Skip to Content

Telecommunications Seeks Its Guiding Light

January 1, 1998

“Pipe it in!” That’s the battle cry in telecommunications these days. The more information that can travel down an optical fiber-whether as phone calls, video games, movies, or symphonies-the better. Now MIT researchers in materials science, physics, and electrical engineering and computer science have collaborated to create a device that could dramatically increase the carrying capacity of optical fibers. The device, a “photonic bandgap microcavity resonator,” uses a microcavity, or defect, in a material called photonic crystal to guide the behavior of photons in much the same way that defects in a semiconductor can be used to control electrical properties.

The device comes at a crucial time, notes Lionel Kimerling, one of the project’s supervisors and director of the Materials Processing Center at MIT. Telecommunications companies need to increase the carrying capacity, or bandwidth, of the fiber base already installed in just about all buildings. Since adding more fiber is costly, telecommunications companies are looking for ways to process signals more quickly-a task traditionally assigned to electronics. Although optical fibers can carry many data streams simultaneously, the signals must be processed sequentially-switching a single phone call entails processing the entire highway of signals to pick out the one needed. That inordinate amount of processing plunks a speed limit on the information highway, effectively limiting the carrying capacity of the fiber. But electronic switching and processing devices themselves are hitting a glass ceiling in their career: although transistors are getting smaller and smaller, there is a limit to how many can fit on a chip.

The best near-term solution, a technique called wavelength division multiplexing pioneered by Lucent Technologies, uses an optical device called a channel drop filter to separate light into color bands, or frequencies. This enables different data streams to be channeled along separate frequencies so that the signals can be processed simultaneously rather than in sequence. The Lucent technology, however, has “a real estate problem,” says graduate student Jim Foresi, who was involved in the fabrication and testing of the MIT device. Current designs are built on a centimeter scale, making precise fabrication difficult and expensive. The devices for 16-channel systems, for instance, are so large that only two of them fit on a 6-inch wafer. At this scale, says Foresi, systems with hundreds of channels would be impractical.

The MIT resonator, or waveguide, could pave the way toward wavelength multiplexing technology on the micron scale-a major advantage for large systems that contain thousands of devices. The device uses a different kind of “filter”-a photonic crystal-to tease out a single frequency from a strand of light. Each waveguide is only 0.5 microns in width; although each captures only one frequency, 100 of them would still be smaller than the Lucent guides. Working in synch, they could let 100 signals travel down the same optical fiber at the same time.

“We all knew this was possible,” says Henri Benisty, research scientist at the Condensed Matter Physics Lab at the Ecole Polytechnique in Palaisseau, France. “The real tour de force was to make the guide small and still obtain fair optical confinement.”

The four-year project began in the physics department, where professor John Joannopoulos and research scientist Pierre Villeneuve explored what Joannopoulos calls the “magical properties” of photonic crystals. These artificial materials exhibit behavior that can be tailored specifically to affect the flow of light. Each perfect crystal acts as a filter: it prohibits entry to a certain range of frequencies, creating what is called a photonic bandgap. If there is a defect in the crystal, however, photons in the prohibited frequencies will be drawn to it. By altering the shape of the defect, scientists can “tune” a crystal to isolate a particular frequency.

“The crystal allows you to custom design electromagnetic states with specific properties. You can pick off just the wavelength you want,” Villeneuve explains.

Together with Kimerling, the researchers began to explore the idea of creating a tiny photonic crystal waveguide operating at optical wavelengths (around 1.5 microns). Previously designed photonic bandgap devices operate at microwave wave-lengths (1 to 100 millimeters) and are 100 times larger. The materials of choice for MIT’s photonic waveguide-silicon and silicon dioxide-are unusual in comparison with those currently used, but their great advantage is ease of miniaturization. “We wanted the waveguides small and to be able to integrate well with the material that chips are made of,” Kimerling explains.

The waveguide was fabricated using a technique called x-ray lithography, which has been used in chip fabrication but never before applied to photonic devices. It consists of a line of eight air holes arrayed on either side of a tiny micro-cavity in a layer of otherwise unblemished single-crystal silicon. The silicon is mounted on a layer of silicon dioxide and a silicon base. The closely spaced air holes act as mirrors: they reflect a very wide range of frequencies through a destructive interference effect. The defect, however, permits one frequency that would otherwise be reflected to enter through the first four holes. It becomes trapped in the micro-cavity for a brief period of time, then exits through the second group of four holes.

The MIT device is not yet ready to hit the market. And one of the things that has to be done before it gets there is to make the waveguide compatible with current technology while minimizing signal loss and power. “One challenge I see is to couple the fiber photons into the narrow guides,” points out French researcher Benisty.

Nonetheless, researchers contend that these tiny devices could lead to new types of switches, frequency modulators, channel drop filters, low-threshold lasers, and light-emitting diodes-essential elements in telecommunications. Ultimately, says Foresi, “what we want is an integrated optics technology that is compatible in materials and dimensions with integrated circuit technology, so we can capture the strengths of both technologies at the same time.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.