RealTime IT News

Nanotech Runs Behind Semiconductors, MEMS in Optical IC Market

The market for integrated circuits (ICs) used in optical switches will be worth more than $5 billion in 2006, according to a report by Pioneer Consulting, and nanotechnology is one of three chip technologies vying for a piece of the pie.

The report, "Optical Chips: Enabling Technologies and Markets from Semiconductors to MEMS, Nano-Optics and Photonic Crystals," predicts that total worldwide optical IC sales will increase from $654.3 million in 2002 to $5.4 billion in 2006. The report used a fairly broad definition of optical ICs, encompassing all chips, regardless of manufacture, used in optical switches.

According to Paul Kellett, senior director of research for Pioneer Consulting, semiconductor chips have a comfortable lead in market share over MEMS and nanotechnology and that should continue for some time to come. But if one of the other technologies offers a clear challenge to semiconductors, Kellett said it would be MEMS.

As for nanotechnology, companies like New Jersey-based NanoOpto are producing sub-wavelength optical elements (SOEs) for use by optical component manufacturers supplying the global fiber optic networking industry. NanoOpto's technology is based on the work of Dr. Stephen Chou, at Princeton University.

But Kellett told NanotechPlanet he believes that any substantial use of optical ICs based on nanotechnology is a minimum of three years away, and then the market will crank up very slowly. Nanotech developments, he said, are four years behind MEMS.

The promise of nanotechnology in optical ICs is the potential to squeeze more transistors onto a single chip. But Kellett's concerns about fabrication are centered around Moore's second law, which states the smaller an IC chip becomes the more expensive it is to manufacture using existing fabrication techniques. Companies who want to sell nanotechnology as the answer for smaller, inexpensive chips have to prove they can keep the fabrication costs in check.

"You've got to crawl before you can run," Kellett said. "Simple SOEs are being produced, but not in any real volume. It really comes down to, I think, to making the fabrication process work the way it is supposed to. The critical thing is to have a fabrication technology that lives up its cost-saving promise."

While chips based on nanotechnology are still primarily in the laboratory stage, Pioneer's report shows an increase in MEMS usage, which will only grow over the report's forecast period (to 2006). Other research firms are also bullish on MEMS chips. In-Stat/MDR forecasts that revenues for MEMS in optical networking will grow from $33.1 million in 2001 to more than $1.8 billion in 2006. MEMS-based product offerings are also expanding beyond switches to include tunable lasers, tunable filters, variable optical attenuators, and dynamic gain equalizers.

Encouraging forecasts related to anything in telecommunications seem a bit out of place in an industry that keeps finding new bottoms instead of a long-awaited recovery, but the capital markets, and their patience with the sector, will make for some interesting business decisions in the future. Kellett said he expects a dip in R&D spending in the telecom sector given the woes of its major players, but that's not where most of the action for optical ICs will be.

"A lot of the companies on the cutting edge are smaller companies," he said. "As long as they have VC money they are going to use it on R&D."

An industry where the technical developments of the little guys outpace the big boys is often headed for consolidation, and Kellett said you can expect to see plenty of that in the future, especially if and when the VC money dries up.

"Clearly one of the trends we will see going forward is bargain-basement shopping," Kellett told NanotechPlanet. "There's a lot of talent out there. It'll be a good market for buyers, but not sellers."