Nature | Nature News Blog

Neutrino experiment affirms faster-than-light claim


18 November 2011: An updated version of this story is now available (see ‘Neutrino experiment replicates faster-than-light finding’).

It is a remarkable confirmation of a stunning result; but most physicists remain skeptical. That seems the most probable outcome of a release of new data expected on 17 November from researchers with the Italian OPERA collaboration, who say they have confirmed their controversial finding that flighty subatomic neutrinos can travel faster than light.

“It’s slightly better than the previous result,” says OPERA’s physics coordinator Dario Autiero of the Institut de Physique Nucleaire de Lyon in France (pictured). He adds that most of the members of the collaboration who declined to sign the original paper because they wanted more time to check the result have now come on board. One of these is Caren Hagner of the University of Hamburg in Germany. She says not only has the experiment’s precision been improved, the statistical analysis is more robust and has been replicated by different groups within OPERA not just the original team. “We gained much more confidence,” Hagner says.

OPERA (which stands for Oscillation Project with Emulsion-tracking Apparatus) made headlines in September with a claim to have clocked neutrinos traveling faster than light, a result at odds with Albert Einstein’s well-established Special Theory of Relativity, which sets light as the ultimate cosmic speed limit. The group used a pulsed beam of neutrinos produced by a particle accelerator at CERN near Geneva which traveled some 730 kilometers to Gran Sasso near L’Aquila, Italy where the particles were detected.

The result was highly statistically significant but following Carl Sagan’s well-known mantra that “extraordinary claims require extraordinary evidence”, most physicists expressed doubts. While OPERA appeared to have conducted its data-taking and analysis carefully, there was rampant speculation about possible sources of error and some made claims of mistakes that the collaboration brushed off.

One set of concerns centered on the relatively long timescale – 10.5 microseconds, or 10.5 millionths of a second – of the proton pulses produced at CERN that result in the neutrino pulses OPERA detects. OPERA did not know whether individual neutrinos received at Gran Sasso corresponded to protons early or late in the proton pulse, creating uncertainty around their detection of them. In October OPERA therefore asked CERN to generate shorter proton pulses lasting just 3 nanoseconds. They have now recorded 20 events in the new data run and say that they have reached a similar level of statistical significance to the first time around, with the neutrinos again reaching Gran Sasso 60 nanoseconds faster than a light beam would do.

OPERA expects the new result to rule out uncertainties due to the long timescale of the proton pulses. But concerns about the experiment’s use of the Global Positioning System to synchronize clocks at each end of the neutrino beam are unlikely to be as easily allayed, The use of GPS is novel in the field of high energy and particle physics and the same system was used for both the original experiment and the new run. Hagner also adds that she’d like to see the time measurement checked using another part of the detector, to increase confidence further.

For most physicists outside the collaboration, however, the key test will be replication by an independent experiment. The one best placed to independently confirm or refute OPERA’s result is MINOS (the Main Injector Neutrino Oscillation Search) at Fermilab in Batavia, Illinois. In response to the latest OPERA result, MINOS issued a statement saying it is upgrading its timing system to match OPERA’s precision and might have preliminary results obtained using the existing system that are relevant to assessing OPERA’s results as soon as early 2012.

“OPERA is to be congratulated for doing some important and sensitive checks but independent checks are the way to go,” says Rob Plunkett, co-spokesman for MINOS.


  1. Report this comment

    Aaron said:

    Hi, shouldn’t “10.5 microseconds, or 10.5 thousandths of a second” be millionths instead?

  2. Report this comment

    Jorge Stolfi said:

    As far as I know, the most serious doubts expressed about the previous announcement were about possible systematic (as opposed to random) errors in the two key measurements: total time and travel distance. From the summary above, it seems that both are still being measured in the same way. So, what has changed?

  3. Report this comment

    John Costella said:

    MINOS as it now exists can only check the 730 km trip from CERN to Italy, not the 18-metre (60-light-nanosecond) trip across the iron “hadron stop” at the end of the decay tunnel at CERN, which may be at the heart of the result. This is because MINOS uses a matched near detector / far detector layout, whereas OPERA measures from the original protons (which are “upstream” from the hadron stop).

  4. Report this comment

    gregorylent said:

    time for science to drop its avoidance of mysticism …

    mystics simply laugh at the conceit that “nothing can go faster than the speed of light” … they know better.

    sure, proof, the scientific method, ta-da, but how else to open up to larger realities than to get out of the self-imposed, and fundamentalist, box?

  5. Report this comment

    Nyjah Huston said:

    Maybe particles without charge, like neutrinos, can go faster. Maybe Relativity only applies strictly to particles with mass and charge. Time will tell.

  6. Report this comment

    Bruce J. King said:

    What is the detector’s event-to-event time jitter relative to the universal time clock? It was disappointing not to find this in any of OPERA’s papers, although I may simply have missed seeing it. With scintillator readout, timing should presumably be at the nanosecond level and so should be good enough to resolve, in the neutrino event sample, the bucket structure from the proton beam. Finding this structure would strengthen the experimental result, both statistically and systematically, and enable further checks to be performed. Has OPERA attempted to do this? Thanks.

  7. Report this comment

    leotec said:

    mmm we need stay attent to these results next weeks

  8. Report this comment

    Ed Gerstner said:

    It’s not a ‘remarkable’ confirmation, it’s just a confirmation. A remarkable confirmation would be if the used a different method of calculating the timing, or a different baseline, or detected Cherenkov emission from the beam.

    Imagine you have a set of lottery numbers. You don’t expect to win. But the first number called matches the number on your ticket. That’s consistent with your eventually winning. But winning is still unlikely. Celebration is certainly premature.

    The latest result is akin to blinking to check that you’d seen the first number drawn correctly. That’s a long way from remarkable.

    More remarkable would be if the next number drawn matches another of your numbers. It still doesn’t mean you’ve won. But it’s a hell of a lot more significant than blinking.

    In this analogy, the next ball to drop will be the MINOS experiment. If that gives the same result, then we can start using words like ‘remarkable’.

  9. Report this comment

    Peter Cozijnsen said:

    It’s San Grasso, not Gran Sasso….

  10. Report this comment

    Ben said:

    reliable measurements of two mechanical slightly differing parameters must be made simaltaniously.

    The results of both experiments show that the method is wrong for the aim of relativity refuting.

  11. Report this comment

    abadidea said:

    gregorylent: What in the?

    “Mysticism” is what held us back since the dawn of civilization. Ditching it for the scientific method once and for all is why we suddenly sprung forward so much in the past few hundred years.

    I hope you’re trolling and not actually suggesting that scientists should use “feelings” instead of “evidence”.

  12. Report this comment

    jessica said:

    Yes! I do believe we need to stay tune to the results

  13. Report this comment

    Krish Munot said:

    Will be particularly interested to see Fermilab’s results in several months.

  14. Report this comment

    Krish Munot said:

    Will be particularly interested to see Fermilab’s results in several months.

  15. Report this comment

    Joseph Kulandai said:

    People doubted on the methodology of time measurement which has affected velocity calculation. Gravity’s effect on clock on different location etc etc..

    Is there any concrete explanation on this?

  16. Report this comment

    Sandy Stewart said:

    Could it be that neutrinos oscillate in velocity from faster than than light © to slower than c? Perhaps they travel at an average c only over long distances, so that c is still the basic limit. Thus the Opera result could be seeing only the faster part of the cycle.

    Not my idea — Asimov used this in his novel Nemesis.

  17. Report this comment

    Robert Lee said:

    It seems to me that the sensible follow-up would be to measure the speed of light using as much of the same locations, equipment and methodology as possible. That would also be the way to (dis)prove the GPS theory.

    What am I missing?

  18. Report this comment

    John Lo said:

    I think the question of the concerns about the experiment’s use of the Global Positioning System to synchronize clocks at each end of the neutrino beam can be solved by measuring the speed of low energy neutrinoes. There is no problem that the speed of low energy neutrinoes should be that of light. If the speed of low energy neutrinoes is lower than the speed of high energy neutrinoes by measurement using the same system. Then the concern about the synchronization problem should be ruled out.

  19. Report this comment

    Ben said:

    Opera`s experiments shows us political physics in action.You publish any idiotic comments but not the “political physics”` comments. Knowing your Cavalli-Sforca games I don`t wonder of your physical game.

  20. Report this comment

    theothercliff said:

    Has the “light slow down travelling through a medium” issue with the air moving at the rotation speed of the earth been taken into account with the GPS signals?

  21. Report this comment

    Dapeng Qian said:

    OPERA neutrinos are “pseudo superluminal”

    According to a new special relativity that has absorbed the uncertainty principle, I have propose a formula for the measurement of neutrino velocity of OPERA. The formula indicates:

    (1) When the energy E=42.9GeV and the early arrival time δt=67.1ns, the neutrino velocity can be calculated as u=(1-1.6×10-40)c, so still does not reach the light speed although u very close to c. Since the neutrino speed is lower than the light speed, why is there still the early arrival phenomenon? This is because the neutrino is not point particle but the 4-dimensional space-time cylinder.

    (2) The mass of muon neutrino can be expected as ~5×10-9eV/c2.

    (3) When the energy is constant the δt does not change with the distance, L, between the detect-point and the generate-point of neutrinos. However, if the neutrino velocity is really faster-than-light, δt is inevitable to the proportional changes with L. The Fermilab or the Super Kamiokande should be able to conduct this experimental verification.

  22. Report this comment

    Murali said:

    Was the method used to measure distance between OPERA and CERN accurate?

  23. Report this comment

    Mahesh Khati said:

    Before 23 years, I had proved mathematically that relative velocity may be more than light velocity. CERN proved experimentally that velocity of Neutrinos may be more than light, if this news will be confirmed then that will be new beginning of physics. So, it is necessary to think different than old concept of science.

    Please read paper “What is matter & dark matter is made up of?” on my web site This paper may help to find solution to this problem & other problems like what is dark matter? & about true relativity. I also proved E= mc^2 without referring Einstein’s relativity. I strongly oppose special theory of relativity because I think that has some basic fundamental problems like non-existence of perfect inertial frame of reference in world

  24. Report this comment

    Eugene said:

    I suggest to measure the pulsed AM or FM wave sent from the same source location and detected at the same neutrino detector location using the same GPS system that detected the neutrinos. Electromagnetic wave is known to travel at the speed of light also.

    Once the detector measures the neutrinos arrival time ealier than the EM wave when the same electronic system is used, there will be no doubt about the result. What do you guys think?

  25. Report this comment

    Bob Hamilton said:

    There were NO neutrinos traveling faster than light in either experiment. If the experiment seemed to indicate that on a follow-up run, it just means they made the same errors both times. Move along people, there’s nothing to see here.

  26. Report this comment

    Niladribihari Sahoo said:

    Personally I think if the datas from OPERA are true,then special relativity is no more valid.Many physicists have raised their voices against it,may be somewhere it is wrong,MINOS is looking forward for a better result than that of time will tell whether this theory is true.

  27. Report this comment

    Niladribihari Sahoo said:

    In standard model,neutrinos are massless entities.But from experiments, we found that neutrinos oscillate between their different flavors,so they must have a non-zero mass which is a necessary condition for neutrino oscillation to occur.As we all know,muon neutrinos are found to move with superluminal speeds from the datas of OPERA experiment, can we still say that neutrino oscillation occurs??

  28. Report this comment

    James T. Dwyer said:

    If, for ANY reason, the estimated 731,278+/-0.2 meters distance traversed by neutrinos was overestimated by about 18 meters then the detected neutrinos DID NOT exceed the speed of light.

    - Whatever the distance estimate used in calculating the comparative time-of-flight for light in a vacuum between CERN and the Gran Sasso neutrino detector, the 61 ns discrepancy would be fully accounted for by a distance of about 18 meters.

    - The actual neutrino propagation path cannot be definitively determined. The neutrino propagation distance cannot be precisely determined.

    - The approximate 730 km estimated traversal distance is dependent on a very complex method using coordinated GPS locations and standard geodesy routines. However, any relativistic spacetime distortions produced by the propagation of particles at near the speed of light within the Earth may not have been consistent with the methods employed by standard geodesy software. In this case the actual path length of neutrino propagation may have been overestimated.

    As a layperson I cannot assess these issues, but some may be addressed in the report: Wolfgang Kundt, (2011). “Speed of the CERN Neutrinos released on 22.9.2011 – Was stated superluminality due to neglecting General Relativity?”,

  29. Report this comment

    Otto Krog said:

    What if the speed of light isn’t constant?

    What if the speed of light varies through time and space?

    That would create some interesting theory. At least I think so.

    Antimatter is the mind and consciousness of all living entities.

    You are your own universe.

    Reality is where the minds (antimatter) meets the physical universe.

    Interested? Then read my philosophical multiverse theory.

    Google crestroyer theory, and find it instantly.