Single microresonators able to generate light pulses at several discrete frequencies could eliminate the need to integrate multiple lasers into communications and spectroscopy equipment. Researchers at Purdue University have developed a way use the microresonators as frequency combs, with precisely controlled emission at evenly spaced frequencies, using the concept of dark pulses. Whereas conventional optical communication requires many lasers to transmit various frequencies, devices using these microresonators might require only a single light source, which is then transformed to emit light at multiple wavelengths. This would reduce cost and make possible more compact optical systems small enough to fit on electronic chips, said professor Minghao Qi. This graphic depicts the optical spectrum of a pump laser used in new microresonator technology and an intriguing optical phenomenon called dark pulses, which might be harnessed to precisely control frequency comb lines. Courtesy of the Birck Nanotechnology Center, Purdue University. "Say you have 40 channels," he said. "If we have 40 individual lasers, together with their individual control circuitry on a single telecommunication chip, then your cost is high. If one of the lasers goes down, you have to replace the entire chip. You could achieve significant cost reduction if you were able to use just one laser to create multiple wavelengths to drive different channels." The microresonators accumulate optical power and enhance the otherwise weak effect of optical nonlinear interaction, which allows for the generation of numerous frequencies. Each mocroresonator has a radius of about 100 μm and is fabricated from silicon nitride, a material compatible with the silicon widely used in electronics. Researchers had previously created "bulk optics" systems, which used mirrors, lenses and other optical components arranged on a vibration-dampened table several feet long to convert and transmit pulsed signals. One challenge in miniaturizing the apparatus is maintaining anomalous dispersion, which makes the high frequency components of a pulse travel faster than the lower ones and was previously considered necessary to generate the frequency combs. "To achieve anomalous dispersion in silicon nitride microresonators ordinarily requires very thick film, which is susceptible to cracking and not practical to manufacture," said professor Andrew M. Weiner. "Here, we show how to generate the combs without anomalous dispersion, so we potentially no longer need the thick films." Dark pulses can be envisioned as a shutter that is normally open to allow light to pass through, but can quickly close to block the light and then open again to turn the light back on. The entire process can be as fast as 1 to 2 ps, almost 100 times faster than the switching speed of the fastest computer microprocessor now available. The technology could enhance research using spectroscopy, as well as enable miniature optical sensors to detect and measure chemicals, as well as optical communications systems that transmit greater volumes of information with better quality and at lower cost. The microresonators can work not only in the near-infrared range required for communications but also in the mid-infrared and visible ranges used for chemical and biological sensing. "If you had a comb, you could probe molecules at multiple wavelengths at once, so it's faster and provides more information about the molecule," said postdoctoral research associate Xiaoxiao Xue. Another potential application is optical clocks, enabling computers to synchronize the operation of the billions of transistors in their microprocessor chips. The technology also could be used in photonics-assisted processing of microwave signals, especially those with large bandwidth that are difficult to process using pure electronic methods. Funding came from the National Science Foundation, U.S. Air Force Office of Scientific Research and DARPA. The research was published in Nature Photonics (doi: 10.1038/nphoton.2015.137) and Laser & Photonics Review (doi: 10.1002/lpor.201500107). For more information, visit www.purdue.edu.