At The Forefront Of HP Research

Williams' group has been researching nanoscale electronics since 1996 and holds about a dozen patents in the field, mostly around a nanoscale circuit called a "crossbar latch." If the technology works as envisioned, the industry may one day be able to literally print out computers less expensively than with today's photolithography.

Meyya Meyyappan, director of the center for nanotechnology at NASA's Ames Research Center, says HP's approach, which envisages an entire molecular computer architecture that sidesteps inevitable defects in nanotechnology manufacturing, looks more promising than other approaches to replacing silicon chips, such as carbon nanotubes. "That's where these guys are smarter than everybody else," Meyyappan says. "They constantly talk about these other ideas that will give the industry the quantum leap it's looking for."

Colleagues describe Williams, who wears a ponytail and lights up when he talks about his newly published research and making beer runs to Mexico at age 13, as brilliant, ambitious, and a no-B.S. scientist. "He's one of the few people in the molecular electronics industry who never says, 'I'm going to solve the world's problems and I'm coming around the corner with the best thing since sliced bread,'" says Robert Curl, a university professor at Rice, where he was Williams' mentor. Curl shared the chemistry Nobel Prize in 1996 for the discovery of Carbon 60 molecules, known as "buckyballs," which laid the groundwork for the nanotech industry.

Before Williams arrived in 1995, HP didn't even have a basic research operation. Two years before he died in 1996, co-founder Packard appointed a committee to form one, at a time when AT&T, IBM, and others were cutting back on basic science. To head the lab, they found Williams, a chemistry professor at the University of California, Los Angeles, who'd come recommended by top academics. A 1994 earthquake had destroyed Williams' lab, and he was looking for something new.

Before UCLA, Williams worked briefly for Bell Labs and earned a Ph.D. in chemistry from UC Berkeley at the recommendation of Curl, who'd spotted Williams as a freshman at Rice in 1970 and let him into his lab. "Everything I am as a scientist I owe to [Curl]," Williams says.

Today, HP Labs' budget is a modest 5% of the company's total $3.5 billion R&D spending, or about $175 million. But former CEO Carly Fiorina, fired by HP's board in February, was an advocate of the labs, and quantum science research in particular. And Williams says HP Labs director Dick Lampman, executive VP Vyomesh "VJ" Joshi, and board member Jay Keyworth are all fans of the group's work, too.

In addition to the long-term goal of building computers that compute with nanoscale wires and molecules, the quantum group's work has potential applications for HP's $24 billion-a-year printing business. Transferring molecular memory technology to the chip that goes inside inkjet printer heads could make them better able to stand up to high voltages, currents, heat, and chemical properties, Williams says.

"You have to have a concrete thing you're shooting for," Williams says. "I and the people in my group are probably the least qualified to talk about commercialization. But it's always fun to speculate. In principle, you could have a simple processor that could fit on the end of a strand of hair. That's within the realms of the physically possible."