February 24, 2009
When it comes to the future of data storage technology, the great debate between bit patterned media and HAMR, DSSC graduate student Charles "Chip" Hogg is betting on bit patterned media. That gamble recently paid off in a big way when the physics PhD student received the Astrid and Bruce McWilliams Graduate Fellowship from the Mellon College of Science (MCS) at Carnegie Mellon.
Established in 2007 by Department of Physics alumnus Bruce McWilliams and his wife, Astrid, the fellowship provides tuition, stipend and fees for up to one year of graduate study, as well as $1,000 for travel or other research expenses. Two graduate students receive the honor each year to support their work in nanotechnology, biophysics or cosmology — areas where MCS students continue to have an increasing impact.
In Hogg's case, that impact will likely come from his work on patterning substances at very small scales. At a basic level, his research involves taking a flat surface and changing it to have some kind of pattern. Researchers typically tackle this problem in two stages: making the pattern in an easily arranged system and transferring it to the surface of the desired material. Under the guidance of Physics Professor Sara Majetich, a group of physicists at Carnegie Mellon has gained expertise in the first stage, making uniform iron oxide (Fe3O4) nanoparticles, each coated with a layer of surfactant — long molecules that act as springs and prevent the particles from clumping. Because of this surfactant, and the uniformity of the particles, they spontaneously form arrays with long-range order, a process known as "self-assembly."
The patterns these self-assembled monolayers form hold great promise for the data storage industry, which continually seeks to increase the number of bits that can be stored per unit area, known as areal bit density. Conventional data storage is limited to a little more than 0.8 terabits per square inch (tbpsi). Hogg's pattern has a density greater than 3 tbpsi, and the method promises to be scalable beyond 10 tbpsi.
With the making of suitable patterns well in-hand, "the focus of my project is really the transfer of that pattern," said Hogg. This transfer can be problematic, though, because of a catch-22 his research uncovered. The surfactant between the particles must be removed, or the etching ions are blocked from reaching the underlying substrate. But that surfactant is what stabilizes the pattern, and as soon as the surfactant is removed, the particles clump together and the pattern is destroyed. Hogg's solution was to flip the problem upside-down: he deposited the layers to be patterned on top of the nanoparticles, effectively taking over the role of pattern stabilizer from the surfactant. Using a series of highly selective etches through ever-thinner layers, Hogg completely removed the original wafer, exposing the backsides of the now-secure particles to further processing. He transferred that pattern into an intermediate SiO2 film using a gentle reactive ion etch (RIE) process that ate away the newly exposed areas between the particles but preserved the areas under the particles. After he removed the nanoparticle mask with a wet-acid treatment he developed, the underlying SiO2 showed the same pattern as the original mask. "Chip is the only person in the world who can make features this small [two nanometer gaps] with a massively parallel writing process," said Majetich, Hogg's advisor.
"Chip is self-critical, curious and thorough — and shows tremendous promise as a researcher," said James A. Bain, professor of electrical and computer engineering and associate director of the DSSC. "His research has involved some of the most creative work in the nanofabrication facility at Carnegie Mellon that I have participated in at the university in a decade."
Hogg's ultimate goal is to prove that this nanomasking technique can be adapted as the drive toward ever-greater bit densities continues. In a world where retooling is expensive, data storage researchers need a technology that can be implemented and scaled to easily achieve the next generation of storage material. Thus far, Hogg's research has exposed a creative technology that can generate high-density material in the short term, and offers the potential for the scalable, small, affordable solution needed for the future.
The McWilliams Fellowship gives Hogg the chance to actually finish this important project. "When I looked at the work I was doing, I could see that it had great and tremendous potential to impact the data storage industry, but we were probably going to run out of money before I finished my doctorate," Hogg said. "This fellowship will make the difference between not finishing the project satisfactorily and really leveraging the work we've already put in and the expertise I've gained. We would not be able to reach the higher impact journals or present this work at important conferences without this fellowship and the grant money it saves by paying for my stipend and the additional $1,000 that will go toward conference fees."