February 24, 2003 2:18 PM PST
Nanotech to pave way for micro-machines
Nanotechnology--the science of making devices with features measuring less than 100 nanometers (or one-ten-millionth of a meter)--will let companies make smaller and cheaper products than ever before, which in turn will lead to new markets, according to Albert Pisano, a professor of engineering and computer science at the University of California at Berkeley.
Communications, for instance, could be greatly affected, Pisano noted. Right now, radios on satellites have to be hardened against radiation, an expensive process. At Stanford University, researchers have shown how small, unhardened radios can transmit and receive messages while in orbit. Although these smaller radios are subject to harm from radiation, they cost far less, so many could be mounted on a spacecraft to compensate for burnouts.
A market for tiny transmitters could also emerge within five years, out of a need to better monitor power lines, freeways and bridges for potential failures.
The movement will also mark a major turning point in mechanical engineering because most of these tiny systems will deal directly with the physical world, like pumps do, rather than getting information from electrical impulses, like microprocessors.
"Most people forget that a radio is two-thirds filter and that filters are mechanical," Pisano said. "Sensors and computing and communications can all be heavily miniaturized."
Batteries, too, could improve through breakthroughs in mechanical engineering. Currently, a kilogram of lithium ion, the same material used in notebook batteries, provides about 400 watt-hours of energy. A kilogram of methane provides 15,000 watt-hours of energy. The difference is that lithium ion is electrically charged, while methane provides thermal energy--that is, heat.
"Can you use thermal energy on such a small scale?" Pisano asked. Apparently so. Some researchers have already developed rotary engines that measure only microns across.
Of course, getting to miniature nirvana won't be easy. In the early '90s, futurists predicted a booming market for MEMS (microelectromechanical systems) similar to the devices described above. Difficulties in mass manufacturing, packaging and other problems kept the market from taking flight.
Pisano, though, noted that new technologies take time. About 15 years passed between the development of the transistor in the late '40s to the true development of the electronics industry. Several challenges need to be overcome, but the progress with nanotechnology makes MEMS more promising.
The first must-have applications for the types of MEMS products Pisano described will appear in about five years, he predicted, because the demand exists. Gillette, for instance, has spent heavily in the last few months on diminutive radio-frequency transmitters.
A number of universities and private companies will be presenting papers on MEMS over the five-day conference here. Some of the major themes include microfluidics, or small machines that can test biological samples, and films or surfaces with sensory capabilities.
Other companies are presenting papers on progress in using chains of molecules to make processors or memory devices. Molecular computing will likely take much longer to develop, but early results are showing promise.
Molecular chips differ substantially from today's silicon processors, noted Phaedon Avouris, manager of nanoscience and nanotechnology at IBM's Watson Research Center. Carbon nanotubes exposed to the air, for instance, carry a positive charge. When hermetically sealed off, they carry a negative charge.
"The properties are not just with the nanotubes, but with the environment," Avouris said.
As a result, IBM is developing different coatings to control this behavior.