by Lorentz microscopy in a TEM.

Experiments were carried out in a CM200 FEG TEM using the OL field in low-mag (LM) mode, based on a procedure described by Verbist et al [1] and Daykin et al [2].

In the article Characterization of the Absolute Crystal Polarity across Twin Boundaries in Gallium Phosphide Using Convergent-Beam Electron Diffraction, by Dov Cohen, Stuart McKernan, and C. Barry Carter, because of a printing error, Figures 4 and 5 were not printed in complete form. The publisher regrets the error. The figures are reprinted here as they should appear.

The new JEOL 3000F high resolution electron microscope at BNL has been optimized for Lorentz imaging. The necessary field-free environment around the sample is obtained by switching off the objective lens in the free-lens control mode, and the associated reduction in magnification is compensated for by a Gatan post-column image filter (GIF) at ∼ 20x magnification. Fresnel imaging is obtained by defocusing with objective mini-lens (OM). The use of low angle diffraction with an aperture located at the back focal plane makes it possible to obtain Foucault images.

In-situTEM dynamic magnetizing experiments combined with Lorentz magnetrc microscopy both in Fresnel and Foucault modes were used to characterize the magnetic structure of some hard and relatively soft magnets, Nd13.75Fe80.25B6 and Pr13.75Fe80.25B6, prepared by different processing routes. The goal of these in-situexperiments was to develop a reliable and effective procedure to search for, identify, and classify the different pinning centers present in real magnets that accord with their “pinning power” versus applied magnetic field.

In the last decade since its development, magnetic force microscopy[l] has emerged as a workhorse in imaging magnetic structures at the sub-micron length scales. It possesses the desirable attributes of robustness, straightforward implementation and a fairly well characterized image contrast formation. In recent years, we have successfully implemented MFM in the presence of a highly controlled external magnetic field.[2] Using this technique, it is possible to follow the sample’s magnetic evolution at various points along it’s magnetization curve. Further, by using standard software implementation, the images can be presented as an animation of the micromagnetic process. We applied this technique to study a variety of slow varying dynamics of magnetic systems, including the dc erasure of thin film recording media[3], the mechanisms of moment rotation and reversal, and the domain wall motion nanostructured magnetic elements[4,5].

In this talk, I will review the rudiments of the technique and show the “dynamics” of the magnetization of cobalt and Permalloy alloys interacting with external fields.

Tremendous progress has been made in the field of magnetic materials research and technology over the past few years. Superior properties and novel scientific questions arise due to our ability to either synthesize artificial structures or to tailor microstructures at the appropriate length scales. Microstructure, generally defined as the morphological arrangement of crystallites with similar or different phase constitutions and of the crystal defects, is controlled by the processing, which in turn affects the phases present, their topology and their dispersion. There are two key length scales that determine the magnetic behavior of materials. One is the characteristic length scale associated with the dimensional characteristics of the magnetic phenomenon under consideration. The second is the size of the microstructural feature of interest - grain size, shape, isolation, interfaces, inter-granular phases etc. It is important to identify the range where these two length scales overlap, for it is there that novel properties and phenomenon are usually observed.

Performance of permanent magnets for a variety of applications is often determined by the maximum energy product (BH)max. In order to obtain high (BH)max permanent magnetic materials have to have large coercivity. In theory the coercive field of ideally oriented, non-interacting, single domain, magnetic particles, assuming Kl is much bigger than K2, was shown to be He = 2K1/Ms - N Ms, where Kl and K2 are the magnetocrystalline anisotropy constants, Ms is the spontaneous magnetization and N is the demagnetization factor. For randomly oriented non-interacting particles the Stoner-Wohlfarth model predicts that the value of Hc decreases to about half. However, experimentally obtained values of the coercitive fields in permanent magnets are 3 to 10 and 2 times smaller for well oriented and randomly oriented samples, respectively. This discrepancy was attributed to inter-particle interaction and the microstructure of the permanent magnets. In order to understand the difference between the theoretically predicted and experimentally obtained results for He we prepared rapidly quenched, Nd-rich, NdxFe14B (2 < x < 150) ribbons.

Magnetic recording forms the basis of one of the largest international industries. Electron microscopy has of course played a significant role in relating magnetic performance to the underlying microstructure. However, relatively little attention has been paid to the ways in which imaging of the magnetic structure can be correlated with processing conditions and microstructural parameters. The present research has been aimed to address this deficiency.

A method was devised to prepare samples with large electron transparent areas (e.g., 250 microns square) from real computer hard discs. Fresnel TEM images were taken in a Philips CM20 FEG equipped with a special Lorentz pole piece, and with the objective lens off. Upon defocusing, the ferromagnetic domain structure strikingly appears. The characteristics of the recorded tracks depend on the magnetic state of the disc prior to magnetic “writing”, and much information about the local direction of magnetization could be obtained from the nature of the magnetic “ripple” structure.

Magnetic granular films, which consist of magnetic fine particles of nanometer size and a nonmagnetic matrix, have been of great interest for new magnetic materials. Their magnetic characteristics strongly depend on the fine structures, besides the combination of magnetic and nonmagnetic constituents, especially their surface topology and the particles size. Although the relationship between these fine topological structures and the macro-magnetic characteristics has been investigated systematically by many researchers, there has remained a desire to observe magnetic microstructures directly. We succeeded in observing the magnetic structure of fine particles in granular films using Lorentz microscopy.

Two kinds of specimens were observed. One was an Fe-Mo granular film. An amorphous film of Fe-Mo was prepared by dc magnetron sputtering on amorphous carbon thin films at a temperature below 180 K. The films were annealed at 770 K within an in-plane magnetic field. The other kind was an Fe-MgO granular film, where iron particles were embedded in a magnesium oxide single crystal.