
ram Mooradian may have the key to opening the door to the next generation of laser technology. His company, Novalux, headquartered in a tree-lined industrial park in Sunnyvale, Calif., says it has new technology that will blast more data along fiber-optic networks than any lasers currently available.
If the claims prove out, Mooradian's research and patents may change everything for a market that is projected to explode to $52 billion by 2005.
Mooradian serves as chief technical officer for Novalux, which roughly translates to "new light" in Latin. He counts among his past jobs heading up the Quantum Electronics Group at MIT's Lincoln Laboratory, and a stint as the CTO of Microcor, a pioneering photonics company.
The almost three-year-old company is poised to unleash what it claims is the holy grail of laser technology: a laser-emitting gallium arsenide chip that may edge out traditional edge-emitter and vertical-cavity surface-emitting laser (VCSEL) chips in terms of output, power requirements, manufacturability and optical quality. The initial goal is the communications infrastructure market, with an eye toward packet-switching devices in particular.
Novalux's claims are backed by some heavyweight industry talent. Besides Mooradian, the management team includes chief executive officer Malcolm Thompson, who served as the chief technologist for Xerox's legendary Palo Alto Research Center (PARC) and as chairman of the United Stated Display Consortium; Charlie Townes, the inventor of the laser and father of fiber-optic light transmission, sits on the board of directors; and Intel fab veteran Gary Oppendahl heads up production.
The startup's key to the future lies in creating a new type of laser chip it calls a NECSEL, or Novalux extended-cavity surface-emitting laser, to be used as a pump laser in various communications systems.
NECSELs share some similarities with the current VCSEL, but the company claims it produces a beam that is larger (about 100 microns wide, compared with the standard VCSEL diameter of about 10 to 20 microns), more powerful, with a better tunable-wavelength capability and needing less power than VCSELs
"Do the math," said Mooradian. "You're looking at fewer parts using less power, spaced farther apart and transmitting more data."
It's not hard to understand why the venture capital crowd likes this guy. His math may also be converted to stock options.
"A conventional VCSEL is a small, low-power device," he explained. "Typically if you get a milliwatt out of a VCSEL that's considered reasonably good power. They're designed for low-power applications, for short-distance communication, optical interfaces and circuit backplanes and so on. But if you try to achieve more output power you have to make a larger device, which then becomes highly multimode."
Novalux has solved that problem, Mooradian said, by formulating larger devices with a controlled mode, resulting in stronger output with the beam fundamentally more coherent and more perfectly round.
This enables NECSELs to focus more power into a single-mode fiber to power amplifiers in fiber-optic deployments, he explained. "This has been a continually difficult problem for edge-emitter designs because it is very difficult to get really high power out of these devices without the risk of overheating and failure."
Mooradian said the delicate structures of the VCSEL devices can disintegrate under the high-power demands that are ever more prevalent in today's multimedia-centric communications environment. Thus, the company sees its products as a particularly good fit in dense wave-division multiplexing deployments.
"The only way that edge emitters can keep up with the power demand or data rate demand is just by adding more and more pump lasers, and it becomes increasingly cumbersome to add more and more edge emitters onto an amplifier," said Mooradian.
Insiders say the company has been keeping very quiet the past couple of years as it gets its intellectual property together and files the necessary patents. And although product details were still sketchy as of press time, Mooradian claimed to have some well-known telecom customers lined up and said Novalux is set to start production. It has its own fab and clean rooms on site in Sunnyvale, and is in the process of acquiring an additional 100,000 square feet of nearby manufacturing space to handle the expected capacity.
Both Thompson, the Novalux CEO, and Mooradian said the NECSEL is applicable to many segments of many markets. For telecom, the company is eyeing long-haul, metro and access equipment makers.
Mooradian, who founded Novalux in 1998, has an extensive background developing laser technology for communications. He earned a PhD in physics from Purdue University; his BS, also in physics, was from Worcester University. Mooradian holds more than 40 patents in photonics and laser fields, and has served as associate editor for both the IEEE Journal of Quantum Electronics and the Optical Society of America. He has also served as vice chairman of the committee on Optical Science and Technology for the National Academy of Sciences/National Research Council.
Thompson and Mooradian agree there is an enormous demand for this technology in communications now and possibly in other markets later, such as high-definition displays. Thompson's strong background in display technology, in fact, was one reason why Novalux wanted him as CEO. "We're focusing on telecom initially," Thompson said, "but there's enormous potential for what we're doing when it comes to displays. We could, for example, use our parts to make low-power RGB lasers and make beautiful, and powerful, displays."
As Novalux achieves its volume manufacturing goals for telco, those same devices could be convertible into powerful digital light sources for various needs, said Mooradian. "The light engine is the last remaining obstacle to brilliant high-definition projection displays," he said, predicting that such devices will eventually find markets in everything from home entertainment to large cinemas, as well as in specialty lighting markets.
"Only about 3 percent of the light output from a tungsten lamp is in the visible range," he said. "So if you have a 10 percent or 20 percent efficient device, just in the wavelength that you want, you not only have the ability to perfectly shape the beam, but to change the color. Say you have a micro-sensor that adapts your headlights to fog, rain, lighting conditions, reflective feedback from signs and road surfaces, to increase your safety margin." Given the preponderance of embedded systems in consumer products, he said, this is a very likely scenario, as is the concept of using NECSELs for signage. "If you have a visible light source, clearly all of these plastic neon signs will ultimately disappear and you will be able to do things that you've never been able to do, like change the colors in real-time. You might have a plastic sign, and you plug it in the wall down below and run plastic fiber up to the sign, so the servicing costs are dramatically decreased."
As to whether the existing telecom infrastructure will need any retrofitting to utilize Novalux's lasers, both Mooradian and Thompson say it's not really an issue. "We'll have to work with the user community to integrate our technology," said Mooradian. Thompson said the company's parts will use standard packaging methods (the first in a 14-pin butterfly design) that will let OEMs integrate them easily.
Mooradian is confident Novalux's technology will be applicable across the entire spectrum. "We'll be able not only to fill the entire usable spectrum of fiber, but be able to boost the signals beyond what's being done now, beyond the erbium amplifier."
He knows competitors will be along shortly, but feels confident the company's patents will create "a large barrier of intellectual properties" for any would-be market intruders. "Even today there's a lot of people that don't know what we're doing, and that was intentional. We wanted to have a lead on this technology," he said.
Mooradian, a lifelong laser man, admits that lasers are funny things. When they were first invented no one really knew what to do with them, and for a long time people had unrealistic, Star Wars-fueled misinterpretations of the technology. But now, as the world embraces fiber optics like never before, "Lasers have become the empowerment for communications as we know it today."
John Poultney is Executive Editor of Communication Systems Design, a sister publication of EE Times.