An array of 4096 optical antennas built on a single silicon chip can steer light in arbitrary directions, holding promise for applications such as holographic televisions and medical imaging devices that can be threaded through tiny blood vessels, among others. Although the concept of such “phased arrays” of antennas goes back more than a century for radio transmitters, scientists recently have been applying this model to optics. Optical wavelengths are much smaller than radio wavelengths, so the antennas have to be fabricated with high accuracy at the nanometer scale. No two-dimensional tunable phased array had been built on a chip before, and previously, the largest passive, or nontunable, array had only 16 antennas. Now, scientists at MIT’s Research Laboratory of Electronics have developed a 64 x 64 grid of antennas on a single silicon chip that takes in laser light, knocks it slightly out of phase to produce interference patterns and then re-emits it via the nanoantennas etched into the chip’s surface. The emission plane (left) and image plane (right) patterns out of a 4096-element nanophotonic phased array designed by MIT associate professor Michael Watts and graduate student Jie Sun to project the university’s logo in the image plane. Images courtesy of Jie Sun. In the 4096 antenna chip, the phase shifts are precalculated to produce rows of images of the MIT logo. The antennas are not like the pixels of a black-and-white monitor that turn on and off in a pattern tracing the logo; rather, they all emit light. If you were close enough (and had infrared vision), you could see a regular array of pinpricks of light. Seen from more than a few millimeters away, however, the interference of the antennas’ phase-shifted beams produces a more intricate image. The MIT team also created an 8 x 8 grid of antennas. The phase shift these antennas produced is tunable, so the chip can steer light in an arbitrary direction. In both chips, the design of the antenna is the same; in principle, the team could have built tuning elements into the antennas of the larger chip. However, there would be 4000 wires coming off the chip, way too many to solder up, said Michael Watts, an associate professor of electrical engineering at MIT. Wiring limitations mean that even the smaller chip is tunable only a row or column at a time. But that’s enough to produce some interesting interference patterns that demonstrate that the tuning elements are working. The large chip, too, largely constitutes a proof of principle, Watts said. “It’s kind of amazing that this actually worked,” he said. “It’s really nanometer precision of the phase, and you’re talking about a fairly large chip.” The new chip consists of tiny antennas arranged in a 64 x 64 grid. “I think it’s one of the first clearly competitive applications where photonics wins,” said Michal Lipson, an associate professor of electrical and computer engineering at Cornell University and head of the Cornell Nanophotonics Group. Within the photonics community, Lipson says, most work is geared toward “the promise that if photonics is embedded in electronic systems, it’s going to really improve things. Here, [the MIT team] has developed a complete system. It’s not a small component: This system is ready to go. So it’s very convincing.” The prototype chip’s tuning limitations are no reason to doubt the practicality of the design, Lipson said. “It’s just physically hard to come up with a very high number of contacts that are external,” she said. “Now, if you were to integrate everything so that it’s all on silicon, there shouldn’t be any problem to integrate those contacts.” The study appeared in Nature (doi: 10.1038/nature11727). For more information, visit: www.mit.edu