The feeble force of light alone can flip a nanometer-sized mechanical switch one way or the other, a team of electrical engineers reports. The little gizmo holds its position without power and at room temperature, so it might someday make a memory bit for an optical computer. Other researchers say it also introduces a promising new twist into the hot field of "optomechanics," which marries nanotechnology and optics. "This a new paradigm," says Markus Aspelmeyer, a physicist at the University of Vienna who was not involved in the research. "People are going to take a good look at this and use it in other schemes."
Since 2005, physicists and engineers have been using light to set tiny structures vibrating and control their motion. Much of their effort has focused on using light to suck energy out of a vibrating beam or cantilever to try to achieve new states of motion that can be described only by quantum mechanics, extending the quantum realm to the movement of humanmade objects. In fact, Aspelmeyer, Oskar Painter of the California Institute of Technology in Pasadena, and colleagues have managed to use laser light to "cool" a vibrating beam to the lowest energy state possible, the so-called quantum ground state, a key step toward achieving more complex quantum states of motion, as they reported 6 October in Nature.
Now, taking a different tack, Mahmood Bagheri, Hong Tang, and colleagues at Yale University have used laser light to pump energy into a tiny bridge of silicon, flipping it between two stable configurations—in effect, making a mechanical switch. The bridge measured 10 micrometers long, 500 nanometers wide, and 110 nanometers thick and was suspended about 250 nanometers above a glass chip. When researchers etched glass out from under it, the silicon expanded, so the bridge bowed either upward or downward, like a playing card squeezed lengthwise between your thumb and forefinger.
The trick was to use light to make the bridge shift between the two positions in a controllable way. The team started by making the bridge part of a longer oval racetrack for light etched onto the chip. Along the back stretch of this racetrack, or "optical cavity," ran a parallel track that could also carry light. Ordinarily, light going through that parallel track could bleed into the racetrack and accumulate there—but only if its wavelength were tuned so that a whole number of wavelengths fit around the track.
The presence of the bridge changes things in two ways, however. First, when the bridge bows down, light traveling through it feels a slowing effect from the glass. As a result, the track fills with light (or "resonates") at a longer wavelength than it does when the bridge bows up. Second, with the bridge in place, light can get into the racetrack even if its wavelength isn't quite right. The reason is that the bridge can vibrate. So light with a slightly too long, less energetic wavelength can enter the cavity by drawing energy from the vibrations, in the process causing the bridge to move less. In contrast, light with a slightly too short, more energetic wavelength can enter the cavity by losing energy to the bridge, increasing the bridge's vibrations.
So here's how Tang and company flipped the switch. Starting with the bridge in, say, the down position, they applied light of a slightly shorter, more energetic wavelength than the down position required. The light then pumped energy into the bridge, eventually causing it to flop wildly between the up and down positions. To put the bridge into, say, the up position, they then applied light of a slightly longer, less energetic wavelength than that position required. That light cooled the bridge and brought it nearly to rest in the desired configuration, as the team reports today in Nature Nanotechnology.
"It's very elegant work," says Michal Lipson, an optical physicist at Cornell University. Such bowed bridges could be used to store information in an optical computer memory—with, for example, bowed up standing for one and bowed down standing for zero. "It's the first time that [such a memory] has been demonstrated in a fairly achievable way," she says. "It's definitely feasible."
The work also introduces a potentially important new element to optomechanics, Aspelmeyer says. Before, researchers worked primarily with simple "linear" mechanical devices, whose vibration frequency remains the same no matter how vigorously they vibrate. Tang's bridge vibrates at a different frequency depending on its configuration and how much it is moving, and such "nonlinearity" opens up new opportunities to control light and motion in more complex ways, Aspelmeyer says.