Speed of neutrinos - latest results?

[DaveC426913]

There was a lot of kerfuffle a decade ago about neutrinos possibly exceeding the speed of light, A re-analysis of events debunked this back in 2012 I believe, but I don't know if the issue is conclusively settled. Has there been any re-confirmation of the speed of neutrinos in any paper later than 2012 or so? Everything I read dates back to then.

 

[mathman]

https://nautil.us/issue/55/trust/the-data-that-threatened-to-break-physics-rp

Try this.

 

[DaveC426913]

https://nautil.us/issue/55/trust/the-data-that-threatened-to-break-physics-rp

Try this.

OK, well that's still about the the OPERA experiment, which is now 9 years old. Nothing new there.

 

[mfb]

It's a low profile measurement. We know the mass range, we have a very strong prediction that they fly so close to the speed of light that we cannot measure the difference. I'm sure we'll get some new measurements eventually, but it's nothing people focus on.

Before MINOS 2007 the last measurement on Earth was done in the 1970s.

 

[vanhees71]

The OPERA issue is completely solved. It was some problem with a wave guide:

https://en.wikipedia.org/wiki/OPERA_experiment#Time-of-flight_measurements

 

[DaveC426913]

The OPERA issue is completely solved. It was some problem with a wave guide:

https://en.wikipedia.org/wiki/OPERA_experiment#Time-of-flight_measurements

Yes I know. And yet I still can't find a recent article that declares the latest determination for the speed of neutrinos.

 

[Vanadium 50]

And yet I still can't find a recent article that declares the latest determination for the speed of neutrinos.

1. There is no "the speed of neutrinos" any more than there is "the" speed of automobiles, airplanes, animals, etc.
2. If the last measurement was 9 years ago, what's the point of writing a paper saying "yup. no change."

Generalissimo Francisco Franco is still dead.

 

[vanhees71]

Yes I know. And yet I still can't find a recent article that declares the latest determination for the speed of neutrinos.

https://arxiv.org/abs/1203.3433

 

[Vanadium 50]

That's 9 years old too. @DaveC426913 has said in #3 he is looking for something more recent.

 

[DaveC426913]

And none of them seem to have a clear answer as to the impossibility of neutrinos traveling at - let alone beyond - c.

 

[mathman]

And none of them seem to have a clear answer as to the impossibility of neutrinos traveling at - let alone beyond - c.

Why not accept special relativity?

 

[DaveC426913]

Why not accept special relativity?

Ironically, I just read a quip somewhere that said "The worst data is better than the best theory." I don't really abide by that, but it makes the point. If the (fixed, corrected, confirmed) data is showing an anomaly, we can't just hand wave it away.

Full disclosure time. I'm not the one doubting relativity, another poster elseweb is. He is examining the SN1987A data (which is, granted, 34 years old now) and its implications. I can't rally much of a refutation if I can't even reference the current understanding of neutrino speeds.

1. There is no "the speed of neutrinos" any more than there is "the" speed of automobiles, airplanes, animals, etc.

Sloppy of me. Maximum speed.

c - and even >c - is within the the margin of error of the measurements.

 

[Vanadium 50]

Full disclosure time. I'm not the one doubting relativity, another poster elseweb is.

Shame on you. You know that PF frowns upon this sort of game.

c - and even >c - is within the the margin of error of the measurements.

So what? The expected flight time is ~10^-20 seconds slower than light. Time resolution is of order ~10^-9 seconds. Of course you're only looking at the resolution of the measurement.

 

[DaveC426913]

Shame on you. You know that PF frowns upon this sort of game.

No, I did not know that.

But, in my defense it's not any kind of game. I don't know what the established facts are about the max speed of neutrinos. I think it's fair to ask for help in finding published information I was unable to find myself (after extensive searching). It's not like I'm involving PF members in any way; I'm just looking for guidance toward a specific fact. PF is my best source when I run out.

And it only became relevant when someone asked why I don't just accept relativity. Which had nothing to do with the question I was asking.

 

[phinds]

Yes I know. And yet I still can't find a recent article that declares the latest determination for the speed of neutrinos.

Dave, neutrinos either are massless, in which case they travel at c, or they have some mass. The VERY strong belief is that they do have mass. As @Vanadium 50 said, there IS NO "the" speed for an object with mass. What is YOUR speed right now? There IS NO "the" speed for a Dave, there are just speeds relative to other objects.

EDIT: OOPS. I missed a couple of posts before posting this (but it's still true and if it has mass, the max speed approaches c as a limit)

 

[DaveC426913]

Dave, neutrinos either are massless, in which case they travel at c, or they have some mass. The VERY strong belief is that they do have mass. As @Vanadium 50 said, there IS NO "the" speed for an object with mass. What is YOUR speed right now? There IS NO "the" speed for a Dave, there are just speeds relative to other objects.

Yes. I was sloppy. I meant max speed recorded wrt to us.

EDIT: OOPS. I missed a couple of posts before posting this (but it's still true and if it has mass, the max speed approaches c as a limit)

Of course, but I wanted data/observations/findings.

 

[vanhees71]

And none of them seem to have a clear answer as to the impossibility of neutrinos traveling at - let alone beyond - c.

I find the answer pretty clear. It's of course very difficult to measure very-low energy neutrinos, where you could really show that they are really traveling with less than the speed of light. That and the fact that the samples measured are so small (just a handful of neutrinos in the ICARUS experiment) makes it very difficult to say more than that everything concerning neutrinos (including astrophysical observations) is in accordance with relativity.

BTW I find the question of experimentally measuring the speed of neutrinos and checking relativity with it legitimate. Even the best established theory has to be tested with all available observations. After all, neutrinos are still the least well known constituents of (the known) matter.

 

[snorkack]

Nothing in special relativity rules out one mass eigenstate of neutrino having rest mass of precisely zero, just as photon and graviton have rest mass of exactly zero.
Are there as yet any observations setting an experimental above zero lower bound on the rest mass of the lowest mass eigenstate of neutrino?

 

[vanhees71]

There are only upper limits in the eV range:

https://pdglive.lbl.gov/Particle.action?node=S066&init=0

 

[snorkack]

There are only upper limits in the eV range:

https://pdglive.lbl.gov/Particle.action?node=S066&init=0

Sounds like neutrinos have imaginary rest mass, since negative but real rest mass would still have a positive square, currently ruled out by evidence?

 

[vanhees71]

Because of this result there was already some rumor about a possible tachyonic nature of neutrinos too. However, it's not such a hard evidence to take such a drastic conclusion too seriously under consideration.

 

[mathman]

Nothing in special relativity rules out one mass eigenstate of neutrino having rest mass of precisely zero, just as photon and graviton have rest mass of exactly zero.
Are there as yet any observations setting an experimental above zero lower bound on the rest mass of the lowest mass eigenstate of neutrino?

If neurino rest mass was exactly zero, speed measurement would be c. Since it is not, the rest mass must be positive.

 

[ohwilleke]

Astrophysical Methods

Some of the more rigorous bounds made and attempted come from neutrino emissions from supernova compared to gravitational wave measurements and the light emitted by supernova. If you see the light or the gravitational waves before you see the neutrinos, but you see them very close in time to each other, then the neutrinos have non-zero but very low mass.

The main limitation of this methodology, however, is that you need to model with considerable precision the expected time gap between neutrino emission, light emission, and gravitational wave emission in this very complex process, and you need to be able to model the distance of the supernova from Earth (which loses precision at greater distances), and if you really want to be fancy about it, you need to consider any gravitational lensing or time dilation that may have occurred en route.

Direct neutrino detection, in and of itself, is also one of the most difficult measurements in physics, because the cross-section of interaction between neutrinos and other kinds of ordinary matter via the weak force and gravity only, is so feeble. The latest OPERA experiment reported in 2021, devoted a paragraph or two to each individual tau neutrino detection event over a decade of observations because there were so few of them (although still enough for statistical significance).

These are hard things to measure and model. But, since supernova are rapid explosive processes and typically happen a very long way away, you can still place some pretty significant bounds upon neutrino mass and speed this way, and since the signals, in practice, tend to come very close to each other in time, as long as your model of the process puts timing of the signals you expect in the right order, your models don't have to be ultra-precise. It turns out that supernovas start to spew an immense burst of neutrinos about two and a half hours before they blow emitting an intense burst of light, and we have decent models of the supernova process to explain why it happens that way. But the systemic uncertainties of this method are appreciable.

Also, this doesn't happen all that often with just the right mix of measurements and parameters that you'd like to have. Even as of 2012 when the article linked above was published, the 1987A core-collapse supernova (SN1987A) in the Large Magellanic Cloud (LMC), 50 kpc away from Earth, remained the gold standard for this kind of measurement.

Astronomers are geared up and ready to go for another perfect storm for this kind of measurement at any time, however. Far more high powered neutrino and gravitational wave and ordinary photon based telescopes are in place than there were in 1987 and they are far better coordinated, via the miracle of the Internet, as well.

One of the most recent astrophysical measurements was a neutrino with PeV magnitude energy detected by the IceCube experiment in 2019 from a "tidal disruption event" as a star was ravaged by a distant black hole that also produced electromagnetic radiation. An open access press release from NASA in 2021, when the analysis was done, explains what they saw. Again, the main difficulty was reconciling the common source light and neutrino signals with an understanding of the process that could have produced them at particular times relative to each other.

Terrestrial Experiments

On Earth, combined results of multiple experiments in 2013 with a combined average neutrino energy of 17 GeV and a distance of about 730 km on average could not discern any statistically significant experimentally observable difference between neutrino speed and the speed of light. This was one of the first post-OPERA neutrino speed measurements.

Review Articles Of Varying Levels Of Formality

Ethan Siegel writing for Forbes Magazine on May 21, 2021 reached the same conclusion (focusing on the method of using Cherenkov radiation to detect neutrinos moving faster than the speed of light in water but below the speed of light in a vacuum, without citing any new experimental results).

As of December 16, 2020, a statement on Fermilab's website from three of its scientists reported that: "Their masses are so tiny that so far no experiment has succeeded in measuring them, while they travel at nearly the speed of light."

A September 8, 2021 retrospective article in Nature Review Physics, looking back at the OPERA goof and its subsequent neutrino physics results, likewise reaffirms that this remains the status quo.

 

[DaveC426913]

Also, this doesn't happen all that often with just the right mix of measurements and parameters that you'd like to have. Even as of 2012 when the article linked above was published, the 1987A core-collapse supernova (SN1987A) in the Large Magellanic Cloud (LMC), 50 kpc away from Earth, remained the gold standard for this kind of measurement.

Astronomers are geared up and ready to go for another perfect storm for this kind of measurement at any time, however. Far more high powered neutrino and gravitational wave and ordinary photon based telescopes are in place than there were in 1987 and they are far better coordinated, via the miracle of the Internet, as well.

I see. That's why it hasn't been repeated. Because it is dependent on the occurrence and conditions of a (relatively rare) natural phenom.

 

[ohwilleke]

I see. That's why it hasn't been repeated. Because it is dependent on the occurrence and conditions of a (relatively rare) natural phenom.

This is certainly one of the most important reasons. The better our telescopes, the more of these events we can see, but for example, with regard to the 2019 event:

“Tidal disruption events are incredibly rare phenomena, only occurring once every 10,000 to 100,000 years in a large galaxy like our own. Astronomers have only observed a few dozen at this point,” said Swift Principal Investigator S. Bradley Cenko at NASA’s Goddard Space Flight Center in Greenbelt,

 

[Hornbein]

I live a few miles from the campus where neutrino mass was discovered. It evokes a Darth Vader Institute of Technology. The guest quarters in particular resemble an Alcatraz cell block.

There is however a very nice park next door.

 

[mfb]

If neurino rest mass was exactly zero, speed measurement would be c. Since it is not, the rest mass must be positive.

Speed measurements can't distinguish between c and the expected "tiny bit slower than c". We know at least two mass eigenstates must be non-zero but this knowledge comes from neutrino mixing, not speed measurements.

There has been no new experiment in the last few years that could have improved the speed measurement notably and there is no serious demand for it either. Without the faulty connection in OPERA - which has been fully resolved - no one outside particle physics would even know about neutrino speed measurements.

 

[ohwilleke]

A recent preprint on the details of core-collapse supernova based neutrino source measurements. https://arxiv.org/abs/2109.13242 The abstract states:

During a failed core-collapse supernova, the proto-neutron star eventually collapses under its own gravitational field and forms a black hole. This collapse happens quickly, on the dynamical time of the proto-neutron star, ≲0.5 ms. During this collapse, barring any excessive rotation, the entire proto-neutron star is accreted into the newly formed black hole.

The main source of neutrinos is now removed and the signal abruptly shuts off over this formation timescale. However, while the source of neutrinos is turned off, the arrival times at an Earth-based detector will depend on the neutrino path. We show here that a modest amount of neutrinos, emitted just prior to the black hole forming, scatter on the infalling material into our line of sight and arrive after the formation of the black hole, up to 15 ms in our model. This neutrino echo, which we characterize with Monte Carlo simulations and analytic models, has a significantly higher average energy (upwards of ∼ 50 MeV) compared to the main neutrino signal, and for the canonical failed supernova explored here, is likely detectable in (10 kT) supernova neutrino detectors for Galactic failed supernovae.

The presence of this signal is important to consider if using black hole formation as a time post for triangulation or the post black hole timing profile for neutrino mass measurements. On its own, it can also be used to characterize or constrain the structure and nature of the accretion flow.

 

[Hornbein]

A recent preprint on the details of core-collapse supernova based neutrino source measurements. https://arxiv.org/abs/2109.13242

Is the idea that the higher energy neutrinos will be faster. Then if we can detect the pulses and the model for timing of the pulse origin is correct we can then calculate the rest mass. I suppose the problem is that at highly relativistic speeds the difference in velocity is tiny. Well, it's not my problem. Let someone else earn their PhD doing this.

 

[jbriggs444]

I meant max speed recorded wrt to us

Mathematical pedant hat on...

There is no maximum speed for a neutrino in any particular frame. For any speed that a neutrino can have there is another possible speed that is higher still.

Like any massive object there is an upper bound on the speed of a neutrino in any particular inertial frame. The least such upper bound is c.

The mathematical distinction is that the "maximum" value in a set of values is actually attained by one of the set members. The "least upper bound" is not necessarily attained by any set member. It's pretty much the same kind of distinction as between an open interval and a closed interval - just a tiny little point on the end.

 

[Vanadium 50]

We have never seen a neutrino from a supernova. We have only seen antineutrinos.

Neutrinos have a short emission time - a time scale of tens of milliseconds. Antineutrinos have a time scale of tens of seconds.

So a nearby galactic supernova would be 1000x better than 1987a? Not exactly - while we know the start time ~1000x better, it would also be ~1000x closer,. so the time difference is ~1000x smaller. And we're right back where we were. Maybe we'll do two or four times better, but not a thousand.

 

[ohwilleke]

Is the idea that the higher energy neutrinos will be faster. Then if we can detect the pulses and the model for timing of the pulse origin is correct we can then calculate the rest mass. I suppose the problem is that at highly relativistic speeds the difference in velocity is tiny. Well, it's not my problem. Let someone else earn their PhD doing this.

The idea is that neutrinos get spewed out in advance of the the explosion itself fully manifesting in a particular pattern, and you need to model the process properly to make meaningful comparisons of different signals arriving from a distant source. You also need to model properly different routes that the neutrino could have taken to get from the source to Earth where it is detected.

 

[ohwilleke]

We have never seen a neutrino from a supernova. We have only seen antineutrinos.

True but pretty pedantic. Obviously, in this context, one is talking about both neutrino and antineutrino observations when one is talking about what neutrino telescopes on Earth are seeing.

Neutrinos have a short emission time - a time scale of tens of milliseconds. Antineutrinos have a time scale of tens of seconds.

So a nearby galactic supernova would be 1000x better than 1987a? Not exactly - while we know the start time ~1000x better, it would also be ~1000x closer,. so the time difference is ~1000x smaller. And we're right back where we were. Maybe we'll do two or four times better, but not a thousand.

This tradeoff is indeed a big deal and creates a sweet spot that is not too near and not too far.