Graphene Could Make Nonvolatile Molecular Memory

European researchers build graphene-based switches

PHOTO: Jannik Meyer

18 August 2008—Researchers have discovered a new way to switch current on and off in graphene, pointing the way to the possibility of molecule-size memory.

Graphene is a 1-atom-thick carbon molecule in which electrons flow 100 times as fast as they do in silicon. In theory, a graphene transistor would be 100 times as fast as the same device made of silicon. One challenge, though, is that graphene is so conductive that it’s hard to stop current from flowing, and such on-off switching is necessary for any sort of transistor.

Now, with help from the scientist in England who first isolated graphene, a group of researchers at AMO, a nanotechnology company in Germany, has come up with a novel mechanism for making a graphene switch. According to research published in the August issue of IEEE Electron Device Letters, the researchers found that by applying an electrical field to the material, they could cause a chemical change that altered the conductance of graphene. They built a transistor-like structure from graphene in which the graphene bridged two electrodes while a third electrode sat between them, separated from the graphene by a thin layer of silicon dioxide dielectric. The group applied a voltage of –5 volts to the gate electrode, causing the conductance of the graphene to decrease by more than six orders of magnitude and essentially cutting off the flow of current. When they reversed the voltage to +5 V, the conductance returned to nearly what it had been before the voltage was applied.

The group fabricated and tested tens of field-effect devices. The researchers called them devices rather than transistors to emphasize that they operate by a different mechanism than conventional transistors. In one test, they switched a device to the off state, removed the voltage, and left it that way for two days before switching it back. This experiment suggests that the system could be used to create nonvolatile memory. Although other types of memory elements are limited in how small they can get, it might be possible to shrink graphene memory down to a single molecule, says Max Lemme, one of the lead authors of the paper from AMO and now a Humboldt Research Fellow at the Center for Nanoscale Systems, at Harvard University. ”With graphene, we believe it can, in principle, be scaled down to a 1-nanometer-by-1-nanometer device,” Lemme says. The switching is not fast enough to be used in a logic circuit, he says, and researchers have not yet shown that it will work for the millions of cycles a memory device would require.

Lemme says the group hasn’t entirely explained how the change works, but thanks to work by Andre Geim of the University of Manchester and Andrea Ferrari of the University of Cambridge, in England, they believe the electric field is causing a hydroxyl molecule to attach to the graphene, changing it to graphene oxide. (The hydroxyl is borrowed from the silicon dioxide dielectric that sits on top of the graphene.) ”This is a controllable reaction,” Geim says. ”You do not destroy the [graphene] backbone itself. You just detach and attach molecules and change the electrical properties.”

”This is potentially a very exciting development in nanoelectronics,” says Michael Fuhrer, a physicist at the University of Maryland’s Center for Nanophysics and Advanced Materials. He says the technique ”opens a new route to nonvolatile memory,” if it can be shown to be fast enough to compete with other emerging technologies, such as phase-change memory.

Lemme says graphene research is still in the very early stages, and it could be more than a decade before their switching mechanism could be applied to some actual circuit. What’s really needed, he says, is a way to deposit graphene on a large scale instead of the small pieces researchers must work with today.

Geim agrees that there’s a need for both more engineering and more fundamental research. Still, he says, being able to electrically control a chemical reaction so directly ”has opened up a completely unexpected direction” in molecular electronics.

About the Author

Neil Savage writes from Lowell, Mass., about lasers, LEDs, optoelectronics, and other technology. In the June 2008 issue of IEEE Spectrum he summarized the recent breakthroughs in graphene transistors.

Advertisement
Advertisement