"Opportunity is missed by most people because it is dressed in overalls and looks like work."

The above insight comes to us courtesy of Mark Twain, and was featured on the closing Power Point slide of Intel's Michael Mayberry, one of the featured speakers at the IPF session on emerging materials and devices beyond CMOS. Mayberry used it to drive home his main point: sometimes the most promising R&D in nanoelectronics isn't the more showy or exciting, but the more mundane and workmanlike.

Mayberry specifically compared the buzz over the potential for carbon nanotubes (CNTs) to revolutionize nanoelectronics with the prounounced lack of buzz over what he considers to be a more promising approach: developing nanomaterials that can be grown in place (via techniques like pattern-assisted self-assembly) on an existing substrate. Even though it's less sexy and exciting, and therefore generates far less "buzz," he believes that approach is much more likely to be easily integrated in future devices, which many of the session's speakers felt would probably be a hybrid CMOS/nano device -- at least initially. (Indeed, Mayberry is among the ranks of IT people who rightly point out that --if you want to get technical about it -- we're already doing nanoelectronics. Certain electronic components are in the nanorealm; for example, gate lengths currently measure 35 to 65 nanometers.)

That's because there are serious scale-up problems when it comes to manufacturing ICs out of CNTs. Take the example of copper nanowires, a promising component for future nanoelectronic devices. The manufacturing process consists of three steps: Surface Preparation, Deposit, and Patterning or Shaping. In contrast, CNTs require a much more complicated process: Grow, Purify, Deposit, Align, Integrate and Verify. Of those, achieving purity and exact alignment are the most difficult. It can be done in carefully controlled laboratory conditions at the small scale, but the techniques aren't nearly robust enough to scale up to the industrial levels needed to make CNTs truly viable for nanoelectronics in the near future. For that reason, Mayberry contends that the IT industry will tend to favor those applications with the fewest alignment requirements.

These and other challenges to scaling up CNT manufacture are well-known, and have been widely reported. Yet somehow those grim realities haven't done much to crush the excitement over CNTs' potential.

The Twain quote ended up resonating with me later in the session as well, during talks that focused primarily on the need for ever-more-precise metrology to characterize the unusual properties of materials at the nanoscale. There's a great deal of scientific ingenuity at work here, but it's decidedly nuts-and-bolts stuff: incremental improvements to existing surface microscopy techniques, for example, or the potential for extreme UV lithography, and occasionally a new emerging technique like chemical force microscopy. This type of work is so critical, particularly since alignment and control of materials at the atomic scale is what will ultimately make nanoelectronics a reality. It might not make for the most exciting headlines, but without dedicated folks willing to don the overalls and sweat over the tiniest details, innovation would be impossible. We tip our hats to them.