IBM has created graphene transistors that leave silicon ones in the dust. The prototype devices, made from atom-thick sheets of carbon, operate at 100 gigahertz–meaning they can switch on and off 100 billion times each second, about 10 times as fast as the speediest silicon transistors.
The transistors were created using processes that are compatible with existing semiconductor manufacturing, and experts say they could be scaled up to produce transistors for high-performance imaging, radar, and communications devices within the next few years, and for zippy computer processors in a decade or so.
Researchers have previously made graphene transistors using laborious mechanical methods, for example by flaking off sheets of graphene from graphite; the fastest transistors made this way have reached speeds of up to 26 gigahertz. Transistors made using similar methods have not equaled these speeds.
Growing transistors on a wafer not only leads to better performance, it’s also more commercially feasible, says Phaedon Avouris, leader of the nanoscale science and technology group at the IBM Watson Research Center in Ossining, NY where the work was carried out.
Ultimately, graphene has the potential to replace silicon in high-speed computer processors. As computers get faster each year, silicon is getting closer and closer to its physical limits, and graphene provides a promising potential replacement because electrons move through the material much faster than they do through silicon. “Even without optimizing the design, these transistors are already 2.5 times better than silicon,” says Yu-Ming Lin, another researcher at IBM Watson who collaborated with Avouris.
Other researchers have made very fast transistors using expensive semiconductor materials such as indium phosphide, but these devices only operate at low temperatures. In theory, graphene has the material properties needed to let transistors run at terahertz speeds at room temperature.
The IBM researchers grew the graphene on the surface of a two-inch silicon-carbide wafer. The process starts when they heat the wafer until the silicon evaporates, leaving behind a thin layer of carbon, known as epitaxial graphene. This technique has been used to make transistors before, but the IBM team improved the process by using better materials for the other parts of the transistor, in particular the insulator.
“Graphene’s properties are very sensitive to its environment,” says Lin. This is why the IBM group focused on designing a new insulating layer–the part of the transistor that prevents short circuits. They found that adding a thin layer of a polymer between the dielectric and the graphene improved performance. The work is described this week in the journal Science.
Walter de Heer, a professor of physics at Georgia Tech in Atlanta who pioneered methods used to work with epitaxial graphene, says the IBM device is a milestone because of its speed and because it was made using practical fabrication techniques. “This is not pie-in-the-sky stuff, this is real,” he says. “This development is really going to turn into a communications device not too long from now.”
“One can apply the same processing technologies to get much closer to a product,” says Avouris. Last year, the same IBM group, and an independent group at HRL Laboratories in Malibu, CA, both made 10 gigahertz graphene transistors using an involved method called mechanical exfoliation. This process involves peeling away layers from a small piece of graphite until a single, atom-thick sheet remains, then setting that down on a substrate and carving it to form a transistor. The problem with this approach is that it compromises graphene’s electrical properties and is not commercially scalable, says Avouris.
The first applications of graphene transistors will likely be as switches and amplifiers in analog military electronics. Indeed, the IBM group’s work is supported in part by the Defense Advanced Research Projects Agency. But the researchers say it will be years before the company begins commercial development on carbon electronics.
De Heer notes that the IBM devices don’t yet realize graphene’s full potential. By carefully controlling the growing conditions, his group has made graphene that conducts electrons 10 times faster than the material used by the IBM team. This higher-quality graphene could, in theory, be used to make transistors that reach terahertz speeds, though de Heer says many things could go wrong during scale-up.
Avouris says the IBM team will work to improve its transistors’ speed by miniaturizing them. The ones it has made so far are 240 nanometers long, which is relatively large–silicon electronic components are down to about 20 nanometers. Avouris also believes that their performance could be improved by making the insulating layer thinner. “The next step is to try and integrate these transistors into a truly operational circuit,” he says.