Investor's Business Daily, August 1st, 2007
If a comedy is ever made about Jeff Welser's job, it might be called, "Honey, I Shrunk the Computer Chips."
Welser is a research director for IBM IBM. He's been working on next-generation chip designs at Big Blue for nearly four years.
Last year, IBM loaned him to the chip industry's main research group, Semiconductor Research Corp. He heads a new SRC unit that works on ways to use nanotechnology in chipmaking.
It's called the Nanoelectronics Research Initiative. Nanotechnology is the science of building electronic circuits and devices from single atoms and molecules, and it could be key to advancing semiconductors and a host of other products.
Nanotech is already being used to make such things as stain-resistant pants and light-but-strong tennis racket frames.
Analysts say NRI is critical if the U.S. is to maintain its global technology lead. The White House agrees. President Bush on July 17 awarded the SRC the National Medal of Technology, the nation's highest high-tech honor, in part for its nanotech effort.
Welser recently spoke with IBD about the benefits nanotech is likely to bring to chips, and some of the hurdles researchers must first overcome.
IBD: How did the NRI come about?
Welser: NRI's mission is to find a semiconductor device that can replace today's CMOS (complementary metal oxide silicon) transistor (the most common type of transistors used to make chips) in the year 2020 time frame. That's a one-sentence mission statement.
NRI is part of the SRC, which was sanctioned by the Semiconductor Industry Association (the biggest chip trade group) to fund university research to further the semiconductor design road maps.
IBD: What's the NRI's role in nanotech research?
Welser: We're funding 25 universities, where they do the research. And we have organized most of the research into three centers. One is the new WIM, the Western Institute of Nanoelectronics. Scientists from four institutions are participating in WIN. They are the University of California at Los Angeles, UC-Santa Barbara, UC-Berkeley and Stanford.
We also are funding individual projects, joining with the National Science Foundation, a federal agency that funds research across the country.
IBD: Is this only for U.S. universities and companies?
Welser: It's U.S. only. We're very much gearing this to help American competitiveness.
IBD: Why can't you just keep shrinking CMOS transistors smaller and smaller, as in the past?
Welser: We're reaching some real physical limits here. Today gate oxide (a tiny chip feature) is at three atoms thick in some advanced circuits. You can't take away many atoms -- the process of scaling or shrinking -- before you hit zero.
And the power these consume is becoming a limitation as well. That heat is getting so high it's difficult to efficiently cool them, particularly in mobile devices.
IBD: What things are you looking at that might replace today's transistors?
Welser: There is research on spintronics (spin-based electronics) transistors. They use the spin of an electron to indicate a one or a zero (the digital language). A spin up might be a one and a spin down a zero.
This is very much quantum mechanics. We can't even measure the spin of electrons yet. It will take years to answer. That's the reason we're starting this process now.
Our goal in the next three to four years is to narrow that to a few promising options, to demonstrate a real device and integrate it into a basic circuit.
IBD: What are the main challenges nanotech researchers face?
Welser: The first is communication between the basic science folks that understand the physics and the engineers like me that understand transistors. We have a completely different set of words, different languages.
Scientists don't necessarily understand how something will work in my cell phone, and I don't understand how things spin -- the physics of it.
A second challenge is testing and characterizing it. How do you measure the spin of a single electron?
IBD: Is there serious competition from other nations in the nanotech arena?
Welser: Absolutely. There are very active groups in Europe and Japan for sure. Some are in parts of Asia, in Korea most notably.
I think we have a good lead, but I don't think we can be complacent.
IBD: A nanotech researcher recently said the chip industry is late to the use of nanotech, that clothing and cosmetics makers are already using nanotech particles. Do you agree?
Welser: He has a point. I don't think we're late to the party. But if you look at how you can apply nanotech, putting in nanoparticles is something you can readily exploit quickly.
The challenge is, we need to create active nano structures that do switching or storing of information. There are billions of these on a chip.
