MISC Research Projects

Patterned film media for signal fidelity, signal stability, and tracking is gaining acceptance as a viable recording scheme, especially with the recent advances in nanolithography. Washington University and Rowan University have developed a technique that is promising for large-scale application to the recording industry. This method enables small-scale features (less than 25 nm); can be used as a resist pattern for selective removal or for liftoff; has large-scale extensibility; and is especially applicable to disk manufacturing since non-rectangular and circumferential patterns can be replicated. While excellent progress has been made, several fundamental issues need to be resolved before realistic systems employing patterned media can be designed.
In complementary research we have proposed a practical new method to fabricate controlled nanostructured surfaces using conventional thin film deposition techniques. This approach modifies the growth of discontinuous films from the vapor phase. Thin films are typically deposited on surfaces maintained at uniform temperatures. Consequently, the initial stages of film growth are characterized by nanoclusters having a random distribution in spatial arrangement as well as in size. We will direct the clusters to occupy pre-determined spatial positions by using inhomogeneous temperatures Ts(x, y). These temperatures will also allow controlled changes in the cluster size from a few nanometers upwards.

Perpendicular magnetic recording is a promising candidate technology for realizing high areal densities. One of the most remarkable features of this recording paradigm is that the demagnetizing fields in the transition region are minimum, as compared with being maximum for conventional longitudinal recording. This may have a huge impact on the recorded signals, recording noise, and recording stability. An often-overlooked phenomenon is that medium noise has magnetization dependent characteristics that are spatially correlated with the signal. Due to the differences in their respective demagnetizing fields, transition noise behavior may be different in longitudinal and perpendicular recordings. Micromagnetic modeling and experimental evidence suggest noisy and jagged transition profiles even for certain high squareness perpendicular media. Medium demagnetizing fields, underlayer features, a finite write head rise time, and the vectorial nature of the head fields might all contribute to this noise. We are detailing results from micromagnetic simulations and developing an analysis that explains the underpinnings of these noise effects in perpendicular recording. Since exchange-coupled perpendicular media may be used instead of conventional exchange-decoupled granular media, we are also investigating high speed dynamic as well as quasi-static switching behavior.

One method to improve magnetic recording areal density is two-dimensional (2D) recording. This scheme explicitly records data in 2D patches, discarding the conventional concept of tracks. The retrieved signal from the patches may suffer from 2D ISI. We are investigating three detection methods and two iterative decoding schemes for a straightforward 2D ISI channel model.

This research explores measurements of transition jitter due to write jitter as well as media jitter. This research is targeted to examine the structure of transition jitter. We have developed a method to separate write and media jitter as well as glean insight into the media jitter structure. Armed with these data we expect new direction toward improvements in channel capacity.

We have developed a method of ultra-high resolution magnetic imaging in situ to the drive. Enhanced magnetization images are produced through deconvolution of the linearized readback by a synthesized two-dimensional head response kernel. This work shows improvements in resolution and detail in both downtrack and crosstrack directions. While the current sub-30 nm resolution of this technique is on the same order as MFM, there are numerous advantages: this is a nondestructive tool, it is performed in situ to a commercial HDD, has ultra-high resolution, high SNR, and this technique can be performed easily (at the manufacturing or development site) without any elaborate tools. We are applying this technique to recording investigations and are exploring failure analysis.

The average database size and associated software support systems are growing at rates greater than the increase in processor performance (i.e., more than doubling roughly every 18 months). This has led to large increases in the time associated with retrieving and processing such large databases. Our work focuses on developing new methods for performing a variety of basic database operations at the level of the disk head and thereby improving database performance of about two orders of magnitude. We are working on a novel mechanism for facilitating such searches. Instead of translating the magnetic signal into bits that are then indexed by the CPU, we propose to recruit the high-speed parallel magnetic sensing systems already present in modern magnetic storage devices to facilitate searches. We propose searching these databases directly, without CPU, RAM, and bus-bandwidth limitations. In our approach an index need not be developed (only to become invalid when the content changes). The database content can be modified, and our “indexing” is done on the fly, using sensors and buffers on the disk. Our results should improve the speed and cost of performing approximate matches within large spaces by orders of magnitude.

This work addresses the fundamental underpinnings of the proposed and yet to be developed spin devices including sensors, memory systems, and reconfigurable logic. We have partnered with NVE Corporation, a world leader in novel spin electronic device. The work in MISC is a scientific investigation of the physical underpinnings of some spin electronic devices. This understanding will lead to the immediate benefit of assisting the industry in directing it toward solving some of their most pressing, current issues and have the long-term benefit of enabling new technology and systems by providing the understanding needed to analyze devices and direct the design of novel materials, devices, and systems and provide the components of practical spin injection devices.