This Week in NW
Switches and chips built atom by atom could lead to smaller, faster, cheaper network products.
BOSTON - Nanotechnology could lead to an automated world of self-restoring architectures, reconfigurable chips, sensor-based wireless networks and desktops that load themselves and bill per [client] usage. That's the vision of Charles Ostman, senior fellow at the Institute for Global Futures. Ostman's view is that IT is in a state of transition, "a move away from the human operations of systems to optimized localized organic nodes."
The word "futuristic" doesn't begin to describe the topics that were under discussion at the recent Nanotech Planet Conference and Expo. Casual conversations drifted between tales of Star Trek prototypes existing today to cell replication and extraterrestrial life forms.
But nanotechnology, or at least the scientific research leading to it, is very real. Conference speaker M.C. Roco says the federal government, which began its National Nanotechnology Initiative funding in 1996, is pouring $570 million into research and development for eight federal departments and agencies this year, up 35% from the $422 million in 2001. Roco chairs the National Science and Technology Council's Subcommittee on Nanoscale Science, Engineering and Technology.
What is nanotechnology
Built on a foundation of physics, chemistry, biology and computer science, nanotechnology is the manipulation of matter at the atomic scale, where one nanometer is one billionth of a meter.
Nanometric building blocks can be used to develop a molecule-sized electronic switch or a miniaturized version of the entire logic system of a computer.
Inside the nanofacturing process
Examples of nanometric sizes
Nanotechnologists at the conference predicted that within 10 years, terabytes of memory cached on a chip will hit the consumer market in toys, games and cameras, marking a transitional step from so-called silicon electronics to molecular electronics.
Wait 50 years, and hybrid molecular electronics, a combination of molecular and silicon components, will replace silicon-based switched networks, leading to a world where the processes done by today's computers will run on a device that consumes far less power and fits in the palm of your hand.
R. Stanley Williams, Hewlett-Packard Labs fellow and director of Quantum Science Research, advises healthy skepticism. "To get a realistic time frame, I use the 'pi rule,'" Williams says. "Take the academic's best estimate of how long something will take and multiply it by pi."
Williams says that while nanoscience exists, there's very little nanotechnology. He divides the topic into passive and active nanotechnology structures. Passive nanostructures, like sensors, accomplish a single, predefined task. Active nanoapplications transfer information between nanostructures.
He says the passive structures, such as those used in fireproof materials and wear-resistant tires, are available now. Active applications, requiring enormous research investments, are still quite a ways out.
Nanoscale approaches to computing offer hope that integrated circuits will continue to get faster and smaller, and will consume less energy. That will mean packing more circuitry onto a microchip less expensively than possible with silicon. It will mean molecular memories with a million times the storage density of today's advanced chips.
While today, assembling nanodevices one molecule at a time, precisely arranging and incorporating millions of molecules into one device, is painstaking, researchers say the next step is self-assembly, molecular designs that automatically arrange themselves into a desired pattern or device.
Williams says there is nothing to fear from this new technology. "Nanotechnology won't make us infinitely wealthy. And nanobots won't take over the world and kill us."
MIT Enterprise Technology review