CERN team claims measurement of neutrino speed >c

[ZapperZ]

That would not work for a lot of reasons, the main one of which is that the neutrinos tunnel right though the Earth in a straight line from Cern to the detector in Italy. There is no equivalent path for light, so the separation of the emitter and detector needs to known somehow. I'll have to dig into Opera faqs, etc to see how the distance was known well enough to measure such a small variation from c.

This is incorrect. The neutrinos DO NOT TUNNEL through the earth. They interact only via weak interactions (and very, very weakly with gravity). "Tunneling" is a different physics entirely!

Zz.

 

[turbo]

This is incorrect. The neutrinos DO NOT TUNNEL through the earth. They interact only via weak interactions (and very, very weakly with gravity). "Tunneling" is a different physics entirely!

Zz.

Noted. Please chalk this up as a poor choice of words. Neutrinos interact so weakly with matter that they can zip through (not tunnel through) impressive amounts of matter without leaving a trace of interaction. Thus, you need a big sensitive detector, and LOTS of neutrinos to get statistically-significant detection-signal. Apparently, Opera was designed with this in mind, and successfully so. Are the results reliable, and are they repeatable with other instrumentation? Time will tell.

 

[PeterDonis]

I'll be interested to see exactly how they calculated what the light travel time should have been. Did they properly account for the fact that the direct path goes through the Earth's interior, and therefore the actual path length will be different than the path length that would be inferred if you just took the differential, in Euclidean geometry, between the two GPS locations, because of GR effects (the difference in spacetime curvature)? My initial guess is that the corrected "through the Earth" path length will be slightly *shorter* than the uncorrected path length you would infer from the differential in GPS locations, which would explain the results. But I haven't done a calculation to see for sure.

 

[atyy]

Newly posted by MTd2 on marcus's quantum gravity bibliography:

http://arxiv.org/abs/1109.4897
Measurement of the neutrino velocity with the OPERA detector in the CNGS beam
OPERA
(Submitted on 22 Sep 2011)
The OPERA neutrino experiment at the underground Gran Sasso Laboratory has measured the velocity of neutrinos from the CERN CNGS beam over a baseline of about 730 km with much higher accuracy than previous studies conducted with accelerator neutrinos. The measurement is based on high-statistics data taken by OPERA in the years 2009, 2010 and 2011. Dedicated upgrades of the CNGS timing system and of the OPERA detector, as well as a high precision geodesy campaign for the measurement of the neutrino baseline, allowed reaching comparable systematic and statistical accuracies. An early arrival time of CNGS muon neutrinos with respect to the one computed assuming the speed of light in vacuum of (60.7 \pm 6.9 (stat.) \pm 7.4 (sys.)) ns was measured. This anomaly corresponds to a relative difference of the muon neutrino velocity with respect to the speed of light (v-c)/c = (2.48 \pm 0.28 (stat.) \pm 0.30 (sys.)) \times 10-5.

 

[DrGreg]

http://arxiv.org/abs/1109.4897
Measurement of the neutrino velocity with the OPERA detector in the CNGS beam

Note the final paragraph:

Despite the large significance of the measurement reported here and the stability of the analysis, the potentially great impact of the result motivates the continuation of our studies in order to investigate possible still unknown systematic effects that could explain the observed anomaly. We deliberately do not attempt any theoretical or phenomenological interpretation of the results.​

 

[Jack21222]

I'm kinda hoping that there is some sort of compact extra dimension explanation to come out of this (because my research advisor would do a literal jump for joy), but I recognize that this is far far FAR more likely to be just some experimental error.

 

[Runner 1]

Well one thing is for certain:

When the paper is released, we'll see a bunch of internet physics experts discover the obvious flaw that multitudes of particle physicists just happened to overlook during 3 years ;)

 

[seerongo]

Whether this is an error in methodology/measurement or it is verified that neutrinos are faster than photons and photons are slower than c and massive, etc., the outcome should be very interesting in any case. This group is not stupid and have had 4 years to figure this out. It seems to me that any outcome is bound to have important implications, even an experimental anomaly, since so many experiments are based on similar methodologies. Anyone here care to speculate on that end of it (since speculation is all we have today)? Comments here so far seem too focus on errors in measuring source/detector separation, equipment latencies, etc, but certainly they have gone over that ground ad nauseum.

For purposes of this discussion if nothing else, can we agree to differentiate the terms "speed of light" and "c", with "c" being the zero-mass SR speed limit? Using them interchangeably can be confusing in a discussion like this.

 

[Dale]

I don't understand why everyone in this thread seems to be assuming that a massive photon will explain this. The value of c is used in so many formulas used by physics. If we'd been using the wrong formulas since the early 1900s, wouldn't somebody have noticed their inaccuracy?

Newton's laws were used for about twice as long before anyone noticed any inaccuracies.

 

[Vanadium 50]

This is a systematic effect. You can take that to the bank.

They don't see a velocity dispersion. By itself, that's a huge problem. If you want to argue that not only are neutrinos faster than light, but they all travel at the same speed regardless of energy, you have to explain why the neutrinos from SN1987A arrived on the same day as the light did, instead of (as the Opera data would indicate) four years earlier.

 

[Vanadium 50]

A massive photon won't explain this. All photons travel at the same speed. If the limiting speed were 1.000025c, we would see more energetic photons move faster, and we don't.

 

[Dale]

A massive photon won't explain this. All photons travel at the same speed. If the limiting speed were 1.000025c, we would see more energetic photons move faster, and we don't.

Yeah, I have to agree here. I just looked at the paper, the effect is too large to have hidden in the noise for all previous experiments.

 

[Jack21222]

+1

I think there is an inverse relationship between the speed at which one dismisses other's works and the number of their own great works.

One can both dismiss and investigate a claim at the same time. I'd be very interested in seeing where the error is. For what it's worth, the CERN team is "dismissing" their own results here. It's still fun trying to pinpoint what could have went wrong.

The point about the SN1987A neutrinos is a big one. I just did the calculations myself... the neutrinos would have arrived 4 years earlier than they did, as V50 says.

 

[PeterDonis]

Differences in time dilation due to a slightly different gravitation field? Bzzzt, wrong, two identical Cs clocks, one at each location with a measured error of 2.3 ± 0.9 ns.

As I read it, the clocks are synchronized using GPS. Just having them identical doesn't account for time dilation if they are at different altitudes (i.e., different levels of gravitational potential), which I believe they are. There has to be some mechanism for correcting their rates to a common standard of simultaneity. That's what the GPS part is for (and it looks like it requires pretty hefty GPS equipment to get that kind of accuracy for the corrections).

Also, I see very precise measurements of distance, but they are all based on GPS location fixes, as far as I can tell. I see a reference to a "common analysis in the ETRF2000 reference frame", but there are no details, just a pointer to a reference at the end of the paper that isn't online. So I can't see if the reference frame they used for their computation of the distance, based on all the measurements, took into account that distance, as well as time, gets distorted when the altitude (i.e., gravitational potential) changes. I would think it would, since they talk about a geodetic survey, which is all about accurate measurements of equipotential surfaces. But it would be nice to have more details.

 

[PeterDonis]

The point about the SN1987A neutrinos is a big one. I just did the calculations myself... the neutrinos would have arrived 4 years earlier than they did, as V50 says.

The paper mentions SN1987A, and notes that the energies of those neutrinos were several orders of magnitude smaller than those of the CERN neutrinos in this experiment. So one could try to account for the SN1987A results and these consistently by postulating a really wacky dispersion relation for neutrinos, that caused virtually no dispersion at energies around the SN1987A energies, but yet caused significant dispersion at the CERN neutrino energies. I don't know if any reasonable physical models for neutrinos would imply such a dispersion relation.

 

[Dale]

They did not measure the speed of photons moving between the same points at the same time. They measured neutrinos to go faster than *c*.

Correct, they are claiming the the neutrinos travel at 299 799 893 m/s compared to the speed of light 299 792 458 m/s. So the massive-photon resolution would require that the invariant speed be something greater than 299 799 893 m/s, but that would have been detectable in other experiments.

 

[Haelfix]

Correct, they are claiming the the neutrinos travel at 299 799 893 m/s compared to the speed of light 299 792 458 m/s. So the massive-photon resolution would require that the invariant speed be something greater than 299 799 893 m/s, but that would have been detectable in other experiments.

The bottomline is there is no way this can work theoretically. You could for instance look for departures from Lorentz invariance, but that's already been searched for ad naeusum through several different channels. No known violation of the SR dispersion relations have ever been discovered, and the bounds are already far in excess of the sensitivity of this experiment.

It also contradicts well established neutrino measurements, like the Supernova ones. Trying to stay consistent with that, leads you into real absurdities (like modifying standard MSW physics in violent ways)

 

[seerongo]

Correct, they are claiming the the neutrinos travel at 299 799 893 m/s compared to the speed of light 299 792 458 m/s. So the massive-photon resolution would require that the invariant speed be something greater than 299 799 893 m/s, but that would have been detectable in other experiments.

Thanks for clarifying that.

So now that the paper clears that up, it appears the photon can resume it's original svelte, speedy status as "c". This sure is starting to look like an error in position measurement. Not as much fun, but still would be important, since they must have been closely studying that possibility all along.

I suppose the little ones could be taking an extra-dimensional short cut or a convenient worm hole, but they'd all have to be taking the same short cut every time for years. I dunno...

Thanks to those who are summarizing the paper's details for us non-physicists.

 

[qsa]

http://indico.cern.ch/conferenceDisplay.py?confId=155620

 

[Chronos]

Agreed, the OPERA team is seeking confirmation [I agree with Pallen it appears unlikely]. Neutrino detection is tricky business and correlating capture with emission is no easy task. I can't help but wonder how many of the detected neutrinos were actually emitted by CERN and how that might skew the measurement. There was a paper about 10 years ago about neutrinos as tachyons by Chodos, IIRC.

 

[seerongo]

Agreed, the OPERA team is seeking confirmation [I agree with Pallen it appears unlikely]. Neutrino detection is tricky business and correlating capture with emission is no easy task. I can't help but wonder how many of the detected neutrinos were actually emitted by CERN and how that might skew the measurement. There was a paper about 10 years ago about neutrinos as tachyons by Chodos, IIRC.

My background is engineering, not physics, but frankly, the method used to correlate the proton extractions with the v detections doesn't seem that bad to me so far, although at first blush, 16,111 detected events doesn't seem too great statistically. I'd like to see more expert comments on that however. Regarding potential contamination, would most contamination come from B-decay, which would be anti's? I think they accounted for anti's, counting about 2% unless I read it wrong. Could someone comment on that? I'm not sure about the potential sources of spurious neutrinos in significant numbers.

I'm more struck by the timing aspects. There seem to be so many places in this system where inaccuracies can gang up on you. This is a pretty complex system with a lot of timing points, all with tolerances. I'd be the last person to second guess this work, but I think that's where I'd look.

 

[Gilles]

I think the question of clock synchronization may be tricky. In GR, there is no absolute definition of simultaneity. Due to differences in gravitational potential, as mentioned, clocks evolves differently at different points. So you must periodically resynchronize them, but how ? there is no unique choice, and the measured time of flight probably depends of how you define the timescale at each point.

 

[Parlyne]

There is one remote possibility I have not seen discussed in this thread.

Is it possible in theory that a neutrino has zero mass and the test is showing tachyonic properties? This would not violate SR.

An unlikely explanation but just wanted to see what an expert has to say.

I'm afraid that's already ruled as a reasonable explanation for this by supernova 1987A. The problem is that the speed of a tachyon is given by

$$v = c\sqrt{1+\frac{|m^2|c^4}{E^2}}$$.

This means that a tachyon's speed increases as its energy decreases. As noted above, the OPERA neutrinos have higher energy than the 1987A neutrinos, meaning that, were they tachyonic, they should be slower, not faster, than the supernova neutrinos. But, in fact, the 1987A neutrinos have a discrepancy from c that is, at worst, something like 4 orders of magnitude smaller than the OPERA discrepancy.

 

[JordanL]

I'm sorry, but I don't quite buy a slight tweak to our definition of "c" as a complete answer.

Anyone suggesting that simply adjusting the "c" constant will fix things needs to explain how 150 years of mathematical and physics equations didn't detect the discrepancy.

Looking to an adjustment to c as the answer to this data, if correct, is... creative. It is not borne out of a dedication to science but a fear of change, as given data like this, that is certainly not the most likely cause, even within our CURRENT theories.

 

[Dmitry67]

What if neutrinos are very high energy tachions, so we never noticed that they are moving slightly faster than c? We can't detect low energy neutrinos, so usually don't see them moving much faster than c.

 

[JordanL]

What if neutrinos are very high energy tachions, so we never noticed that they are moving slightly faster than c? We can't detect low energy neutrinos, so usually don't see them moving much faster than c.

I believe someone else already asked this question. The answer that was given is that tachyons decrease in energy as they increase in speed. Using the neutrinos detected from the referenced supernova, were they tachyonic, the neutrinos should have been traveling even faster than the ones CERN is talking about. Instead we saw them arrive simultaneously with the photons.

 

[Demystifier]

The idea that neutrinos are tachyons is not new. Many papers already exist:
http://xxx.lanl.gov/find/hep-ph/1/ti:+AND+tachyon+neutrino/0/1/0/all/0/1

 

[TrickyDicky]

It is reasonable that after this kind of announcements people starts getting nervous and all kind of silly things are said. Maybe it is not so normal that knowledgeable people first reaction to this apparently "FTL neutrinos" be that SR must be modified, or everything that was measured so far to a certain accuracy is now wrong. It is not. Let's listen to Vanadium 50 here.
First thing to rule out is obviously some kind of error in the measurement, and this is explicit in most posts. Even if no measurement error is found, we must first look for explanations that are compatible with the accuracy level of thousands of previous experiments that can't just be ignored.
So far little attention has been focused to the special nature of the subject particle, the neutrino and the way it is measured, I would say that this is the weakest link of the chain if no obvious claculational or silly error is found so I think the first serious theoretical searches must come from this side rather than question relativity.

 

[Chi Meson]

This is a systematic effect. You can take that to the bank.

They don't see a velocity dispersion. By itself, that's a huge problem. If you want to argue that not only are neutrinos faster than light, but they all travel at the same speed regardless of energy, you have to explain why the neutrinos from SN1987A arrived on the same day as the light did, instead of (as the Opera data would indicate) four years earlier.

Thank you for posting this! I was pouring through all this info, with this same obvious fact in mind, wondering what I missed. The neutrino burst is part of the standard method for studying Type 2 supernovae in other galaxies, and they all arrive, exactly according to precise calculations, after the light gets here. So granted there's plenty I don't know or understand about the data, but place me in the camp that thinks a systematic error is to blame, rather than derailment of SR.

But hey, I'm a good little scientist--I'll leave the door open.

 

[MarcoD]

Yes, it would be a big problem.
The problem here theoreticians don't seem to make up their mind what speed neutrinos should travel at, when they were supposed to be massless they were expected to have light speed, and supernovae detections so seemed to verify, when agreement was reached that they had mass they obviously should be slower than c, but as Demystifier pointed out there were several people that hypothesized that they should be FTL.
One has to wonder what they really are all measuring, is it really neutrinos? Is there a serious agreement about what its speed should be?

The fact that particles arrive at the same moment from supernovae is a compelling argument it is a fluke, unfortunately. Success to you all.

 

[suprised]

To be sure: it is not CERN who is claiming this, but a team outside of CERN, and all what CERN does is to provide a platform for today's press conference. Unfortunately so, and many colleagues strongly object this. Of course this is being mixed up all over in the media, as usual. Incidentally neither the General Director nor the research director will be present.

 

[xts]

Seems that I am only one here who bothered to read OPERA preprint: http://arxiv.org/ftp/arxiv/papers/1109/1109.4897.pdf

Just some points after reading:

1. There is no information what reference frame they use for analysis and how they covered relativistic effects in their analysis:
CERN? Gran-Sasso? Centre-of-Earth? Solar System?
Please note, that SR time dilation between CERN and Gran-Sasso frames is 10 times stronger than the effect they report. How the clocks were corrected for dilation?
There is also no information if GR effects were taken into account.

2. There is no discussion about systematic errors which may be caused by delays in readout electronics and scintillators itself (except of light propagation, which is the only one discussed). The systematic error caused by DAQ and detectors is estimated as for few ns each, which seems to be too optimistic.

3. Detailed experimental setup is delegated to other paper not available online.

 

[dgwsoft]

There would be a race to determine the mass of the photon. It would be a huge surprise, but I think it would be a bigger hit for QED than SR or GR - the latter rely only on the fact that there is a spacetime structure speed limit. Whether a particular particle reaches it is irrelevant.

I would definitely take the bet against this being confirmed.

My first thought was perhaps photons do no travel at "the speed of light", ie photons have (rest) mass.

According to wikipedia http://en.wikipedia.org/wiki/Photon#Experimental_checks_on_photon_mass the experimental limit is at least as good as m < 1e-14 eV/c^2

I could not find a formula to convert photon mass into speed, but I think I have worked it out:

(v/c) = SQRT( (1+d^2)/(1+2d^2) ) where d = Lmc/h (L = wavelength, m = photon rest mass, c = "cosmic speed limit for which we need to find a new name", h = Planks constant).

For small d this approximates to v/c = 1 - d^2/2

Using the mass given above and for a green photon of wavelength 500nm that comes out as one part in about 10^30, much smaller than the 20 parts in a million quoted for the neutrinos.

To look at it the other way, for a photon to be traveling 6000m/s slower then true "c" would require it to have a rest mass of about 1.5e-2 ev/c^2 which would have been noticed.

However my SR is a bit rusty so if anyone wants to check this I would be grateful.

(AIUI it is not significant that light is observed to travel "at c" because since there is no evidence (as yet) that photons have mass, we have just taken "c" to be the speed of light).

 

[Vanadium 50]

The paper mentions SN1987A, and notes that the energies of those neutrinos were several orders of magnitude smaller than those of the CERN neutrinos in this experiment. So one could try to account for the SN1987A results and these consistently by postulating a really wacky dispersion relation for neutrinos, that caused virtually no dispersion at energies around the SN1987A energies, but yet caused significant dispersion at the CERN neutrino energies.

It's even wackier than that. You have to argue that you have no velocity dispersion at the 10^-10 level or so for MeV neutrinos that vary by a factor of ~3 in energy, and no velocity dispersion at the 10^-6 level or so for GeV neutrinos that vary by a factor of ~3 in energy, but between those two energies the velocity changes by 25 x 10^-6.

There is no information what reference frame they use for analysis and how they covered relativistic effects in their analysis:

It's not relevant. Essentially what they are doing is measuring the Lorentz-invariant interval between the production and detection of the neutrinos, and comparing that to a null interval. Since interval is a Lorentz invariant quantity, it doesn't matter what frame they worked it in.

2. There is no discussion about systematic errors which may be caused by delays in readout electronics and scintillators itself (except of light propagation, which is the only one discussed). The systematic error caused by DAQ and detectors is estimated as for few ns each, which seems to be too optimistic.

If the experimenters are competent, this is easy to do, and as such not worth much space. You get the electronics timing by checking the time difference between input and output on a scope. The detector timing is a little trickier, but signal formation time for plastic scintillator and even a slow phototube is a few nanoseconds. Timing in the detector relative to itself to 1-2 ns is commonplace.

 

[TrickyDicky]

It's even wackier than that. You have to argue that you have no velocity dispersion at the 10^-10 level or so for MeV neutrinos that vary by a factor of ~3 in energy, and no velocity dispersion at the 10^-6 level or so for GeV neutrinos that vary by a factor of ~3 in energy, but between those two energies the velocity changes by 25 x 10^-6.

Exactly. So I would agree with the cautious Susan Cartwright, senior lecturer in particle astrophysics at Sheffield University when she says "Neutrino experimental results are not historically all that reliable, so the words 'don't hold your breath' do spring to mind when you hear very counter-intuitive results like this."
Most likely they didn't measure what they thought they were measuring.