One of the characteristics of clever science is to look at a new material from every which way. So it is with graphene. Graphene is a sheet of carbon atoms, in a layer one atom thick, arranged in the pattern of a honeycomb. It sounds simple, and is anything but. Its super-thinness in this precise configuration gives graphene amazing electrical and quantum mechanical properties – as well as being one of the strongest materials known. Scientists invent a new application for graphene just about every month. One of the most persistent applications is to make a semi-conductor out of it; it seems like graphene ought to make transistors and other electronic components, just like or even better than those made of silicon. But there’s a catch: Graphene doesn’t have a band gap.
Put simply, the band gap is a space between the layer of electrons bound to an atom’s nucleus, the valence layer, and a layer of electrons that are free to move about from atom to atom, the conduction layer. In many metals these two bands are on top of each other and electrons flow freely between them and between atoms – which makes metals a good conductor of electricity. In other materials, there is a large gap between the valence and conductive layers, which electrons cannot cross – making such material an insulator. Finally there are materials with a small band gap that given enough of a kick, the electrons can jump between the bands – these are called semiconductors. Most of modern electronics are built upon the band gap properties of semiconductor materials, principally silicon. Without a band gap, graphene doesn’t make a semiconductor, although there are many avenues of research to get around this deficiency.
While it’s true that a layer of graphene has no bandgap, it occurred to scientists at the National Institute of Standards (NIST, USA) that having two layers of graphene might have a space between them that could act like a band gap. As described in the journal Nature Physics [24 April 2011, paywalled, Microscopic polarization in bilayer graphene] two layers of graphene (a bilayer) placed on a third non-conducting layer (a substrate) produces bandgap-like effects. However…
There is a catch to this too: Electrons do not move across this bandgap as easily as, say, silicon. This would make graphene transistors slower than silicon – not a winning property. So the research reported in this paper is unfinished business. The researchers need to experiment with a wide variety of substrates under a variety of conditions to see if they can tweak the ‘artificial’ bandgap into greater efficiency. As far as it goes, the bilayer-on-a-substrate approach seems promising, but with a new material like graphene, there’s no guarantee of success beyond the laboratory.
SciTechStory has covered other approaches to graphene transistors, none of which have yet become commercially viable. They’re worth covering because of the promise of graphene to produce cheaper, faster, more durable semiconductor devices.
[SciTechStory post: Graphene gets spintronics]
[SciTechStory post: Working toward a triple-threat graphene transistor]
[SciTechStory post: Graphene transistors]