Things to come?

Each year, the editors of Technology Review, published by MIT, take a look at what they see as important emerging technologies for readers to watch for in the future. As in the past, the editors have picked 10 potential breakthroughs that could have significant impact on our lives and the world around us. The list of hopefuls includes a method to produce cellulosic biofuels more efficiently and cheaper to lessen our dependence on fossil fuels, the creation of maps of the neural circuits of our minds to better understand how the brain functions, the development of atomic magnetometers for use in portable MRI machines, and the commercialization of wireless power.

[ILLUSTRATION OMITTED]

One technology that looks very promising is the use of a form of carbon to fabricate ultra high-speed transistors. These transistors could operate in the terahertz region or possibly beyond, according to researcher Walter de Heer, a physics professor at Georgia Tech. The transistors are made of graphene, a material found in common pencil lead.

Graphene presents a very low resistance to the flow of electrons. As stated in the "10 Emerging Technologies, 2008" article in the April issue, "... electrons move through graphene with almost no resistance, generating little heat. What's more, graphene is itself a good thermal conductor, allowing heat to dissipate quickly." These capabilities allow the graphene transistors to operate at much faster speeds than silicon-based devices.

Carbon nanotubes also show promise in this area, but they essentially are miniscule cylinders made from sheets of graphene. However, there are some drawbacks with nanotubes, according to the article. For complex circuits, the selection and placement of the nanotubes are very important, and the methodology has not been fully developed. Also, working with graphene is much easier.

Graphene also can be fabricated in much the same way as today's silicon-based devices. I would assume that some or much of the specialized equipment used for silicon could be modified slightly, if necessary, to produce graphene devices.

As you might expect, all is not smooth sailing with graphene. Unlike semiconductor devices, graphene devices can't be completely turned off. This would limit their applicability to only a few technologies such as imaging and communications. The conductivity of graphene can be changed, but it can't be switched off.

However, de Heer has modeled a technique using narrow graphene ribbons that would enable it to have characteristics similar to the switching properties of silicon. Other techniques under consideration include "chemically modifying graphene and putting a layer of graphene on top of certain other substrates," as reported in the article. It's noteworthy that HP, Intel, and IBM are looking into possible applications of graphene at their respective companies.

Another technology picked by the Technology Review editors to watch is the dumbing down of certain semiconductor devices to save valuable system power. A professor of computing at Rice University has developed a concept called probabilistic semiconductor technology that would lower supply voltages on certain parts of chips, allowing them to potentially malfunction some of the time. In this condition, the devices would not meet all of the specs but be perfectly acceptable for certain applications where precision is not needed such as with cell phones and music players. Using this technology, the professor expects a tenfold increase in battery life.

Stay tuned. It will be fun to see how many of these technologies actually come to pass.

Paul Milo

Editorial Director

pmilo@evaluationengineering.com