Optical components

Inplane's in the Money

Inplane Photonics Inc. emerged from stealth mode today (ta-da!) to tell the world it's developing integrated components based on erbium-doped waveguides (see Leading Scientists Launch Inplane).

Inplane (formerly Key Optics) also announced that it has raised $20 million in funding so far from Jerusalem Venture Partners and Morgenthaler. The funding was awarded in installments, the first back in March 2001, about five months after the company was incorporated.

The coming-out announcement is timely because Inplane shipped its first product -- a gain block for erbium-doped waveguide amplifiers (EDWAs) -- to a customer last Friday (March 1), John Kostibas, the company's VP of sales and marketing, told Light Reading.

EDWAs promise to cut costs in metro networks by providing a smaller and lower-power alternative to using Erbium Doped-Fiber Amplifiers (EDFAs). If they can be made small enough, they might also kick-start the development of much more complicated integrated components on a chip, because they compensate for the optical losses that result when many different optical elements are strung together.

The shipment is to a single customer (hey -- you have to start somewhere). It's an indicator that Inplane has a fighting chance of surviving in the increasingly crowded waveguide amplifier space. Other companies with similar products include France’s Teem Photonics, Denmark’s Cisilias A/S, Australia’s Redfern Integrated Optics, and the U.S.’s Molecular OptoElectronics Corp. (MOEC) and Northstar Photonics Inc. (see Which Amp is the Champ?).

But, although Inplane is the newest of the bunch, its founders are veterans of waveguide amplifier technology. The startup's CEO Joseph Shmulovich and CTO Allan Bruce worked together at Bell Labs since 1988, developing technology that looks as if it's found its way into Inplane. The pair filed their first patent on an "erbium-doped planar optical device" back in April 1991 and numerous other patents since.

Kostibas contends it is Inplane's in-house expertise, not just in components, but also in manufacturing, which sets it apart from the competition. When pressed he adds, "We are using waveguides with very high [refractive] index contrast, which allows us to bend the radius [of the waveguide] more tightly." As a result, Inplane can place the waveguides on its chips in a more compact arrangement, making it possible to get more chips out of the same wafer, and thus making it easier and cheaper to manufacture in volume.

It turns out that Inplane, like Symmorphix Inc., manufactures its components using physical vapor deposition (PVD) -- a technique used to lay down the thin-film transistors that operate pixels in a flat panel display. Both companies contend that PVD gives better results than other processing techniques (see Symmorphix Joins the Amp Camp). Indeed, it's the manufacturing technique that enables Inplane to make more compact devices -- PVD makes it possible to include higher concentrations of erbium and other dopants in glass than other approaches do.

"The most important message is that we are not based on one product," says Kostibas. "Soon you will see the amplifier combined with other elements." He adds that Inplane plans to show a prototype tunable dispersion compensator in Q2 2002.

— Pauline Rigby, Senior Editor, Light Reading
Dr.Q 12/4/2012 | 10:51:02 PM
re: Inplane's in the Money Welcome, Inplane, to the EDWA marketplace.

Now the hard questions start:
-How much gain?
-How much power?
-How small?
-How cheap?

Inquiring minds want to know...
lifer 12/4/2012 | 10:50:51 PM
re: Inplane's in the Money Amen, brother.

how is it that waveguides on a substrate
will occupy any less space
and/or perform any better
than a coil of fiber?

perhaps the integrated device
will be more convenient to manufacture.
you do not have to fusion splice
discrete wdms, isolators and taps.

but will these integrated components
have any where near the performance
of the "discretes"?

probably more effective to master
the manufacture of the fiber "spaghetti"


Dr.Q 12/4/2012 | 10:50:15 PM
re: Inplane's in the Money 'lifer' asks
>how is it that waveguides on a substrate
>will occupy any less space and/or perform any
> better than a coil of fiber?

The tradeoff is one of performance. The Erbium physics are the same in fibers as planar waveguides--the gain per cm initially increases as the Erbium concentration rises, but above a critical Erbium density the efficiency of converting pump light to signal light drops (due to Erbium ions being close enough together to interact).
In fibers it is straightforward to control the density, and to use long fiber lengths to get more Erbium ions. To get more gain, you use more Erbium and more pump light.
In a planar waveguide you can't go to (much) longer lengths. Thus you have to increase the Erbium concentration, and increase the pump light much faster than in an EDFA.
The result is that you get LESS total gain, LESS output power and require MORE pump power in an EDWA than EDFA.
The advantage of EDWA is size (a few cm for packaged EDWA chip, vs several inches for EDFA module), and cost. The EDWA chip is made in batch process. The EDFA is hand assembled with long coils of fiber.
Cost will drive the EDWA to be a big winner in metro and access markets. Performance will keep the EDFA dominant in the long haul systems.

- Dr. Q
Sign In