PORTLAND, Ore. Despite tough talk from solid-state memory makers about the demise of hard disks, ingenious disk drive engineers have used trial-and-error to keep one step ahead.
Now, physicists have proposed theoretical advances to keep the status quo. Recently, a physicist at the University of California, Santa Cruz and a researcher at Hitachi Global Storage Technologies collaborated to create an angstrom-scale model of magnetic domains that they say will take the trial-and-error out of magnetic media design.
"We've used several separate approaches to understand magnetic media theoretically," said Joshua Deutsch, professor at the UC Santa Cruz. "Now we just need disk drive researchers to verify these predictions experimentally."
Deutsch began modeling magnetic media on his own, but he started modeling disk drive media in earnest after giving a seminar to Hitachi Global Storage Technologies. Deutsch and Andreas Berger of Hitachi Global Storage Technologies created a detailed model of switching in magnetic media that includes the physics of wobbling, or "precession," that they could pack into a single computer program.
The crux is that disk drive researchers have had to depend too heavily on experimentation because they model magnetic spins as a simple binary operation that they call an a avalanche. However, because magnetic domains on the media's surface are real molecules, the actual physics is more like flipping over a spinning top, followed by a period of damped wobbling.
"Researchers studying magnetic avalanches use much simpler models that don't try to correctly capture the detailed motion of spins at a small scale," said Deutsch. "Our work applies a more accurate micro-magnetic approach to the problem of avalanches."
Deutsch's and Berger's model contains many detailed physical phenomena, but central is its model of spin precessionthe wobbling that resembles the way a spinning top stabilizes its spin. After a magnetic domain, or bit, switches from a one to a zero, or vice versa, it stabilizes over a few nanoseconds. During this time, its magnetic field exerts a destabilizing influence on all of its neighboring magnetic domains, a phenomenon Deutsch and Berger claim is largely ignored by current disk drive researcher's models.
A more detailed model could help disk drive researchers design media that quickly dampens magnetic domain precession while maintaining optimal signal strength for the tiny disk drive read heads. Detailed models could become even more critical to disk drive media designers as the industry marches ahead with perpendicular recording techniques that contain magnetic domains vertically in layers as thin as 500 angstroms. Over the next few years, magnetic media designers hope to pack as many as six angstrom-scale layers with alternating damping characteristics, called exchange-coupled composites, with a goal of boosting density from today's 200 gigabits per square inch to future densities as high as 1 terabit per square inch.
Instead of exhaustive trial-and-error techniques designed to measure the ability of new magnetic materials to damp out avalanches, Deutsch and Berger model the underlying causes of damping, even including the influence of electrons and physical vibrations, called phonons, in the rotating media.
Now that they have the world's most comprehensive model of magnetic avalanches, Deutsch and Berger are counting on disk drive researchers to provide independent verification of their model.
"We already have created a model with which we can calculate rather than just measure," said Deutsch. "It is just getting practical to [perform] the ultrafast measurements needed to verify our model experimentally, so I hope that such measurements will be done in the near future."