Graphene holds the promise of 10-times faster speed than silicon chips, plus the ability to be integrated with exiting semiconductor fabrication techniques. Unfortunately, graphene transistor channels need to be less than 10-nanometers wide in order to open up a bandgap suitable for digital circuitry, delaying its entry into the International Technology Roadmap for Semiconductors to beyond 2017, when sub-10 nanometer lithography becomes available.

Now professor Feng Wang at UC Berkeley claims to have demonstrated a technology that can electrically tune graphene's bandgap, enabling it to be used for digital transistors long before lithography hits sub-10 nanometer sizes.

"We have for the first time demonstrated that you can use an electric field to open or close a bandgap in graphene," Wang said. "There is no other material available today that can do this, only bilayer grapheme."

Electrically tunable bandgaps have been predicted for a double layer of graphene in the past, and others have reported evidence that the theory is sound. But Wang's group is the first group to report laboratory confirmation that the technique really works.

Wang used exfoliation to fabricate two parallel graphene monolayers atop each other, then attached gate electrodes to the top and bottom of the bilayers. Electrical connections for the source and drain were made along the edges of the bilayer sheets. By varying the gating voltages on the top and bottom gates independently, the team was able to demonstrate an electrically tunable bandgap that varied between zero (a metal) and 250 milli-electron volts (a semiconductor). That was only a fraction of the size of bandgaps in current semiconductors (germanium and silicon have bandgaps of 740- and 1,200-meV, respectively) but wide enough to fabricate digital circuitry.

The researchers speculated that a new kind of graphene gate array would be possible using the technique to dynamically reconfigure millions of gates, each with both top and bottom electrodes, by retuning their bandgaps on-the-fly.

"All you need is dual gates at all positions, then you could change any location to be either a metal or a semiconductor electrically," said Wang.

Wang, along with his post-doctoral fellow Yuanbo Zhang, his graduate student Tsung-Ta Tang and their UC Berkeley and Lawrence Berkeley National Laboratory colleagues, are next planning to demonstrate working circuits using the technique. The group is particularly interested in demonstrating optical emissions as high-energy electrons shed photons to jump the band-gap.