D. S. Lo

Origin of Stripe Domains in Ni–Fe Films

Stripe domains have been observed in Ni–Fe films having positive, zero, or negative magnetostriction and with either tensile or compressive stress. By using the model of Saito et al. and assuming their anisotropy constant K is attributed to both the shape anisotropy K0 and the magnetostrictively induced anisotropy Ks, we can explain our experimental results quantitatively. Here K0 and Ks are determined from the stress, the magnetostriction and the geometry of a columnar grain structure (defined by Iwata et al.). From a critical thickness (the minimum thickness above which the stripe domains will be formed) vs film composition curve, it has been found that K0 is about an order of magnitude larger than Ks at a composition of 85% Ni. Therefore, the shape anisotropy, especially in the neighborhood of 81.7% Ni, is determined as the predominant factor in the perpendicular anisotropy which causes formation of stripe domains. Dependence of stripe domains on annealing, film thickness, substrate temperature, and va...

Magnetization Creep of Cross‐Tie Walls

Cross‐tie walls are known to have lower creep thresholds than non‐cross‐tie walls. This indicates that a special creep mechanism exists for cross‐tie walls but does not for non‐cross‐tie walls. Such a mechanism is described: In the presence of a pure dc easy‐axis field the magnetization antiparallel to the field buckles more, while the magnetization on the other side of the wall buckles less. This causes magnetostatic charges on the wall which cause the main wall itself to buckle in a zig‐zag fashion with bends at the Bloch lines between the cross‐ties and at the Bloch lines in the cross‐ties. When a hard‐axis field is applied, the Bloch lines between the (fixed) cross‐ties move along the wall, a situation resulting in magnetostatic charges causing the apex of the zig‐zags to move along with the Bloch lines. This distorts the symmetrical zig‐zags into asymmetrical ones having alternately long and short legs. At this point, each segment of the zig‐zag lies at a different angle from the easy axis than it di...

Variation of Stripe‐Domain Spacing with Applied Field

Theory has predicted that because of the exchange and demagnetizing energies, the stripe‐domain spacing is a function of the amplitude of the (periodic) component of the magnetization normal to the film plane. Experimentally the stripe width was reported not to vary when the normal component of the magnetization was changed by applying a field parallel to the stripes. This disagreement, which was attributed to the coercive force of the stripes, has prevented a close comparison between theory and experiment. It has been found that the application of an oscillating field above some critical value in the plane of the film and normal to the stripes overcomes the coercivity associated with the stripes and allows the domain spacing to relax to that of a lower energy state. A change of spacing up to 50% has been observed in some Ni–Fe films. Experimental curves of the stripe width as a function of amplitude of the normal component of magnetization in the stripes are in rough agreement with the theoretical curve ...

Lorentz Microscopy in Polycrystalline Ni–Fe Films 2000‐Å Thick

The use of Lorentz microscopy has been reported on polycrystalline films 1450 A and thinner. In this paper the use of Lorentz microscopy on polycrystalline films in the thickness range between 1200 and 2000 A is presented, and some of the more unexpected results discussed. The intermediate wall cross‐ties observed in a 1450‐A film as reported earlier were not found in a low‐dispersion Ni–Fe film 2000±50 A thick (the thickness measurements were made carefully with torque magnetometer, Tolansky apparatus, and loop checker with calibrated standard), but when the film was annealed in vacua so that the crystallites became enlarged and Hc and α90 increased, the intermediate wall cross‐ties were observed. The conclusion is that the biaxial inhomogenities caused the 180° Bloch wall to become a 170° wall in some segments and a 190° wall in others. Then, according to intermediate wall theory, these non‐180° walls must have a Neel component which alternates along the length of the wall, thereby causing the cross‐tie...

Magnetic Switching in Ni‐Fe Single‐Crystal Films

Quasistatic magnetic switching in Ni‐Fe single‐crystal films for different orientations of the reversal fields were studied both experimentally by Lorentz microscopy and theoretically. It was found that the switching processes can be explained quantitatively by a theory combining the Stoner‐Wohlfarth rotational switching curve and the theoretical wall motion switching curve derived in this paper from a simple model. Eight typical switching processes are illustrated by Lorentz micrographs with explanation for two types of films; one type has a higher ratio of coercive force to biaxial anisotropy field than the other. Both wall motion switching and rotational switching were observed. In some cases, the complete switch takes place in two separate consecutive stages, each stage corresponding to a different easy axis direction 90° apart, each stage being by a different switching mechanism. Cross‐tie walls were observed in single‐crystal films in the proper thickness range. Although the films studied had nearly...

Printed Circuit Board Evaluation by Holographic Interferometry

Application of real time holographic interferometry for the examination of a printed circuit board is described. The procedures and the resulting holographic interferograms of two specific studies are presented: (1) The heat conduction of a spacer used for mounting transistors on a heat sink, and (2) the detection of inner layer short circuits on a multilayer printed circuit board.

Stripe Domains in Thin Co–Al Films

Stripe domains were observed in sputtered Co–Al films 350–700‐A thick. This type of domain structure has been reported previously only in much thicker ferromagnetic films. The structure of the films determined by electron diffraction showed the stripe domains were associated with a bcc structure similar to iron. Saturation magnetization was measured with a vibrating sample magnetometer and was found to vary linearly with film composition, but had no relationship to the formation of stripes. The domains were observed by Lorentz microscopy and could be removed by a saturating field, but reappeared, contrary to the case reported on Ni–Fe films when an ac field was applied.