Warp speed is still out of reach for spaceships, but two new experiments have pushed a pulse of light beyond the speed limit of 300,000 kilometers per second set down in Einstein's theory of special relativity. Although physicists had pulled off similar feats before, two teams now report ways to beat down the loss of light so that much more of it appears to break the universal speed limit. Don't worry, the tricky experiments don't really violate relativity. But the techniques might, in principle, slightly speed up optical communications.
Here's how you make a pulse of light appear to travel faster than light speed. The pulse can be thought of as a kind of rogue wave of electromagnetic radiation zipping through space so that if you made a graph of the pulse's intensity, it would start at zero, smoothly climb to a single peak, and then decline back toward zero. But such a peak can also be thought of as a collection of waves with a range of wavelengths, all continuously oscillating up and down and piled on top of one another. At the center of the pulse, the various waves all line up and reinforce each other (see figure). In contrast, near the leading and trailing edges of the pulse, the different waves get out of sync and cancel one another out.
Now suppose you run the light pulse through a special material that slows down some wavelengths of light more than others. That can change the way the waves line up and, ironically, shift the spot at which the various waves reinforce each other forward, making the peak appear to jump forward faster than light. The peak can even appear to emerge from the back of the material before it enters the front. None of this violates relativity, however, as that would require the individual waves to run faster than light speed. More generally, physicists now interpret relativity to mean that information cannot be transmitted faster than light. And it's the unvarying speed of the first overlapping light waves, not the exact position of the pulse's peak, that determines the ultimate rate at which information can flow.
Vitaliy Lomakin of the University of California, San Diego, and colleagues at the Public University of Navarre in Pamplona, Spain, put this idea into practice by sending microwaves into a 35-micrometer-thick, holey sheet of copper sandwiched between two 0.79-millimeter-thick Teflon discs. As a consequence of this design, the peak of a microwave pulse can emerge from the other side of the device even before it enters the metal sandwich.
But the metal doesn't naturally let much light through. That's where the Teflon comes in. The two layers maintain the brightness and direction of the waves while the metal's pattern of holes builds up the signal in these strong waves. Whereas previous experiments might have seen less than 1% of the light pulse break the cosmic speed limit, the interaction between the metal and Teflon allowed the team to send 10% through about 100 picoseconds early , an advance soon to be described in Physical Review B. "This is achieved with a remarkably thin structure that can be fabricated easily in a wide spectrum from microwaves to visible light," Lomakin says.
Li Zhan and colleagues at Shanghai Jiao Tong University in China say they could make a pulse arrive even earlier with optical fibers, which are already used for high-speed data communications. This team sent an infrared light signal clockwise through a loop of optical fiber and measured it at two sensors, one near the point where the light entered the fiber and the other 10 meters on. Ordinarily, the signal passing through the silicon fiber would show up in the first sensor and then reach the second sensor 48.6 nanoseconds later. However, Zhan and his colleagues managed to speed the signal up so much that it arrived at the second sensor 221.2 nanoseconds before it reached the first one.
In this experiment, the optical fiber itself played a role similar to that of the holey plate. To speed up the light signal, a second light wave ran counterclockwise through the optical fiber. The presence of that additional light changed the speed at which light waves of different wavelength zip along to modify the alignment of the waves. Typically, this second wave absorbs so much of the signal light that pushing the peak ahead even 1 nanosecond reduces the peak's intensity by about 20%. By contrast, the researchers managed to push the light forward by 211.3 nanoseconds before losing that much light , they report in a paper in press at Physical Review Letters.
Although information can't really travel faster than light speed, Zhan argues that communications could make small gains in the speed at which signals are detected. The receivers in optical communication systems also react to the peak of a pulse, not its leading edge. Pushing the peak closer to that edge might save only a few hundred nanoseconds, but Zhan says it could one day make the difference in the whirlwind world of high-speed stock exchange.
"This may be true, but they have not done the experiment," says Daniel Gauthier, a specialist in fast light at Duke University in Durham, North Carolina. Based on his research and that of others, he expects the pulse's information-carrying shape will be mangled in the process. Günter Nimtz, an expert in faster-than-light-speed phenomena at the University of Cologne in Germany, agrees that changes to the pulse shape would be problematic, but he suggests that with some knowledge about the wavelengths in the pulse, and the material it moves through, the receiving end could recover the information.