Scientists Speed up Light

An anonymous reader writes "With off-the-shelf components, scientists have managed to speed up light beyond the 'universal' constant of c, or roughly 300 million meters/sec. This, and the previous ability to slow light down could shake up the telecom world, according to the story at Science Blog." Also, all those posters with 186,000 miles per second as a speed limit need to be amended. At least entropy is still around!

Comments

Overhyped as always

[trip11]

Everyone say it together with me: "Phase velocity vs Group velocity" There are no photons in this experiment that are traveling faster than the speed of light. Only collections of them that 'appear' to be doing so. Think of this as an example: I space people out in a line, each of them two light minutes apart from the people next in line (all at rest with respect to each other). Now I go about talking to them and informing them of my plan. At 12:00 the first person waves, at 12:01 the second person waves, at 12:02 the third person waves, and so forth. My "wave" is propogating, therefore, at twice the speed of light. This is the same thing that this experiment is doing more or less. By spending extra time setting up the experiment, you can make it appear that a light pulse travels faster than c, but like my "wave" it is only an appearance.

Re:Overhyped as always

[Anonymous Coward]

Information transfer *is* what's limited by c. It then *follows* that a particle cannot travel faster than light, but that's a simple case. In general the limit applies to an experiment only if that experiment could be used to transfer information.

Re:Overhyped as always

[lgw]

There are some experiments in which photons are travelling faster than "the speed of light", because c is defined in a vacuum, and a vacuum is not the lowest impedance available.

Even in a vacuum, light doesn't travel as photons for the entire journey (at least, if you believe in quantum). Light spends some of its time as electron-positron pairs which exist very briefly, before annihilating to product a new photon. As the electron-positron pair travels slower than the speed of light, light in a vacuum (which is how we've defined c) travels slighty slower that the speed of a photon.

When you shine a light between very closely spaced conductive plates, that reduces the available "wavelengths" of the electron-positron pairs (I don't like that terminaology, but it makes the temporary electron-positron pairs less likely to occur), so the light spends more time as photons. Therefore light is travelling faster than "the speed of light".

But not really, it's just that c is standardized on the wrong empirical constant. What you care about is the speed of photons, not the speed of light in a vacuum.

Re:Overhyped as always

[exp(pi*sqrt(163))]

Don't explain it. Show it!

Re:Overhyped as always

[LionMan]

Other posters have already stated this:
information can not travel faster than the speed of light.
The fact that phase velocity can be faster than c, as this article points out (which has been known for a long time! read about anomolous dispersion. we've known about that for a long time now.) can _not_ improve telecom by speeding up information transfer. Advanced techniques (better fiber optics, optical routers, etc) which still abide by the c speed limit are the only way to reduce your ping time.
Anyway, the current bottleneck is not the fiber part of telecom. The optical-electronic interface and the electronic switching is the real culprit. Once optical switching and routing is prevalent, then more technology spent on optics will really pay off.

Re:Overhyped as always

[Guppy06]

"Why does it matter if a signal can get someplace faster than it could have gotten there via photons?"

Special relativity.

"If I send a message from A to B and it gets there instantaneously, and B sends a message back to A a fixed time later, it will be received by A after A sent the first message. No time travel."

No, B's signal will get there before A sent the original instantaneous signal. It is not a matter of "B's clock looks like it's behind A's," it is that B's clock is behind A's, and that instanatneous signal will actually arrive in B's (and A's) past. What you're doing is assuming a preferred frame of reference, that one station's measure of time is more valid than the other's, and special relativity says that cannot be.

"The only causality broken would be that A could tell B "there's some light headed your way, it's going to show you our sun going nova, the light should get there in about 4 years,"

No. If A was 4 light-years away, B would get the signal 8 years before the nova got there, or 4 years in the past. A, looking at B, would see B's calendar and see that B is 4 years behind (and it's not that B "appears to be" 4 years behind, otherwise it would be possible for A and B to get different measurements for the speed of light), so that instantaneous signal would reach B 4 years ago.

You're assuming A's reading of A's clock is more valid than A's reading of B's clock, and that's not allowable in special relativity.

"The pulses are encoded to distinguish one from the other (e.g. a time stamp as in NTP). When the pulses come back to A, A knows precisely how far apart they are, and so can send out a pre-arranged signal, and then (at the appropriate time) do something "simultaneously" with the other station (of course, B can also calculate the time difference, so they both know what time it is "now" on the other station). If there's a third station C, also at exactly the same time rate as the other two, is there any way that C won't be able to be in synch with both A and B (i.e. getting a consistent time difference for the two)?"

Time and space are not constant, only the speed of light is constant. Time and space change in relation to each other to maintain that constant ratio. According to C, moving at relativistic speeds, the distance between A and B is different than what A and B measure. And while A and B may see C as being half-way between them, C, moving towards A, will see itself as being closer to A than to B. Similarly, while A and B may believe they are in synch with each other, C, moving towards A, will see A's clock ticking faster than C's, and B's clock ticking slower than C's.

As in the barn and the pole "paradox",, if C is going at relativistic velocities towards A and away from B, A's signal will reach C before B's does, so they will not be simultaneous.

A's and B's clock interpretations cannot be "more right" than C's interpretation, or otherwise A's and B's measurement of the speed of light would be "more right." However, even if C is moving at half the speed of light towards A (accordin to A), A's light-based signal will reach C (according to C) at speed c (not 1.5c), and B's light-based signal will reach C at speed c (not 0.5c).

"Why can't A, B and C all "go off" simultaneously?"

It can only happen if A, B and C are all at rest with respect to each other. Otherwise, time dillation throws the clocks out of psynch. Once C moves, C's clock will start to tick slow (according to A and B), and even after C comes to rest again, C's clock will still be behind.

repost?

[rkruse]

Hasn't this already been done before?

Bet you any money...

[gowen]

... it's "only" the phase velocity. This has been done before, and, since information is carried at the group velocity, there aren't any serious "light-cone" repercussions for Einsteinian limits on causality.

Nothing too new...

[Space cowboy]

There's more than one measure of the speed of light - the phase velocity and the group velocity. It's the group velocity that can't travel faster than c, the phase velocity is free to travel faster assuming dispersion is allowed. In any event, information travels at the speed of the group velocity, which is why the write-up mentions that Einstein ain't wrong just yet ("only a portion of the signal is affected").

If you look at this treatment of wave velocity, it's reasonably clear ([grin] - at least if you've done undergrad physics, but then in that case you'd know all about it anyway :-)

A good quote from the above link:

Unfortunately we frequently read in the newspapers about how someone has succeeded in transmitting a wave with a group velocity exceeding c, and we are asked to regard this as an astounding discovery, overturning the principles of relativity, etc. The problem with these stories is that the group velocity corresponds to the actual signal velocity only under conditions of normal dispersion, or, more generally, under conditions when the group velocity is less than the phase velocity. In other circumstances, the group velocity does not necessarily represent the actual propagation speed of any information or energy. For example, in a regime of anomalous dispersion, which means the refractive index decreases with increasing wave number, the preceding formula shows that what we called the group velocity exceeds what we called the phase velocity. In such circumstances the group velocity no longer represents the speed at which information or energy propagates.

The phenomena is also discussed in Feynman's Lectures on Physics ( vol 1, Chapter 48-6) in a bit more rigor - these books ought to be required reading of any physics undergrads :-)

Simon

Re:Nothing too new...

[Geoff St. Germaine]

Good post. I recall a lecture I had from a PhD from Los Alamos when I was doing my undergraduate degree about the group velocity exceeding c, but they could still not transmit information at that velocity. The information velocity isn't the phase velocity, but it isn't necessarily the group velocity either.

Re:Nothing too new...

[Space cowboy]

[grin] not really. You need a reasonable grounding in wave theory before you get to phenomena like standing-waves (eg: a string attached at one end, and agitated at the other) or superposition (eg: the "beating" sound of two similar-frequency sounds) and group/phase velocities are slightly farther on than that.

Let's try though: Imagine a slightly-complicated (3 ups and downs) wave in your head (or on paper), now repeat it three times - add the same wave to the start and the end of the original. You ought to see a sort of symmetry - three complicated waves (which are very self-similar) one after the other. Let's assume this is a wave travelling through space from A to B.

[aside: You also need to know that any complicated wave can be decomposed into a bunch of simple sine waves (at different frequencies), all superimposed on top of each other. Physicists call the simple sine waves the component frequencies of your wave]

The speed of information (group velocity, under normal conditions) is determined by the speed at which those 3 groups (hence the name :-) of waves arrive at the receiver. When the medium through which the wave is travelling has a constant refractive index [wave theory thing, just accept it as a property of the medium for now], the group velocity is equal to the phase velocity.

However, when the wave travels through a transparent medium (water, glass, transparent aluminium (!), etc.), the refractive index tends to change slightly with frequency. This is why different frequencies of light are split when going through a prism. In this case, the group velocities of the different colours of light are lower than c because of the refractive index of glass.

But, you say, here the group velocity is *higher*, well, the group velocity itself is usually a function of the wave's frequency, and you can create media with exotic refractive indices (this is the province of non-linear optics). Both of these can result in group velocity dispersion for different component-frequencies of the wave. The result is that the 3 waveforms in your head smear over time as a result of different frequency components of the pulse travelling at different velocities on their path from A to B.

So, now consider your 3 waves after they've been travelling for a certain time T. They now overlap in space as different frequencies from each of your 3 starting-waves travel at different speeds to the destination, so individual frequency-components (which ones depends on the refractive index) of the wave can arrive faster than c at the receiver. This is what the write-up meant when it said that only a portion of the signal is travelling faster than c. Crucially however, each one of the 3 waves does *not* travel faster than c as a whole, and in fact almost always travels slower.

At least, I rather hope the above is correct - I've not read or used any of this stuff for ~15 years :-)

Simon.

Don't have to change the constant

[Azarael]

When people have 'c' recorded, it's assumed that it's referring light in a vacuum and it's not messed around with. So the values can stay the same.

Cesium Chamber Experiment from Before

[vectorian798]

Is this really that new? This has happened before. Read here: CNN: Light can break its own speed limit

And before we all start yapping, I quote from the (CNN) article:

This effect cannot be used to send information back in time," said Lijun Wang, a researcher with the private NEC Institute. "However, our experiment does show that the generally held misconception that `nothing can travel faster than the speed of light' is wrong.

Not quite

[pauljlucas]

All 4 basic forces: electromagnitism, gravity, strong nuclear, and weak nuclear ... forces propogate at the speed of light in their reference frame.

They propagate at the speed of light in all reference frames, i.e., the speed of light is the same to all observers.

(However, including the nuclear forces is moot since they have no influence nor can they be observed outside the nucleus of an atom.)

Re:Not quite

[Anonymous Coward]

They propagate at the speed of light in all reference frames, i.e., the speed of light is the same to all observers.

In all inertial reference frames. That is (in short) the ones that are not accelerating or decelerating.

Re:Not quite

[ultranova]

(However, including the nuclear forces is moot since they have no influence nor can they be observed outside the nucleus of an atom.)

Really ? I can see the Sun shining just fine.

Perhaps you meant that they can't be directly observed, only indirectly by the way of their consequences ? But surely you realize that this is true for all forces except electromagnetic - even with the proposed gravity sensors, you can't actually see the gravity waves, you can just see weights moving.

Anyway, including nuclear forces is extremely important, because if a nuclear force that propagates faster than c, you could simply arrange multiple nuclear particles into a line - meaning that faster-than-light communication would be just an engineering problem, not one that requires violating basic physics.

Re:Not quite

[ultranova]

Creating stable wires made out of neutronium is an "engineering problem" which will require a whole lot of new basic science to accomplish!

Undoutedly. But "very very difficult" is completely different than "impossible".

You can not violate basic laws of physics; if you can, then they weren't basic laws of physics, you just thought they were.

You can do anything not expressly forbidden by basic laws of physics; it is just a matter of doing a lot of research first.

That's the difference between "engineering problem" and "violation of laws of physics": engineering problems can be solved by throwing enough money at them, but the laws of physics can't be bribed.

Re:A question

[drstock]

You can't use newtonian physics for speeds so close to c. Newtonian physics states that you just can add speeds, ie x = y+z.
This isn't correct which gets noticable when speeds approaches the speed of light. Instead use relativistic physics: x = (y + z)/(1 + y*z/c^2).

So your example becomes:
v = (0.75c + 0.75c)/(1 + 0.75c*0.75c/c^2) = 0.96c

Link to Actual Paper

[statemachine]

I'm not sure if anyone already posted the actual paper. ScienceBlog only links to itself and references a future printed publication. Well, here it is:

http://www.opticsexpress.org/abstract.cfm?URI=OPEX -13-1-82

Domino block analogy

[Alwin Henseler]

Set up say, 1000 domino blocks in a row. Then tip the first one over. Given constant size, weight, spacing of individual blocks, and a horizontal surface, you will observe blocks falling down at a constant rate/speed ('c'). Given that constant rate/speed, tipping over the first block will cause all blocks to fall down, tipping over the last block some time later. Time delay calculates as distance divided by 'c'.

Now, create 'extreme conditions', where the first domino block is down, the last one is still standing, and halfway down the row, blocks are falling, but not quite down on the floor. Then, observe the 'wave front' of falling domino blocks. It will appear to move faster than the previously determined 'c'. How come?

Look more closely: as each block falls down, there's a fixed delay before it hits the next block. But what happens under our 'extreme conditions'? At the exact time a previous block would have hit the next one (under normal circumstances), that next block is already falling down! The time it takes for the 1000 blocks to fall down, is less than what normally would be expected.

Did this 'c' constant get violated? Nope, it still took the same amount of time for each block to fall down. Was the maximum 'c' speed exceeded? Nope. After tipping the first block, it still took the same amount of time before this 'information' was passed on to the next block. With a set of 1000 blocks all standing, the time needed for an initial 'disturbance' to be passed on to the last block, is still limited by 'c'.

So these 'extreme conditions' are like pre-tipping each block, and let you observe something that appeared to move faster than 'c'.

Nice for the lab folks, but other than that, sensationalist journalism. Wake me up when trans-atlantic ping times (sending actual packets with random data) dive below the time dictated by the speed of light.

Re:Domino block analogy

[NidStyles]

What you just described would explain a slower than (c) speed, not a faster than (c) speed.

Re: Information transfer *is* what's limited by c

[falconwolf]

My impulsiveness got the best of me. Someone else on this thread asked me to read Wikipedia's article on quantum entangement. When I did I found this:

Entanglement produces some interesting interactions with the principle of relativity that states that information cannot be transferred faster than the speed of light. Although two entangled systems can interact across large spatial separations, no useful information can be transmitted in this way, so causality cannot be violated through entanglement. This occurs for two subtle reasons: (i) quantum mechanical measurements yield probabilistic results, and (ii) the no cloning theorem forbids the statistical inspection of entangled quantum states.

NOTHING goes faster than the speed of light. Period.

If you're talking about the speed of light in a vacumn then this may interest you. Scientists have apparently broken the universe's speed limit. Sending a laser light through cesium vapor they were able to beat the speed on light in a vacumn. Farther on it says:

"This effect cannot be used to send information back in time," said Lijun Wang, a researcher with the private NEC Institute. "However, our experiment does show that the generally held misconception that `nothing can travel faster than the speed of light' is wrong."

Falcon