At Universität Rostock in Germany, researchers may have found a way to squeeze more bandwidth out of today’s telecommunications infrastructure. They have demonstrated the existence of molecules of temporal solitons, or permanent solitary waves. Such molecules could be used as a third “letter” for optically encoding data, an increase of 50 percent over the binary on-and-off encoding that is currently used.Investigators at Universität Rostock have experimentally demonstrated the existence of soliton molecules — stable bound states of two bright solitons on either side of a dark one. Image by Fedor Mitschke. Used with permission. Discovered more than a century ago, solitons can be thought of as solitary waves that can travel great distances and interact with each other without losing energy. Used to transmit data in optical fiber telecommunications, they come in two varieties: bright and dark. The former is a bright peak, and the latter is a notch in a bright setting. Dark solitons are found in fibers with normal dispersion, and bright solitons show up in anomalously dispersive fiber. To date, data in optical fibers has been coded as either a bright soliton or as nothing at all. There have been hints that it might be possible to have both bright and dark solitons simultaneously in the dispersion-managed fiber now favored for telecommunications. Using numerical experiments, the Rostock researchers found this to be true. Their theoretical calculations revealed a soliton molecule — two bright pulses on either side of a dark notch in a stable bound state. The calculations also showed that if the two bright pulses are too far apart or too close to each other in time, they relax to a favored separation as they propagate through a fiber.The scientists then conducted experiments to verify that soliton molecules exist. They constructed a scale model of a real telecommunications fiber, shortening the length from tens or hundreds of kilometers to a few tens of meters. Using a Ti:sapphire laser, an optical parametric oscillator and a Mach-Zehnder interferometer, they varied the time between pulses from a few hundred to more than 1000 fs. They measured the result after the pulses traveled through the fiber using an autocorrelator and an optical spectrum analyzer.What they found was that a soliton molecule did form when the separation was about 450 fs. They could vary the power somewhat and still create the molecules. The results agree very well with predictions based on numerical calculations.The researchers hope that the discovery will give telecommunications a three-letter alphabet of a bright soliton, no soliton and a soliton molecule. Translating these initial findings into practice will take time.