Become a fan of Slashdot on Facebook


Forgot your password?
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. ×

Single Photons Do Not Exceed the Speed of Light 196

GhigoRenzulli writes "A group of physicists at the Hong Kong University of Science and Technology (HKUST) led by Prof Shengwang Du reported the direct observation of optical precursor of a single photon and proved that single photons cannot travel faster than the speed of light in vacuum. HKUST's study reaffirms Einstein's theory that nothing travels faster than light and closes a decade-long debate about the speed of a single photon. ... Discovery of superluminal propagation of optical pulses in some specific medium 10 years ago has evoked the world's dream of time travel, but later scientists realized that it is only a visual effect where the superluminal 'group' velocity of many photons could not be used for transmitting any real information. Then people set their hope on single photons because in the strange quantum world nothing seems impossible — a single photon may be possible to travel faster than the speed limit in the classical world. Because of lack of experimental evidence of single photon velocity, this is also an open debate among physicists. To tackle the problem, Prof Du's team measured the ultimate speed of a single photon with controllable waveforms. The study, which showed that single photons also obey the speed limit c, confirms Einstein's causality; that is, an effect cannot occur before its cause."
This discussion has been archived. No new comments can be posted.

Single Photons Do Not Exceed the Speed of Light

Comments Filter:
  • by jasomill ( 186436 ) on Monday July 25, 2011 @06:39PM (#36877722)

    QED says that the path light travels is a path of least action

    Perhaps this is pedantry, but wouldn't that be classical mechanics? The classical "action" is mass times velocity times length, so it would vanish identically for massless particles —though modern formulations tend to substitute Hamilton's principle (discovered, incidentally, by Lagrange): admissible paths are critical points for the map from paths to real numbers given by integrating the system's Lagrangian over the path. But still, this only holds for quantum systems in the limit as the Planck constant goes to zero —hence Feynman's formalism that effectively reduces Hamilton's principle to a "stationary phase approximation" of an infinite-dimensional path integral.

    When it comes down tobrass tacks, I'm pretty sure QED doesn't say anything about the "path" of a single photon — to the extent paths are introduced at all, one considers integrals over spaces of paths, including, in the usual formulation, paths where "photons travel faster than light."

    The end result, as I recall, indicates that the probability that any given experiment would reveal a photon traveling faster than light is zero. And I'm not really sure how you would "prove" the difference between "zero probability" and "unimaginably small nonzero probability" experimentally; I'm pretty sure these "virtual tachyons" are just "unobservable intermediate results" in the formalism, that "faster than light photons" implies a violation of local conservation of energy that is generally held to be true by hypothesis.

    So I'm confused by the summary and the press release from the outset, rather generally, since it's impossible in principle to "prove" that something cannot occur by "direct observation." Observation of what? All possible photon trajectories?

    So you're right — it's certainly not obvious!

  • Re:Obvious? (Score:5, Interesting)

    by Altrag ( 195300 ) on Monday July 25, 2011 @08:59PM (#36879042)

    By measuring the relative velocities of all of the galaxies, it can be extrapolated that everything (at least, everything within our visible radius) expanded from a single point. Which is not to say a central point -- for all observational purposes, the earth is the center of the universe thanks to the fixed speed of light. We can see back 13ish billion years in every direction -- there's no directional bias that would suggest we're not at the center.

    However, the same thing happens in alpha centauri as well -- they'll be able to see a few lightyears further in one direction than we can, and a few lightyears less in the opposite direction because they're also at the "center" of the universe from their perspective. Of course, we can't make use of this feature by say, having an observatory in Alpha Centauri because the time it would take for AC to send their data to us would be no less than amount of time it would take the light from that piece of the universe to reach us directly. But in as much as we can imagine a universal "now", AC will have a slightly different view of the universe than we do -- yet we're both still justified in claiming we're at the center, thus eliminating any fundamental concept of "center" beyond just calling it an observational bias.

    So back on topic, there's another couple of things that we can figure out:
    - There is stuff we cannot see. Anything beyond our past lightcone is forever lost to us unless the universe turns around and starts collapsing again. It may well be that the universe is larger, perhaps many many orders of magnitude larger than what we can observe. Its possible that our specific singularity is a minor fluctuation in some phenominally larger structure (Ekpyrotic []). But short of inventing FTL travel, we'll never be able to confirm that experimentally.

    - The universe will eventually disappear. This is a combination of the finite speed of light with eternal expansion. Eventually the expansion even between the nearest pair of "objects" will exceed the speed of light (Observable Universe []). At that point, there will simply be no universe left. I say "objects" as I'm not entirely sure at what scale gravity is able to overwhelm expansion and keep things held together. Definitely within a single galaxy, but to the scale of clusters and superclusters is something I'm less certain of. So the whole universe wouldn't disappear beyond the light cone horizon, just most of it. We're not at this point yet however -- we can see back to a time when the universe was completely opaque to light (Surface of Last Scattering []) and we'll need to develop instruments that measure gravitational waves in order to see back any further.

    - If expansion is speeding up as the big rip theory [] proposes, then it will eventually get to the point where the "force" of expansion exceeds gravity, then EM and finally the strong force, ripping everything apart even down to the subatomic level and there will literally be nothing left until the next big bang.

    To sum up, we can make a pretty good estimate of how far back our observable universe goes, but whatever might be outside of our observable universe is entirely up for grabs, and the only way we can ever investigate it is to discover FTL travel, which has a very good chance of being fundamentally impossible (basically, we'd need not only new physics, but new physics that can be applied to macroscopic objects such as probes or people.)

Practical people would be more practical if they would take a little more time for dreaming. -- J. P. McEvoy