New Graphene Discovery Could Finally Punch the Gas Pedal, Drive Faster CPUs

Few substances have excited the computer industry as much as graphene. Few substances have proven as maddening and difficult to work with as graphene. The collision of these two facts is why, 14 years after Andre Geim and Konstantin Novoselov characterized and isolated the substance, we’re still waiting to see it used in semiconductor manufacturing. A recent breakthrough, however, could finally change that.

There are, broadly speaking, two major problems with graphene. The first problem is the difficulty in producing it at scale. The second is its electrical conductivity. The latter might seem like an odd problem, given that graphene’s phenomenal electrical properties are the reason semiconductor manufacturers are interested in it in the first place. But graphene’s unique capabilities also make it difficult to stop the material from conducting electricity. Silicon has a band gap — an energy range where it doesn’t conduct electricity. Graphene, in its pure form, does not. While a handful of methods of producing a band gap in graphene have been found, none of them have been suitable for mass production. That may finally change, thanks to a team from the Catalan Institute of Nanoscience and Nanotechnology (ICN2), who have found a way to create a graphene bandgap that’s identical to silicon’s.

“What we show in our work is that it is possible to fabricate a graphene ‘like’ material, but with a gap that is very close to the one of silicon, Aitor Mugarza, a research professor and group leader at ICN2, told IEEE Spectrum. “In addition, by simply modifying the width of the graphene strips between the pores (the number of carbon atoms), this band gap can be controlled. The fabrication method is relatively simple and can be extended to wafer-scale growth.”

One key component of the work is that the advances were driven by bottom-up construction rather than top-down methods. It’s difficult to hit the nanometer scales of modern semiconductor manufacturing with the top-down method, but building the structure from the bottom is much easier to scale to mass manufacturing, according to Mugarza. The research team claims their technique scales to the atomic scale, with lateral dimensions “on the order of 1nm.” A video of the construction process is embedded below.

So, with this discovery, are we all set for graphene production? Not exactly. There’s still a long road between the present day and any chance of seeing graphene transistors. There are major questions around substrates, contacts, and mass manufacturing. If Intel, Samsung, TSMC, and GlobalFoundries can’t build graphene in sufficient quantities, we’ll never see it adopted for anything but the most esoteric projects.

But at the same time, this genuinely does seem to be a major step forward for mass graphene production and integration into logic. Without the ability to create an effective band gap, we were never going to see graphene in transistors at all. And of all the methods of creating a band gap we’ve discovered, the bottom-up method seems to have the best chance of working at scale and creating the desired characteristics. That qualifies as a genuine breakthrough in our book, even if it’s not large enough to clear the runway entirely and prompt immediate commercialization.

Leave a Reply

Your email address will not be published. Required fields are marked *