ILLUSTRATION BY JAMES SCHLESINGER

Basic industrial scientific research in the United States has gone the way of the rotary telephone. Now that the likes of Bell Labs and IBM have hung up most of their fundamental research programs, someone else has to pick up the slack if the U.S. is to compete in the world, argues David Awschalom of the University of California, Santa Barbara (UCSB). So, Awschalom had a few discussions and made a few calls.

The professor of physics, electrical, and computer engineering needed little persuasion to convince two other California universities, two U.S. defense agencies, national labs, and 10 industrial partners to establish the Center for Nanoscience Innovation for Defense (CNID) at the UCSB campus. The center's broad aim is to rapidly transfer innovations from basic research in nanoscience to the defense and private sectors.

DARPA and Defense Microelectronics Activity (DMEA) are the two U.S. agencies sponsoring CNID. UCLA and UC Riverside have joined UCSB and Los Alamos National Laboratory among the federal research institutions sharing $13.5 million in federal funds for CNID. The 10 industrial partners are Boeing, DuPont, Hewlett-Packard, Hughes Research, Motorola, NanoSys, Northrop Grumman, Rockwell Scientific, Raytheon, and TRW. With a second round of funding, the complete cost of the project is expected to surpass $20 million over three years. Two buildings, one with a cleanroom, will be constructed on the UCSB campus.

DARPA was already on board, since the defense experts share Awschalom's viewpoint. "DARPA deserves enormous credit for having the courage and the inspiration to make this happen. They have forward-looking program managers who are also outstanding scientists who I think are deeply concerned about the level of national science, [and] technology, and science education, and about how to attract the best students."

In an interview at his office overlooking the Santa Barbara campus, Awschalom spoke in measured professorial tones that could not disguise his enthusiasm for CNID and the great need to compensate for the decline in basic science.

"Look at the research activities of the major industrial labs now in the United States compared with even five or 10 years ago, when each of the laboratories—Bell Laboratories, IBM, Xerox PARC, of course, and Hewlett-Packard—had wonderful fundamental science programs incorporating hundreds of chemists, physicists, and materials scientists working on the fundamental issues that ultimately enable new technologies. With changes in the economy and changes in technology, many of these programs have disappeared."

Awschalom describes a golden age of research science. "For many of us doing fundamental science, when you look at the plethora of Nobel Prizes that have come from condensed-matter physics, for example, in materials science, a significant majority has come from these industrial-laboratory programs, whether it's high-temperature superconductivity or the invention of the scanning tunneling microscope."

POSITIVE SPIN: David Awschalom says spintronics is one of several promising research areas. PHOTO COURTESY OF UCSB

One of the main reasons for such success is the multidisciplinary nature of basic scientific endeavors, the physics professor maintains. It's an approach that CNID hopes to replicate, putting top graduate students from a variety of disciplines in contact with a wide range of experts in the hope that sparks fly.

Two years ago, UCSB and UCLA formed the California NanoSystems Institute. With this initiative, the state matches every $2 in private support with $1 in funds, up to $100 million. CNID will qualify for the state matching funds. The center will act as a conduit, Awschalom says, for industry to recruit highly trained students in nanoscale science and engineering while giving students access to private-sector R&D. In addition to the commercial sector, the transfer of knowledge is expected to ultimately benefit national defense.

Stu Wolf, a DARPA program manager, says the project could broaden its focus beyond the collaboration of the initial partners. "The cost of establishing a first-rate research infrastructure is beyond the reach of many institutions that have excellent researchers," he emphasizes.

"The idea of the CNID program is to create a structure where students, when they get their degrees, can not only learn from colleagues in different disciplines, it's also a portal for industry," Awschalom notes.

He says the center will offer "one-stop shopping" for those seeking solutions to technical problems. "Here, you can go to one place and say, 'I have a problem. I have defects moving under this chip, and I don't know how to get rid of them.' You can come here, and we will try to help with good students, good technology, good infrastructure, and we'll see what we can do."

CNID will concentrate on four areas: spintronics and quantum information processing, nanoscale electronics in semiconductors and molecules, nanophotonics for communication and computation, and nanomechanical sensors and devices.

Awschalom is the director of UCSB's Center for Spintronics and Quantum Comuptation Processing. He oversees DARPA-funded research designed to understand and control electron spin in semiconductor materials. One fascinating application "we're very keen on" involves taking "three different areas of technology—storage, logic, and communication," he explains. "Right now, you use semiconductors for logic on the computer. You might use...a magnetic disk—a 200-gigabyte disk on your Dell computer for storage and then photons or optical fibers to transmit the information. What if all of those [functions] could be [put] into one multifunctional electronic chip that combines logic, storage, and communication?

"Why is that interesting? Well, if you talk to Intel and ask, 'Why don't we have a 30 gigahertz chip, their first response is that it will just melt," Awschalom adds, laughing. "And it will melt because a large part of the chip is interconnect wiring. Suppose, though, you've added a chip that had almost no interconnect because all of the interfacing was done within the same device—the storage, the logic, the communication—so there might be just a fraction of the wiring now. Some of that would have a big impact on technology."

A more radical proposition is based on quantum mechanics and involves using the spin of the electron, he says. "If you take several of these electrons and mix them in the right way, you can entangle these different quantum mechanical states and effectively do extremely complex, traditionally high-speed calculations by designing nanometer-scale systems."

The center director emphasizes that many "physics obstacles" must be overcome in order to realize this application, "which is why it's an active research area. But the potential payoff is extraordinary." Cryptography is one use that holds exciting potential. Research groups around the country have prototype programs in place to explore the fact that electrons recombine in semiconductors to produce photons, Awschalom notes.

"Photons can have polarization," he explains. "That's a natural way to think about propagating information securely. The basic tenet of quantum mechanics is that the act of observing information can change it. So, if you want something that can prevent eavesdropping and you send quantum states down some channel and you try to read it, you change it. The person on the other end knows for a fact somebody is trying to intercept [the information]. It's a prototype for ultimately doing really 100% secure information processing."

Awschalom says cryptography will be one of the first uses of quantum information processing. "There are so many reasons to want that: for currency exchange, for banking, and, of course, for defense applications. But even in today's marketplace you want to use your ATM card securely. You want to pay with your credit card securely."

CNID research also may ultimately find a way to help chipmakers with yields and defect detection in the nanometer realm, Awschalom believes. One potential use is high-resolution imaging techniques for examining defect densities and device structures. Finding and removing defects "moving on and off contacts and gates" because of electromigration are obvious major concerns. The development of new defect-free materials at the nanometer scale is another project that researchers are pursuing.

DARPA's role is limited to "agreeing that areas we've chosen to investigate are interesting." Awschalom emphasizes that the defense agency, which will monitor the progress of the research, does not absolutely require the work to have dual-use applications. "I think if you did, you'd pretty much guarantee losing the best students. The exciting thing is you need a certain amount of flexibility and randomness thrown in. But you want some people around... who, when you do make a random discovery, can make the connection and say, 'Oh, that random discovery can have this fantastic impact.'

"If you don't have that oversight, that discovery might be sitting in some journal of science for a decade before someone notices it. That's what used to happen in industrial labs. When there was a fundamental scientific discovery, some senior manager would look at it and say, 'Boy, that's wonderful science, but the person in the building next door to you has been looking for two years for something like that. You guys have to get together.' Without this type of connection I think it's going to be hard to rapidly move scientific discoveries to the marketplace."

Awschalom calls the CNID project an experiment "we hope will work well." As a cautionary warning, the director points out that other regions have targeted basic science research. "In the areas that we're focusing on in this new center, there's absolutely no doubt that other parts of the world are accelerating their programs. In materials sciences—in areas of semiconductors, and magnetism and electronics—Japan, for example, has launched numerous multiuniversity programs that are 10-year ventures in these areas. The comparable number of programs in the United States are substantially fewer.

"Oddly enough, many of the leaders of the national programs in Japan were trained as postdocs at Bell Labs and IBM, and went back to Japan. I give them enormous credit for doing this, but I think that most of condensed-matter science is based on materials in the United States. If you don't have the materials, you're losing the platform for research."