Out of the Blue: Midrange Perspectives

  • Smaller Small Medium Big Bigger
  • Default Helvetica Segoe Georgia Times

More than two decades ago, a shrunken, molecular Hollywood submarine, with a full complement of miniaturized scientific doodads (including Raquel Welch, cast as an inviting target for clinging antibodies), was injected into the bloodstream of a dying man with an inoperable blood clot. The movie's premise was that the sub's aortic wanderings would eventually bring it to the site of the clot where the mini-medical team would perform the first true in vivo surgery, saving the day, the patient and, of course, the breathy Ms. Welch from a near-rapacious attack by those wicked antibodies.

I thought I had permanently jettisoned that memory to the bad-movie bit-bucket, when an article by science and technology writer Lance Frazer revived it by posing an engaging prospect: the near-term possibility of building computers with molecular components.

The incredible shrinking AS/400 could well benefit. Since its introduction in 1988, the AS/400 has gone through an impressive process of compression and miniaturization, and IBM's latest announcements clearly demonstrate that the trend continues. Consider the size of the box itself: most models are now not much bigger than a suitcase while their corpulent predecessors, the S/36 and S/38, resembled old-fashioned chest-freezers.

In 1988, the AS/400 used 1MB memory technology. Five years later, 16MB have been squeezed into the same space. With improved mounting technology, memory capacities doubled again. IBM's 256MB memory board is the latest advancement in memory compression.

DASD density has undergone a similar contraction. Five years ago, a single rack could hold up to 4.8GB of DASD. Today, 94.4GB spin in the same space. How much further can miniaturization go? The answers are the purview of a burgeoning science called nanotechnology.

Nanotechnology is the science of microminiaturization, and it could produce marketable molecular components within a generation. That's the good news. The bad news is that without a contraption called a scanning tunneling microscope- a device developed by two Nobel Prize-winning IBM scientists-you won't be able to see them.

Nanotechnology seeks to assemble and direct individual atoms. It views matter, and the biochemical processes of living cells, as a petri dishful of molecular robot parts which can be manipulated and programmed to do, well, just about anything. Frazer insists that scientists will one day build "gears, wheels and motors smaller than dust particles...made atom by atom.

"The basic task of nanotechnology," says Frazer, "is to make single atoms perform a desired task, rather than relying on the properties exhibited by atoms in groups." The possibilities are so great that universities, governments and business research laboratories are investing billions to be the first to make those molecules sit up and roll over on command.

Dr. Eric Drexler, Stanford University lecturer, sees nanotechnology as the "industrial base of the 21st century." Whether IBM shares that vision is not clear. But it is involved with extensive experimentation in related fields at its three research facilities; the T.J. Watson Research Center in Yorktown Heights, New York; the Almaden lab in San Jose, California; and a European research facility in Zurich. There are presently some 3,000 researchers employed by IBM, and that number includes some of the most respected scientific minds in the nation.

What impact could nanotechnology have on the AS/400? Though it may be a while until your upgrade is delivered in a matchbox, Frazer foresees "molecule-sized computer chips with millions of times the speed and capacity of present-day devices." Or how about the ability to store a single bit of data on a single atom (in the AS/400 it now takes about a billion atoms to store a single bit). Or the ability to store data and images in three dimensions.

This last innovation (although not nanotechnology in the strictest sense) already exists and is called "volumetric memory." It was pioneered by Robert Birge, director of Syracuse University's Center for Molecular Electronics. "Six memory cubes," says Birge, "each no more than 5 cubic centimeters in volume, could store the entire cataloged Library of Congress."

IBM is also experimenting with holographic storage systems. IBM scientists have developed a set of photorefractive polymer films that can record several bright, erasable images in the same spot. IBM hopes these polymers will lead to a new type of rapid, random-access storage in which huge amounts of information are stored as holograms in incredibly small volumes. Several billion bits of data-more than the entire bulk of the Encyclopedia Brittanica-shrunk onto a dollop of polymer film the width and thickness of a dime. For the AS/400, holographic storage would mean virtually limitless data-storage capacity in a space no larger than a cigarette pack.

Gerald Present, senior communications specialist at the Watson Center, says IBM's research is not typically product-oriented; but includes investigations into a variety of disciplines including solid-state physics, chemistry, optics and molecular electronics.

The door to the submolecular world of nanotechnology was first pried open at IBM's Almaden lab where, according to Michael Ross, staff communications specialist, scientists have been studying how atoms behave on different surfaces. Their recent breakthroughs will likely provide the foundation for future computer technologies.

Two of those breakthroughs were pioneered by physicist Donald M. Eigler. In 1990, Dr. Eigler learned how to position individual xenon atoms on a nickel surface. He placed them in a pattern forming the letters "IBM." A small feat only in the literal sense: the xenon letters comprising the IBM logo were about 500,000 times smaller than those on this page; each atom separated by only about 50 billionths of an inch. Eigler then built the first-ever atomic cluster, one atom at a time, constructing a chain of seven xenon atoms.

His next breakthrough was the creation of the first "single atom switch." Eigler was able to move a single xenon atom back and forth across a gap between two electrodes. Momentously, he was also able to measure a change in the electrical current flowing between the electrodes based on the position of the atom. Conceptually, then, Eigler had created an atomic on/off switch; and switches, of course, are used as the fundamental logic elements of computers and storage devices.

But it is the potential of nanotechnology to combine living and nonliving matter into functional devices that strains the scientific envelope. Frazer reports that "the biochemical processes of life are [being] harnessed and altered to serve the needs of science." For example, "light-sensitive proteins found in salt marshes are now being made into biomolecular computer components." That seems harmless enough, but what of the future? Is the silicon chip slowly being replaced by the Frankenchip? Perhaps.

The most bizarre nanotechnological prospect is the potential of preserving human consciousness beyond death. The theory, courtesy of Stuart Hameroff of the University of Arizona, goes like this: each cell in the brain is essentially a very sophisticated computer with its own computing mechanism. Hameroff believes that this mechanism is comprised of protein polymers which have the ability to respond to external stimuli. What if you could "interface" with these mechanisms and create what Hameroff calls a "biocomputational system patterned after the brain itself?" Hameroff believes you could then "download consciousness" and preserve it.

Nanotechnology also has vast medical potential, similar in concept to Hollywood's rendering, if lacking the incredulity of intravenous submarine warfare. It may be possible, according to Frazer, to program "molecular motors" such as those that "move the flagellum of a humble bacteria, [to] power man-made microscopic structures for use in medical delivery systems." Medication could then be applied directly, and exclusively, to infected cells. Who knows, perhaps they'll even genetically engineer a Rambo-molecule that identifies and kills cells responsible for bad acting. Hold on Raquel, help is on the way.

Victor Rozek has 17 years of experience in the data processing industry, including seven years with IBM in Operations Management and Systems Engineering.