APS News

Graphene's Unique Properties Offer Much Potential

Graphene holds lots of promise for future nanoscale electronics.
Graphene holds lots of promise for future nanoscale electronics.

A two-dimensional sheet of carbon, called graphene, has many of the same interesting properties as one-dimensional carbon nanotubes (CNTs), according to several papers presented at the APS March Meeting in Baltimore. Electrons can move at high speeds through the material–so fast that their behavior is governed by relativity rather than classical physics. They also suffer little energy loss, making graphene an ideal candidate for future electronics applications, especially at the nanoscale.

To date, much attention has focused on CNTs as holding the most promise for nanoelectronics because they conduct electricity with virtually no resistance. But there are some serious obstacles to scaling up CNT-based devices to high-throughput manufacturing. For example, scientists have yet to find a way to produce nanotubes of consistent sizes and electronic properties, which is key to achieving sufficient control for device applications. It is also difficult to integrate CNT into electronic devices using processes suitable for high-volume production. And there is high electrical resistance that produces heating and energy loss at junctions between CNTs and the metal wires connecting them.

Their use in next-wave microchips is among the most promising short-term applications for graphene. When rolled into CNTs or formed into ribbons or patterned planes, graphene is a terrific platform for electronics. Electrons move quickly and suffer very little energy dissipation even at room temperature. In fact, they act almost like massless particles. Making smooth interconnections between separated devices on a chip might be easier with graphene, and scientists hope to be able to further exploit the material’s unusual quantum effects.

“Nanotubes are simply graphene than has been rolled into a cylindrical shape,” says Georgia Tech's Walt de Heer. “Using narrow ribbons of graphene, we can get all the properties of nanotubes because those properties are due to the graphene and the confinement of the electrons, not the nanotube structure.” The width of the ribbon controls the material’s band-gap. Other structures, such as sensing molecules, could be attached to the edges of the ribbons, which are normally passivated by hydrogen atoms. The ribbon width confines the electrons in a quantum effect similar to that seen in CNTs.

According to de Heer, graphene will provide a more controllable platform for integrated electronics than is possible with CNTs since graphene structures can be fabricated as large wafers using existing lithographic techniques. Continuous graphene circuitry can be produced using standard microelectronic processing techniques, which gives scientists a road map for high-volume graphene electronics manufacturing. “There is a huge advantage to making a system out of one continuous material, compared to having different materials with different interfaces–and large resistances to cause heating at the contacts,” he said.

Single sheets of graphene were only isolated in 2004 by a group of researchers led by Andre Geim of the University of Manchester, sparking a wave of related investigation into the material. De Heer’s team starts with a wafer of silicon carbide, and then heats the wafer in a high vacuum to drive silicon atoms from the surface. What’s left is a thin continuous later of graphene. Next, they spin-coat onto the surface a photo-resist material and pattern the surface using optical lithography or electron-beam lithography, followed by conventional etching processes to remove unwanted graphene.

De Heer’s team has managed to create feature sizes as small as 80 nm–well on the way towards their goal of 10 nm–using electron beam lithography. Electrons move with very little scattering through the resulting graphene circuitry. The researchers have also shown electronic coherence at near room temperature, as well as evidence of quantum interference effects. They expect to see ballistic transport when they make structures small enough.

From a fundamental research perspective, graphene is equally rich in potential. For instance, it exhibits effects previously thought to occur only in the plasmas surrounding very dense neutron stars. Also, in graphene, electron velocity is independent of energy, so the electrons move as if they were light waves–they act like massless particles, even though the material contains particles known as massive chiral fermions, and particle theory has previously maintained that any particle with chirality must have mass.

This extraordinary property was explored further in November 2005 experiments making use of the quantum Hall effect (QHE), in which electrons, confined to a plane and subjected to high magnetic fields, execute only prescribed quantum trajectories. The experiments were conducted by Geim’s group, and by a team at Columbia University led by Philip Kim.

The QHE studies also revealed that when an electron completes a full circular trajectory in the imposed magnetic field, the phase of its wave function is shifted by 180 degrees. This is a modification known as “Berry’s phase,” and it serves to reduce electron energy loss. In a new twist to the story, Geim reported that he’s observed a new version of QHE while studying the effect in graphene bilayers, resulting in a doubled Berry’s phase of 360 degrees. This translates into even less energy loss than previously reported.

Geim compared his results to certain cosmologies in which multiple universes can co-exist, each with its own set of physical constants; in graphene, he said, where electrons move in a light-like way, with a fast speed (yet still somewhat less than the speed of light in a vacuum), the parameter which sets the scale of the electromagnetic force–that is, the fine structure constant–has a higher value of 2.0 rather than the customary 1/137.

The next step is to learn more about the fundamental physics of graphene, rather than focusing on potential applications. For example, de Heer reported that a plot of resistance versus an applied magnetic field had a fractal shape. He admitted that he can’t yet explain this unusual finding.

As for the applications, he said that on an all-graphene chip, linking components with the usual metallic interconnects (which tends to disrupt quantum relations) would not be necessary. So the wave nature of electrons could be more fully exploited for quantum-information purposes. Thus far de Heer’s group has attempted to build circuitry in this way, and has even made a few rudimentary graphene structures, including a graphene planar field-effect transistor. They have also built a working quantum interference device, which would be useful in manipulating electronic waves.

Meanwhile, research in CNTs marches on. A March 23 paper in Science by IBM researchers reported that they have succeeded in fashioning an electronic circuit around a single CNT molecule, obtaining switching frequencies of 52 MHz, roughly equivalent to Intel’s old 486 microprocessor chips. The approach could be used to simplify the manufacture of molecular electronic circuits.

APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.

Editor: Alan Chodos
Associate Editor: Jennifer Ouellette
Staff Writer: Ernie Tretkoff