Mario Paniccia and his research team became overnight sensations in the high-tech world when they succeeded in building a super-fast silicon photonic chip that uses light waves instead of electricity to send data.

Chasing the speed of light

Endicott native Mario Paniccia '88 leads the drive for the world's fastest silicon photonic chip

As a kid growing up in Endicott, N.Y., in the 1970s and 1980s, Mario Paniccia always aimed to head out to Silicon Valley some day. He dreamed of settling there, starting his own high-tech firm and building a comfortable business.

Now, a couple of decades later, Paniccia is a shining star in Silicon Valley. The 37-year-old physicist and Binghamton University alumnus made headlines around the world earlier this year when his Intel research team unveiled the exciting development of a super-fast silicon photonic chip that could someday transform computing and communications by using light waves instead of electricity to send data.

Geared to switch light on and off like an optical shutter, this new type of chip can encode light beams passing through silicon at speeds up to 1 gigahertz (or 1 billion cycles a second), as much as 50 times faster than the previous silicon switching record. Intel scientists say the creation of this new silicon-based photonic device could lead to the development of ultra high-speed fiber-optic equipment at much lower prices than before, allowing computers to span different locations and beam high-definition video and images hundreds of times faster than they do now.

"With silicon, you significantly change the cost economics of the market," said Paniccia, who heads Intel's photonics technology laboratory in Santa Clara, Calif. "Our research goal is to ‘siliconize' photonics."

Paniccia, a solid-state physicist with master's and PhD degrees from Purdue University, came up with the idea for the breakthrough optical modulator after puzzling over the high cost of building photonic devices and systems. These components are so expensive now because they're typically made with such esoteric materials as indium phosphide, lithium niobate or gallium arsenide, not relatively cheap silicon.

"It just kind of evolved," he said. "I had a couple of ‘eureka' moments . . . I did a back-of-the-envelope calculation that this could work."

With backing from Leslie Vadasz, an original Intel employee and former top executive who's now a director emeritus of the company, Paniccia put together a small, select team several years ago to work on the top-secret project. Even though a number of his researchers didn't think silicon could be used to beam light at such blistering speeds, he galvanized them to at least try.

"Half the people who came onboard didn't believe it could be done," he said. "They came onboard because it was an interesting project . . . When I put the stake in the ground at 1 gigahertz, I almost had a mutiny."

Paniccia and his crew also faced many unique hurdles. For one thing, they couldn't use modeling, a tried-and-true scientific technique, to study their invention because they were attempting something totally new.

"The process didn't exist," he said. "We had to develop a process from scratch. There were no models."
But that didn't faze Paniccia and his talented team. "I kind of thrive on risks," he said. "If it's not a risky project, I'm not interested."

Toiling away in the photonics lab, the crack researchers persisted. By mid-summer 2003, after conducting their preliminary speed measurements, they realized they had hit paydirt. But Paniccia didn't realize how much paydirt they had hit until the prestigious scientific journal Nature agreed to publish their paper on the silicon modulator last fall.

"This is the first time Intel has ever published anything in Nature," he said.

"Half the people who came onboard didn't believe it could be done. . . . When I put the stake in the ground at 1 gigahertz, I almost had a mutiny."

"I had no idea it'd be that big a deal." In fact, the development was such a big deal that 169 newspapers, magazines, news services, trade journals, TV networks, Web newsletters and other media outlets quickly ran extensive stories about it, including The New York Times, the Wall Street Journal, Newsweek, USA Today, Reuters, the Associated Press, CNN and ABC News.

This silicon-based, ultra high-speed modulator, developed by Pannicia and his research team at Intel, can encode light beams at up to 1 billion cycles a second.

As a result of this and earlier optical discoveries at Intel, Paniccia, who still gives Binghamton Prof. Robert Pompi credit as his mentor, is now quite the celebrity scientist. Invitations to speak and present his work have streamed in from academic and research conferences across the globe. In the fall, for instance, he's slated to speak at the opening of a conference in Hong Kong. "I'm going to do my world tour this year," he joked.

Paniccia is also such a star at home that his nearly 3-year-old son, Michael Sergio, thinks he's a good buddy of Intel CEO Craig Barrett. Used to watching a tape of his father and Barrett appearing on stage to introduce the silicon modulator, young Michael constantly asks Paniccia whether he's seen Barrett that day.

Such accolades don't go to Paniccia's head, though. The son of Italian immigrants who never went past fifth grade in school, he didn't grow up in working-class Endicott dreaming of fame and fortune. He just wanted to go out to Silicon Valley and work on new technologies.

" Growing up like that really balances you," he said. "I never get caught up in the money or the power or the title." He also credits his "extremely supportive" wife, Rachel, whom he met at Purdue.

Still, Paniccia can't help but be amazed by his success. On a trip back home to Endicott this past spring, where he received nearly a hero's welcome from family and old friends, he marveled at where his dream has led.

"Sometimes you pinch yourself," he said. "Most people don't get this opportunity . . . Growing up, who would've thought?"

-- Alan Breznick '79