Scientific breakthroughs can’t be scheduled. Sometimes things happen fairly quickly, like when IBM Zurich researchers Gerd Binnig and Heinrich Rohrer invented the scanning tunneling microscope, which for the first time allowed scientists to “see” individual atoms. The two physicists came up with the idea in 1981, demonstrated their invention two years later and won the Nobel Prize in Physics in 1986. In contrast, racetrack memory, a revolutionary new method for storing digital data, has come along more slowly. Its path from concept to reality demonstrates the patience and the passion that’s required of researchers who aim to produce world-shaking inventions.
Earlier this month, the scientists at IBM Almaden Labs who are working on the racetrack memory project reported an advance that gives hope that their efforts could pay off in the not-too-distant future.
Racetrack memory is a method for storing data on three dimensional microchips. It promises to pack 100 times more data in the same space as either a hard disk drive or a traditional memory chip at a much lower cost than either–while using much less energy. If and when racetrack memory is commercialized, it could represent a major advance affecting every aspect of IBM’s Smarter Planet agenda, including sensors, networks, mobile devices, and data analytics. For instance, it could make it possible to store every movie made in a year on a single portable device.
Almaden researcher Stuart Parkin had been thinking about alternatives to traditional memory and storage devices for many years, but the racetrack memory idea came to him about eight years ago when IBM sold its hard disk drive business. It occurred to him that the basic technology for the hard disk drive had been around for decades. The same was true for the technology behind memory chips. He saw that the most significant limitation to both methods was that they were two dimensional, limiting the amount of data that could be stored in a small space. “I asked, ‘Is there a better way of designing a disk drive without moving parts?’” he recalls.
Parkin came up with the concept of using a dense web of tiny magnetically charged nanowires attached to the surface of a silicon semiconductor for storing data. He combined that with the idea of reading and writing data via stationary sensors embedded in the silicon. The data is stored in so-called domain walls on the nanowires. Imagine racetrack memory as a tiny roller-coaster and the bits of data as people climbing on the cars (domain walls) and taking a wild ride up and down the nanowires.
Because the invention was so novel, Parkin received his first patent quickly–in 2004. As he and his colleagues received additional patents, they began talking and writing about their inventions publicly–including in two papers in Science magazine in 2008. Their newest missive, also in Science, explains how they established that domain walls can be moved along the nanowires precisely enough to used to store bits of data.
At this point, Parkin says, the racetrack memory concepts have been proven in experiments. The next step is to create a prototype racetrack memory device that could lead to mass production and commercialization. “We know there is no scientific roadblock in the way that could impede racetrack,” says Parkin. “The question is time and money.” He says it would be possible to produce a prototype in three years and commercialize the technology in five.
Will IBM come up with the funds to put racetrack memory on the fast track? Will it seek partners to help fund the project and spread the risk? These are questions I can’t answer.
But what’s clear already is that a novel new technology has emerged into the light, and, if and when it comes to market it could revolutionize computing and consumer electronics.
This story also illustrates one of the eternal truths of scientific exploration: Scientists stand on the shoulders of other scientists. Parkin and his research colleagues couldn’t have produced their breakthrough without the nanotechnology tools that resulted from Binnig and Rohrer’s invention of the scanning tunneling microscope.