Romney R. Katti

Magnetic force microscopy images of magnetic garnet with thin‐film magnetic tip

We present magnetic force microscopy images of YGdTmGa/YSmTmGa magnetic garnet, using a thin magnetic film deposited on Si3N5 atomic force microscopy tips. We have found correlations between the topography and the magnetic domain structure. We show that by using either magnetized Fe‐Ni bilayer tips versus unmagnetized single layer Fe tips that the image contrast shows domains versus domain walls, respectively.

Stripe stabilization in vertical Bloch line memory

Partially grooved, long, rectangular grooves on the garnet surface were used to stabilize minor loop stripes in a vertical Bloch line memory. These stripes reside beneath the stripe confinement groove. The test chip contained 10 or 20 minor loop grooves, 10 read/write (r/w) gate grooves, and a major line groove. The film thickness was 4.76 μm, and the groove was 0.5 μm deep. Each minor loop groove was 500 μm long, four widths were tried: 3.5, 4, 4.5, and 5 μm. Two groove periods were tried: 10 and 20 μm. For stripe initialization, a bubble was nucleated in every minor loop groove by a single current pulse at a 90 Oe bias field. As the bias field was decreased slowly, bubbles began to stripe out at 80 Oe. At 75 Oe, all 10 stripes were stretched to the full groove length. These stripes were stable at bias fields between 70 and 80 Oe. Stripes decreased in length at higher fields, and escaped from the confinement groove at low fields. When stripes escaped from the groove, it always occured from the end not fa...

Major line operation in vertical Bloch line memory

The operation of a major line in a vertical Bloch line memory using a partial garnet grooving architecture was studied experimentally and numerically. The major line was fabricated on a 5‐μm bubble garnet using three conductor layers and a 10% garnet grooving. The major line contains a bubble generator, a bubble propagation track, a bubble expander for bubble detection, and a bubble annihilator. The minimum current for bubble generation was 350 mA, at 50 ns pulse width. The bubble propagation track was a typical dual conductor design with a 5‐μm‐wide conductor and a 20‐μm circuit period. The minimum drive current was 5 and 10 mA at an operating frequency of 250 and 500 kHz, respectively. The bias field margin was from 84 to 100 Oe. The bubble expander was a modification of the propagation track. The meandering conductor and the groove width were gradually increased to stretch the bubble into a stripe. The bias field margin was from 82 to 84 Oe. The numerical model includes the effect of the garnet groovin...

Partial grooving in vertical Bloch line memory

Partially grooved regions in magnetic garnets, used to stabilize minor-loop stripe domains in vertical Bloch line memories have been investigated using arrays of grooves in garnets with a zero-field stripe-width of 2.36 mu m and thickness of 2.26 mu m. Rectangular grooved regions were used to confine the minor loop stripes, using a groove depth of 10% and groove widths of 1.5 mu m and 2.00 mu m. Major line regions were defined by two successive 10% grooving steps. Bias field margins were measured experimentally and computed using a two-dimensional, magnetic-domain computer simulation. The computed margins were shown to be in quantitative agreement with experimentally measured bias field margins. Interpretation of the simulations and the experimental data in these samples indicates that magnetostatic effects dominate magnetostrictive effects in defining bias field stability. >

Space data storage systems and technologies

The acquisition and transmission from space to Earth of data is the central part of many space missions. A data storage subsystem, based on a data storage technology, is typically used to buffer data from the spacecraft and its instruments before data are returned to Earth. Because of requirements, constraints, and the design of the mission, the selection of the appropriate data storage technology for this function is based on many considerations, such as reliability, readiness, availability, mass, volume, power consumption, capacity, data rate, nonvolatility, environmental stability, radiation insensitivity, vibration insensitivity, data management flexibility, and cost, among others. Issues are reviewed for space data storage technologies, including magnetic tape recording and solid-state memories, for past, present, and future missions, such as Voyager, Cassini, and Pluto Flyby. >

Partially grooved domain stabilization structures for vertical Bloch line memory

Bias field stability ranges were measured and numerically simulated for magnetic domains in garnets stabilized by partially grooved rectangular and ring grooves. Simulation results agree favorably with experimental results when the finite slope effects of the groove walls are included. As bias fields increase, destabilization in rectangular and ring domains occurs by runout and midstripe domain buckling, respectively. While ring domains are stable at lower bias fields than rectangular domains, bias field stability ranges are approximately equal. For the same partial grooving depth, rectangular domains are preferred because they offer higher storage density potential for vertical Bloch line storage, as long as the bit propagation margins at the stripe ends are sufficient. >

Thickness and roughness dependence of DC modulation noise in thin film magnetic recording disk media

DC modulation noise was studied as a function of magnetic-layer thickness and surface roughness for longitudinal, low-noise, nominally 1000-Oe, plated CoP and CoNiZnP, thin-film magnetic recording disk media made from controlled fabrication runs. DC modulation noise is generated when a reverse longitudinal DC field is applied to a uniformly magnetized disk. A model which allows autocorrelation functions and noise spectra to be calculated and which agrees with measured noise data (with correlations greater than 0.99) was developed. Trends observed with measured data and the model indicate that inhomogeneities in the medium, and not surface roughness, are the dominant sources of noise. >