Reprogrammable photonic circuits based on a novel programmable material might speed the rate at which engineers can develop working photonic devices, researchers say.
Electronic integrated circuits (ICs) are nowadays key to many technologies, but their light-based counterparts, photonic integrated circuits (PICs), may offer many advantages, such as lower energy consumption and faster operation. However, current fabrication methods for PICs experience a great deal of variability, such that many of the resulting devices are slightly off base from the desired specifications, resulting in limited yields.
One potential way around this problem is to develop PICs that are reconfigurable or programmable to help compensate for any slight variability during fabrication. A key ingredient for a reconfigurable PIC is an optical material whose refractive index—the amount by which it slows light passing through it—is adjustable between two or more states.
However, many of the switchable optical materials that previous research examined required continuous heating, and thus a constant power supply and intricate systems to control this heat. Other materials suffered degraded performance upon switching in terms of signal loss.
Now scientists have found a switchable material that they say avoids the shortfalls of prior work and could lead to practical reconfigurable PICs. "This is the first programmable photonic circuit where you can program the photonic material itself and reset it and [it] requires no power to keep its programmed state," says Oded Raz, an electrical engineer at the Eindhoven University of Technology in the Netherlands who led the work.
The material in question is hydrogenated amorphous silicon, which is currently used in thin-film silicon solar cells. Previous research on a phenomenon dubbed the Staebler-Wronski effect found that light or heat could alter the optical and electrical properties of hydrogenated amorphous silicon, whereas slowly cooling in the dark could partially restore its optical properties.
Although the Staebler-Wronski effect is undesirable with thin-film silicon solar cells, the scientists reasoned it might prove useful in reconfigurable PICs. "It's nice to take an effect that is considered to be a liability in one context and to flip it on its head and change it to very beneficial in another context," Raz says.
The researchers investigated how a thin layer of hydrogenated amorphous silicon changed in response to cycles where they soaked it in near-infrared laser light for 100 hours or more and then it slowly cooled, or annealed, for four hours in the dark. They found the light could increase the material’s refractive index by 0.3 percent whereas annealing reversed this change, a change they found was due to how light and heat led the material to expand in volume.
The scientists then created reconfigurable optical switches using microscopic rings of hydrogenated amorphous silicon. They found they could reversibly change the refractive indexes of these devices with no detectable increase in optical loss. In addition, in experiments with freestanding membranes of hydrogenated amorphous silicon, they found there was long-term stability with these programmed states, each lasting at least a month.
The key criticism of these findings is likely how much light exposure it currently takes for switching, Raz says. Furthermore, a 0.3 percent change in refractive index is very small, "and saying this on its own can solve all the problems with photonic devices is a stretch," he adds.
Still, Raz notes there was a lot of research in the 1980s on how to reverse the Staebler-Wronski effect, "and we believe we can actually use all those insights on how to make this effect smaller to actually make this effect larger and respond faster." In addition, future research with similar materials such as amorphous silicon germanium or amorphous silicon carbon might reveal that they prove better at switching than hydrogenated amorphous silicon, he adds.
If future research can boost the strength of the switching effect, the improvements in production yields could also be significant, Raz says. Whereas previous yields of relatively basic photonic components might range from 10 percent to 20 percent, programmable optical materials could improve yields of such devices to between 50 percent and 80 percent, he suggests.
Boosting yields could in turn reduce the prototyping time needed for photonic devices. Currently it might take six to nine months to go from concept to fabrication of a PIC, but due to variability during fabrication, the final product might not do exactly what researchers wanted it to, "and you would have to make another run," Raz says. "Having programmable materials could allow for faster prototyping."
Ideally, programmable optical materials could lead to a photonic version of a field-programmable gate array (FPGA), which are electronic ICs that users can reconfigure after manufacture. This could help bring prototyping times "down to maybe 10 hours or two weeks, compared to one year," Raz says.
The scientists detailed their findings in the 19 March issue of the journal Advanced Optical Materials.
Charles Q. Choi is a science reporter who contributes regularly to IEEE Spectrum. He has written for Scientific American, The New York Times, Wired, and Science, among others.