Let's Keep Nano Options Open

Although announcements about leaders such as IBM, TI, Intel, Micron and AMD calling for proposals to specifically fund nanoelectronics research centers seem to indicate that the semiconductor industry is looking beyond CMOS, appearances can be deceiving.
 
Unfortunately, these centers do not go far enough. Their nanoelectronics work is mostly focused on CMOS-related technologies and not much else. Most funding companies are considering using nanotubes as transistors, which means current nanotech research goals will probably take them to 2015 or 2020, but not much beyond. This is our own fault: We will do anything to retain CMOS, even grudgingly forcing it to become a nanotechnology platform.

We no longer rely on government funding for nanotechnology centers. Now industry, as represented by the Semiconductor Industry Association (SIA), has taken steps in the right direction, resulting in the National Nanoelectronics Initiative (NRI), initially funded half by NSF and half by the Semiconductor Research Corp. (SRC); there is now more industry funding for existing centers to work on nanoelectronics.

Ideally, nanoelectronics projects should be funded in a way similar to the NSF's, where selection is carried out by peer review. The process takes several months, and comments are provided to candidates. Even those who are unsuccessful get feedback, and everyone knows why they were funded or not.

Unlike the NSF, the SRC and SIA do not provide applicant feedback. If a proposal is declined, that is all that the candidate is told. SRC funding comes from industry, and each company has one or more representatives who vote on proposed projects. This sometimes leads to a situation in which a candidate who has access to board members could contact them about a proposal before the selection process. Some insiders indicate that it is not unusual for some board members to go into meetings having already decided what to fund.

This is not necessarily wrong. Although global in scope, our industry is small and technology leaders are known to all of us — they have often worked in academia as well as industry. It is not improper that those who provide the funding prefer that it be used to advance mature technology that will keep them in business for years with a minimum of disruption. However, this can mean that proposals made by candidates lacking this acquaintance may have more than an even chance of not getting funded. The downside is that some innovative projects might not get funded because they cannot avail themselves of this approval process.

It might be worthwhile to consider opening the doors wider, by allocating 30% — or even less — of the funding for novel projects proposed by candidates who might not be so close to our industry. Because current nanoelectronics research goals seem aimed at extending CMOS technology's life — often at the expense of potentially revolutionary new approaches — we cannot afford to be complacent about pursuing a goal we know will eventually become unattainable.

Nobody seriously suggests that projects that might take 15 or more years to come to fruition be funded at the expense of CMOS, but there should be more than a perfunctory financial nod in their direction. Considerable nanotech research is taking place, but most of it does not focus on nanoelectronics. Stan Williams, a senior HP fellow and director of Quantum Science Research at HP Labs, has pointed out that, short term, the biggest nanotech opportunities are not in the semiconductor industry. "Nano composites are being introduced that have combinations of properties that no naturally occurring material has ever had, like hardness and toughness. This is leading to all sorts of improved products, like car bumpers and drill bits, and as the cost of these materials falls, they'll find broader applications."¹ Williams added that true nanotechnology — using new properties enabled by the nanoscale — will not significantly impact our industry until the end of this decade or the beginning of the next.

Surely we can do better.