As much sense as it might make for the universe and its myriad life-fostering properties, it's annoying to have a limit such as c, the speed of light in a vacuum. There are lots of limits and constants in the cosmos, really, but the speed of light is just so there, keeping us humans from our most fervent sci-fi dreams: time travel, instantaneous communication, deep-space exploration. c is the rule, the most fundamental rule, that enforces loneliness in the universe.
And so we take notice of even the unlikeliest suggestions for getting around it, as when researchers at CERN clocked neutrinos, moving between its base on the Swiss-French border and an underground laboratory in Italy, at faster than light speed. Those results were, of course, bunk, but we all took a moment to dream of the possibilities, even with a full century of good science declaring in no uncertain terms that we shouldn't even consider such a reality.
Quantum physics has, however, come up with some intriguing simulations or approximations of faster-than-light phenomena. There is first this notion of entanglement, in which two particles or groupings of particles can be linked over a distance (up to 143 km in the most recent experiments), such that some action taken on a particle (a measurement, usually) is reflected in the other particle. It's not really faster-than-light communication, though, as a classical, slower-than-light back-channel is still needed in order to interpret the information gathered from the second, distant particle.
A second approximation of post-light speed quickness arises from what are known as fast-light materials. A new test of the concept is described this week in the journal Nature Photonics, courtesy of a research team based at NIST's Joint Quantum Institute (JQI). While the stunning illusion is of a particle in a beam beating out c, it's not quite that easy. The particle, as it moves through a certain medium, becomes dispersed into a bunch of different amplitude components. It's the same thing as regular visible light passing through a prism and emerging as a spectrum of distinct colors. The beam is shifted forward, but that just means the waveform's peak emerges have traveled faster than light.
It's sort of like hiding a financial gain on one account by marking a bunch of tiny losses in other, different accounts that can be resolved later on. It winds up such that there's a zero net gain from light speed after a full accounting, but the full accounting is delayed. The question explored in the JQI paper is whether or not it might still be possible to glean useful FTL (faster than light) information from such a system of "borrowing."
Quantum entanglement experiments typically look at relatively discrete particles properties like spin (up or down) and polarization (horizontal or vertical, for example). These are not the only possibilities for marrying particles though, and the JQI team looked instead at the continuous—not either/or, but all of the combinations between "either" and "or"— property of phase. Phase, if you were to look at the waveform peaks of a beam of light traveling through space, is the movement of those peaks forward and backward in time. That is, the peaks of the light waves are altered so that they arrive sooner or later than they were previously.
The result of entangling full beams of light is that a momentary fluctuation in one beam, possibly the result of a quantum fluctuation, will be reflected in the other beam. (As for the fluctuations, recall that the quantum world is uncertain, and so is always "fizzing" with particles that pop in and out of existence; the unchanging, constant alternative is too certain for quantum reality.) What the JQI experiment did is create two "daughter" beams from one larger beam, each one consisting of just a single photon and neither powerful enough to measure directly. One is known as the conjugate beam, the other the probe beam. The experiment then sends these two beams on two different paths. The path for the probe beam is an unencumbered route through a plain vacuum (or relative vacuum), while the probe beam passes through a cell filled with rubidium vapor. The result of this is typically "slow light," and the conjugate beam will wind up traveling slower than its partner, as expected.
There is an odd alternative, however, called anomalous dispersion. When you fire some beam of light through a medium, whether that's a prism or just some random crap, you calculate the resulting propagation as refraction. The refractive index is simply the ratio of c, the speed of light in a vacuum, over v, the speed of light through some substance. So, the more and more a substance slows down a beam of light traveling through it, the larger the index is. If the beam isn't slowed down at all, then you just have an index of "1." Easy enough.
The anomalous version occurs when the speed through a substance ends up being faster than the speed of light through a vacuum, giving a ratio below 1. This is a regular feature of ultraviolet light moving through glass, a pulse of light appearing to arrive faster than should be allowed by the constant speed of light. The cleverness behind the phenomenon is this: a pulse of light is not actually just one big unified wave pattern fluctuating in amplitude as it is a combination or sum of smaller waves of different wavelengths (frequencies) adding together into the big waveform.
It helps to see it:
Image: "The black curve shows the sum of the red, green and blue sine waves. When all of the waves are delayed, but the longer wavelengths [red] are delayed more and the shorter wavelengths [blue] are delayed less, then the overall pulse appears to be advanced in time!"/UCLA
You can interfere with the propagation of the constituent colors independently, such that you're delaying one band of color more than another, with the result being the pulse's peak appearing to have traveled faster than light. But what really happened is that the light's been manipulated such that longer wavelength red waves arrive artificially late to their destination. The pulse arriving early, in this case, does not mean that the beam has overall arrived early. It's a trick.
But it's a trick that seems like it might have potential for real faster than light communication. If a pulse can be registered after a beam has been subject to anomalous dispersion, does it matter whether or not there are still some frequencies of light lagging behind? This is what the experiment ultimately tested: whether or not that forward-shifted peak can be used in an information sense.
For that to be the case, the beam needs amplification post-dispersal, just as any very weak electrical signal might be boosted before filling the requirements for detectability (within a given system or device). The problem, the JQI team discovered, is that boosting a signal (adding gain) as weak as required for the storage of quantum information (which is very small, one photon or less) means that the boosted signal is effectively destroyed by quantum noise (remember the fizzing?) inherent in any amplification scheme. That is, there is no gain signal pure enough to boost less than a single photon simply because signal purity violates the uncertainty of the quantum world. (At much larger scales, functional purity in a gain signal could be achieved because the quantum fluctuations would be easily overridden.)
What we have now then is another barrier to FTL communication: quantum noise. We can slip through a pulse of light that is indeed breaking c, but it gets washed away in a sea of spontaneous fizz. "We did these experiments not to try to violate causality," notes the JQI's Paul Lett in a press release, "but because we wanted to see the fundamental way that quantum noise 'enforces' causality, and working near the limits of quantum noise also lets us examine the somewhat surprising differences between slow and fast light materials when it comes to the transport of information."