University Communications

It's the molecular-sized computer, however, that has captured the attention of the world, or at least The New York Times and other hot media (of Heath, Vanity Fair rhapsodized in its end-of-the-millennium 1999 Hall of Fame: "Like William Blake, he spies the world in a grain of sand"). The idea emerged after Heath was recruited by UCLA chemist Stan Williams to help Hewlett-Packard launch a basic research division. While at HP, Heath and Williams met Phil Keukes, an HP physicist, who introduced them to Teramac, an HP computer capable of astounding performance while being as riddled with defective circuits as Swiss cheese is with holes. This made Teramac exceedingly relevant to Heath's dreams of synthesizing computers from the bottom up. "If you're going to design a computer by chemical synthesis," says Heath, "you're inevitably going to have plenty of defects. It's going to be far from perfect. But Teramac had a quarter-of-a-million defects in it and it still did a trillion operations per second. And these were hardware defects, not software bugs. If a Pentium even has one, it's trash. And so the very fact that this worked suggested that we ought to take some time and learn about this machine."

Heath, Keukes and Williams, along with HP's Gregory Snider, spent the next two years studying Teramac and writing a paper for the prestigious journal Science on a computer architecture that would be resistant to defects and that could be synthesized using the techniques of nanotechnology. The essence of their plan was to grow an ordered crystal of redundant wires and switches, locate the defects and then, in effect, download the logical architecture of the computer in such a way that it wired itself around those defects into working circuits. The Science article came out in June 1998 and "made a splash," says Heath. "I think broken computers resonate with people, and here was one that was broken that actually worked."

Heath and his group, along with Professor of Chemistry Fraser Stoddart's group, then spent the next year trying to synthesize their thinking machine. While Heath had thought his quantum dots might be useful, that turned out not to be the case. Instead they have been working on a device built of tiny quantum wires and switches made from Stoddart's molecules that can be turned on and off through quantum mechanical processes. They published their first example of such a device last summer, which was when they graced the front page of The New York Times. "It was amazing," says Heath. "I thought we did something pretty significant, but I didn't think it was that significant."

Still, the 1999 version of the computer was only what Heath calls a demonstration of architecture. While they could set their molecular switches on or off, they could only switch them once. So Heath and his colleagues have been working again with Stoddart's and Professor of Chemistry Fred Wudl's '64, Ph.D. '67 groups to concoct molecular switches that can be switched repeatedly, of which they now have four. "You can switch two electronically," says Heath, "one chemically and one optically. But they're all based on very similar architectures. Molecules just do things like that."

<previous> <next>