Nov 23 2008
A team at Rice University has figured out that a strip of graphite only 10 atoms thick can be broken with a jolt of electric current -- and repaired with another. Over and over.
That's called a switch, and it could start a revolution in data storage as part of Professor James Tour's quest to bring molecular computing to everyday devices. Think about a handheld device that can hold every movie you want to see -- along with all the others you're likely to see -- and you'll get a taste of what Tour's new type of solid-state storage could achieve.
The research is available online in Nature Materials. In the paper, Tour and postdoctoral researchers Yubao Li and Alexander Sinitskii describe a memory device that takes advantage of the conducting properties of graphene. Tour said such a device would have all kinds of advantages over both today's state-of-the-art flash memory and other up-and-coming technologies.
For starters, it would increase the amount of storage in a two-dimensional array by a factor of five. The individual bits could be made smaller than 10 nanometers, compared with the 45-nanometer circuitry in today's flash memory chips, and the new switches can be controlled by two terminals instead of three, as in current chips.
Addressing each bit with two wires instead of three is key, because simplifying chip architecture makes three-dimensional memory practical. Graphene arrays can be stacked, multiplying a chip's capacity with every layer, said Tour, Rice's Chao Professor of Chemistry as well as a professor of mechanical engineering and materials science and of computer science.
Even better, being essentially a mechanical device, such chips will consume virtually no power while keeping data intact -- much the same way today's e-book readers keep the image of a page visible even when the power is off.
What distinguishes graphene from other next-generation memories is the on-off power ratio -- the amount of juice a circuit holds when it's on, as opposed to off. "It's huge -- a million-to-one," said Tour. "Phase-change memory, the other thing the industry is considering, runs at 10-to-1. That means the ‘off' state holds, say, one-tenth the amount of current than the ‘on' state."
Electrical current tends to leak from an "off" that's holding a charge. "That means in a 10-by-10 grid, 10 ‘offs' would leak enough to look like they were ‘on.' With our method, it would take a million ‘offs' in a line to look like ‘on,'' he said. "So this is big. It allows us to make a much larger array."
While generating little heat itself, graphene memory seems impervious to a wide temperature range, having been tested from minus 75 to more than 200 degrees Celsius with no discernable effect, Tour said. "Below that, it seems to stick. After all, it's a mechanical motion, and minus 75 C is pretty cold."
That temperature range allows graphene memory to work in close proximity to hot processors. Better still, tests show it to be impervious to radiation, making it suitable for extreme environments. If you're headed to Mars in a couple of decades, this is what you'll want.
Tour said the new switches are also fast; in fact, they react faster than his lab's current testing systems can measure. And they're robust. "We've tested it in the lab 20,000 times with no degradation," said Tour. "Its lifetime is going to be huge, much better than flash memory."
Best of all, the raw material is far from exotic. Graphene is a form of carbon. In a clump it's called graphite, which you spread on paper every time you use a pencil.
The technology has drawn serious interest from industry, said Tour, who's now working on manufacturing techniques. He said it's possible to deposit a layer of graphene on silicon or another substrate by chemical vapor deposition. "Typically, graphene is very hard to think about fabricating commercially," he said, "but this can be done very easily by deposition. The same types of processes used right now can be used to grow this type of graphene in place."
Tour has no illusions about what supports the market for memory. "What really drives technology is entertainment, the big-market stuff," he said. "Cameras, games, cell phones … it's not like somebody's saying, ‘Let's make better memory for science.' It's for massive consumer use.
"I'm not very good at making predictions, but someone at Intel once told me that from the time a technology is successfully tested in the lab, it takes about eight years until it's on the market."
Eight years isn't very long to wait for a handheld device that could hold all those movies and so much more. But Tour's new -- and successfully tested -- technology might make it possible.
"This shows a lot of promise," he said, "in spite of the many obstacles to development that are always on the way to a sellable product."