- The Washington Times - Thursday, November 14, 2002

From IBM Corp., something that may be a really big deal, if it can get the technology to work in the real world. Specifically, molecular-sized computer circuits that actually compute. Translated, this means seriously tiny computer parts, small enough that hundreds of billions of them could fit on the end of a finger.
OK, Fred, you say, that's cute, but why do I care?
The idea is that, if it works, the technology will have profound effects, but it isn't easy to say just what effects. We have never had anything like this.
We are used to having computers get lots faster real fast. Moore's Law, enunciated by Gordon Moore, one of the co-founders of Intel, says that the number of transistors that can be put on a chip doubles every two years. Depending on whom you talk to, you can get slightly different formulations to fit changing circumstances, but this one is close enough. It's why your computer is obsolete six months after you buy it.
For the increase in power to continue (and you do want your computer to obsolesce fast, don't you?) the little switches that make up computers-transistors actually have to get smaller. Smaller means faster for various reasons. For example, computers run so fast nowadays that the time for a signal to travel from one side of a chip to the other slows the machine down. Which is crazy, but it's true.
The first electronic computers used vacuum tubes about the size of light bulbs. A modern desktop computer made of tubes would be the size of something large, like Montana, and wouldn't work.
Later came separate transistors the size of aspirin tablets. These were better, but useless today. Now we have we have tens of millions of photo-engraved transistors on a chip the size of a fingernail. They keep getting smaller, so computers keep getting faster.
The problem is that the methods used to make them will, sooner or later, run into the laws of physics. Making the current technology smaller stops being hard and becomes impossible.
Now IBM has managed to make circuits that actually calculate by stringing together individual molecules, specifically of carbon monoxide on copper. A row of individual carbon monoxide molecules knock each other over like falling dominoes. ("Knock over" isn't the technical term, but we'll use it anyway.) The falling rows can be made to interact in ways that represent computer operations.
If you want small, molecular switches are just the thing. IBM says a circuit made this way is 1/260,000 the size of one made normally. One of the circuits the company made, and it worked, was a three-input sorter. You could put 190 billion of them on the eraser of a pencil. (You probably wouldn't want to, but they would fit.)
You might call it minimalist computing. Instead of using huge numbers of atoms in a bunch, you use the smallest number that can do the job.
Said Andreas Heinrich, a physicist at IBM's Almaden Research Center in San Jose, Calif. "The molecule cascade is not only a novel way to do computation, but it is also the first time all of the components necessary for nanoscale computation have been constructed, connected and then made to compute." That's slick.
It's also useless, practically speaking, just at the moment. There's no way to reset the circuit after you have used it. You do a calculation once, and then have to build another circuit. It's like having a light that can be turned on but not off again.
Further, the particular circuits were made and operated within 10 degrees of absolute zero, not much good in the average office. But the rascals worked, once anyway. And IBM's people are scurrying around trying to get them closer to practicality. If they do and I wouldn't put it past them the world will become a stranger place.


Copyright © 2018 The Washington Times, LLC. Click here for reprint permission.

The Washington Times Comment Policy

The Washington Times welcomes your comments on Spot.im, our third-party provider. Please read our Comment Policy before commenting.

 

Click to Read More and View Comments

Click to Hide