F. B. Humphrey

Large Barkhausen discontinuities in Co‐based amorphous wires with negative magnetostriction

Magnetic properties, such as domain patterns and anisotropy, were measured for negative magnetostrictive Co‐Si‐B amorphous wires exhibiting large Barkhausen discontinuities and the results are compared to those of Fe‐Si‐B wires with positive magnetostriction. The Co‐based wire was found to have a bamboolike domain structure at the wire surface. It was also shown that the amorphous wires prepared by the in‐water quenching technique store tensile stress in the radial direction. The magnetostrictive anisotropy due to residual stress will produce an axial component of magnetization in conjunction with the two‐dimensional geometry of wires making both Co‐ and Fe‐based wires exhibit large Barkhausen discontinuities along the axis of the wire.

Magnetic Properties of Co-Si-B Amorphous Wires Prepared by Quenching In-Rotating Water Technique

The authors measured the magnetic properties of Co-Si-B amorphous wire with negative magnetostriction, prepared by quenching in-rotating water, and results were compared with the properties of Fe wires with positive magnetostriction. The Co wire was found to exhibit large Barkhausen discontinuities as well as the Matteucci effect, as does Fe wire. It was also found that the Co wire has a bamboo-like domain pattern at the surface and has a reverse domain propagation mobility larger than the Fe wire.

Observation of vertical Bloch line propagation

The presence of vertical Bloch lines in the wall of a stripe domain of a magnetic ‘‘bubble’’ material can be inferred by observing the transient distortion of the straight wall driven by a uniform field. Because the local mobility depends upon local structure in the wall, rapid motion of the straight wall results in transient bulges of the wall. These hard wall bulge sections were seen by applying a large uniform bias field pulse and then observing the transient shape of the initially straight wall using a high speed optical sampling microscope with 10 ns pulse laser illumination. Bulge phenomenon, such as generation, annihilation, and propagation along the stripes were observed in an as‐grown sample, Sm0.4Y2.6Ga1.15Fe3.85O12 with a thickness of 6.26 μm, 4πMs=125 Oe, γ=1.54*107 Oe−1, Hk=2190 Oe, α=0.208, and Q has a value of 17. Several different types of bulges have been seen. They can be different in size and can appear either on one side or both sides along the stripe domains. Shorter bulges usually mo...

Computer simulation of magnetic bubble logic devices

Operation margins for a dual‐conductor current access magnetic bubble AND/OR gate, a one‐bit full adder and a string comparator were simulated by a computer program. These gates employ bubble‐bubble interaction to perform logic functions. A bubble replicator using a third layer conductor was also simulated. For the same lithographic feature size, the comparator requires 4.5% the area of the same function implemented in silicon. The simulation procedure is based on the Hayashi method where the bubble domain wall is divided into 18 to 72 wall segments which move according to the total effective field at the center of each wall segment. The field at every point is calculated by a finite difference routine. The logic functions simulated exhibited a 40‐Oe bias margin when the drive current is 1.6 mA/μm.

Nonuniform wall motion in magnetic stripe domains and its effect on vertical Bloch line propagation

Nonuniform transient motion of magnetic stripe domain walls was observed in bubble garnet material. A number of films with 2–8 μm stripe width were used for investigation. The nonuniformity of motion varies from place to place on a submicron scale in the ‘‘perfect’’ single‐crystal garnet film. Observing the development through a time sequence shows that the nonuniform motion begins within the first wall width of displacement. The degree of the nonuniformity increases as the effective drive field decreases. This nonuniform wall motion is due to the existence of local fluctuations of wall coercivity within the garnet crystal film. The response of the stripe domain wall due to dc bias field variation shows behavior similar to Barkhausen jumps indicating the intrinsic nonuniformity of the garnet material. The nonuniform wall motion may seriously affect the reliability of the propagation of vertical Bloch lines (VBLs) in a VBL memory.

Bloch lines generation in moving stripe heads

The generation of vertical Bloch lines (VBL’s) at the head of a moving stripe domain with and without an in‐plane field was studied. For an as‐grown sample, the velocity saturation is independent of in‐plane field direction and the same velocity as an implanted sample when the in‐plane field direction is the same as the stray field in the implanted layer. When the in‐plane field direction was reversed, the wall saturation velocity is higher. Overshoot starts when the wall velocity approaches the saturation velocity and increases with drive field as the stripe head becomes asymmetrically distorted. Eventually, the stripe will move at an angle. The presence and motion of VBL’s was inferred from the shape of the stripe head during and at the end of the applied pulse field, the bending of the stripe, and the presence of overshoot. In the ion‐implanted film, after the generation of a pair of VBL’s on the stripe head, the positive (negative) VBL was pushed into the straight part of the stripe when the in‐plane ...

Potential barrier study of the border‐line defined by ion implantation in bubble garnets

This paper describes an experimental investigation of the potential barrier existing at the border of an implanted region. The technique consists of taking high speed photography pictures of a bubble moving across this barrier or away from it with the help of gradient drive fields. Results are presented for samples implanted with Ne+ 200 keV (0.5–5)×1014 unannealed and annealed subsequently up to 700 °C. A bubble is in a stable position in the implanted region when its domain wall is pinned at the border. The magnitude of the potential barrier separating the two regions is found to be 8×10−8 erg. The energy minimum corresponding to the bubble stable position is very sharp. When ion dosage increases up to 5×1014, the forces holding the bubble into the minimum do not appear to saturate. Under annealing from 300 to 700 °C, all the forces involved by the barrier decrease monotonically. A model is presented which gives a good account of the shape and the magnitude of the potential barrier.

High‐frequency propagation and failure of contiguous disk magnetic bubble devices

High‐frequency propagation and failure of contiguous disk devices in an ion‐implanted 3‐μm YSmLuCaGe garnet film were investigated by using high‐speed optical sampling. The minimum drive field, the bias field operating margin, and the bubble position with respect to the instantaneous rotating field direction were found to change with the frequency of the drive field. The shape and position of bubbles during the propagation were measured so that the instantaneous velocity dependence of the frequency of the drive field could be calculated. This study was done on a selection of different tracks so that the dependence of the geometry of the tracks could be included. It was found that more uniform instantaneous velocity leads to lower minimum drive field and a higher rotating frequency limit. The cause of high‐frequency failure was examined and found generally to be the limitation of the maximum velocity of bubbles causing a phase lag greater than 90°. This limitation can be caused either by a low drive or by ...

Annihilation of vertical Bloch lines during creep in magnetic bubble materials

Additional motion of a magnetic bubble following a gradient field pulse may be induced through the application of a relatively small uniform z‐field pulse. This phenomenon, known as ’’creep,’’ is studied in detail using a high‐speed optical sampling microscope. Results for a (YGdTm)3(FeGa)5O12 sample indicate that the distance through which the bubble may translate during creep is roughly quantized in units of about 0.6 μm. This distance is attributed to the annihilation of a single pair of oppositely wound vertical Bloch lines (VBL’s). Observed motions of 1.2, 1.8, 2.4, ... μm correspond to the annihilation of 2, 3, 4, ... VBL pairs, respectively. Based on the Bloch curve model, corresponding theoretical values are obtained and show close agreement with experiment. In addition, experimental values for the rate of VBL annihilation are obtained. In each case the annihilation time for a single VBL pair is between 40 and 65 ns. All experimental results support the assertion that VBL annihilation must be acco...

Radial expansion dynamics of a magnetic bubble domain

High‐speed laser photography was used in two measurements on the radial expansion dynamics of unichiral, S = 1 magnetic bubbles during and after the application of large 20‐, 40‐, and 57‐Oe expansion pulses. The bubble diameter was photographed at varying intervals after the beginning of a 120‐ns pulse and the chirality change resulting from the application of pulses of widths from 17.5 to 120 ns measured. Plots of the probability of chirality change versus pulse width were used to infer the rate of rotation of the magnetization within the wall. Results are compared with the one space dimensional Schryer‐Walker model. The features of high‐speed initial motion followed by a slower uniform average velocity agree between the model and the data, but the large transients predicted when the pulse leaves the bubble in the unstable Neel wall state were never observed. It is necessary to increase the Gilbert damping parameter a by fourfold in order to get quantitative agreement between the model and the diameter d...