How Do You Detect Neutrino Oscillations?
Updated: Oct 9
This is the third article on neutrinos.
Before looking at how neutrino oscillations are detected, let's ask a basic question.
Why do Neutrinos Oscillate?
When a neutrino is born, it is created in a pure flavour state, (either electron-neutrino, muon-neutrino or tau-neutrino) but in a quantum superposition of three different mass states.
When neutrinos travel, because of their tiny masses and the fact that they hardly interact with the environment through which they travel, they maintain a state of coherence. In this context, “coherence” means that the different states in superposition, maintain a fixed phase relationship, allowing their quantum amplitudes to interfere with each other. Neutrinos travel as a superposition of all three neutrino masses.
When a neutrino does interact with matter, the superposition collapses into a pure flavour state, based on quantum probabilities at the measured distance from the source. This results in oscillation. Theoretically, once a neutrino is born, always with a pure flavour state, it will oscillate until the superposition collapses due to an interaction.
However, this isn't always the case as the wave packets associated with the different mass states have different speeds and can gradually separate, reducing, or even eliminating, oscillation over inter-galactic distances, for example.
The mechanism that creates oscillation is known as “neutrino mixing”, and although a neutrino may have a pure flavour state, its mass is indeterminate. Oscillation experiments measure differences between the squares of the neutrino masses, while other experiments are needed to constrain their absolute masses.
Neutrinos are something like 500,000 times lighter than an electron, making the effect they have on standard measuring equipment almost undetectable. As a consequence, their masses are always reported as the highest possible value, so far recorded.
Why don't quarks and charged leptons oscillate like neutrinos?
Quarks also exhibit “flavour mixing”, but only within the confined domain of hadrons. Charged leptons do not exhibit the same sort of flavour mixing as neutrinos.
There are three important differences between the fundamental fermions (quarks and charged leptons) of the standard model and neutrinos;
Neutrinos are produced in definite flavour states, which are quantum superpositions of the three mass states. As neutrinos travel, the different mass components propagate at different speeds resulting in travelling flavour oscillations.
Free quarks can't travel far as the powerful colour interaction traps them inside hadrons almost instantaneously. However, within a hadron, their flavour can change through the weak nuclear interaction, but this is not the same as the travelling flavour oscillations of neutrinos.
The six quarks have very different masses and the three charged leptons have substantially different masses. These differences affect the phases and coherence of quantum states. However, for charged leptons, their flavour states coincide with their mass states, which means that charged leptons do not exist in superpositions of mass states and so, do not oscillate.
In fact, because of “quark mixing” between pairs of quarks in neutral mesons, composed of a quark and an anti-quark, the two particles do oscillate into their own anti-particles. This only occurs in the confined environment of mesons and are known as “neutral meson oscillations”, but they are quite different to travelling neutrino oscillations.
How Can You Detect a Neutrino has Oscillated?
Imagine a neutrino flux being produced by a powerful accelerator.
Let's suppose that at the accelerator, the neutrino flux components may be measured as being predominately muon-neutrinos in a nearby detector.
Let the rest of the flux travel hundreds of miles to another, remote detector.
We know, through measurements, that the accelerator is producing mainly muon-neutrinos. If the detector at the remote site also measures muon-neutrinos, then it's probable that no oscillations have occurred. Alternatively, if measurements show that muon-neutrinos are no longer the dominant measurement, then oscillation must have occurred.
Measurements must prove that the majority of the flux reaching the remote site is now different from the majority determined at the accelerator site.
This is how neutrino oscillations are confirmed.
In practice, the distance at which the muon-neutrino will have completed a full cycle of changes, known as the "oscillation length", can be calculated and from this, the distance to where the neutrino is most likely to have changed flavour, can be determined.
The Oscillation Length
There is an approximate relationship between the distance at which a full cycle of oscillations occur and both the energy and the squares of the neutrino masses. If a neutrino travels this distance, then it oscillates back to its original flavour. The relationship, in what's called the simple two-flavour model, is:
L(osc) = 2.48 E/Δm2
Where; L(osc) is the length over which a full oscillation cycle occurs, E is the energy of the neutrino at birth and Δm2 is the difference between the squares of the two mass values involved.
The following graph, derived from the simple two flavour model, shows the probability of measuring a particular flavour at distances from the source. If it was a muon-neutrino created at birth, where probability is 1 (Orange trace), then the probability is again 1 at the oscillation distance for a 1GeV muon-neutrino.
At the mid-way distance, the probability of it still being a muon-neutrino becomes 0 and the probability of it becoming a Tau-neutrino has become 1. This is based on the simplified two-flavour model: in reality, the muon-neutrino could also oscillate into an electron-neutrino.
So, calculate the oscillation distance for a 1GeV muon neutrino and halve the value to get the distance at which the muon-neutrino should have changed flavour and place your remote detector there.
Δm2 has been determined experimentally for muon-neutrinos and tau-neutrinos and the equation yields an oscillation distance of 616 miles for 1GeV muon-neutrinos.
In summary;
Source at 0 miles: the beam is almost pure muon-neutrinos, created in the accelerator as there has been no distance travelled over which amplitude interference can occur.
Maximum Phase Interference at 308 miles: the beam has undergone almost complete phase inversion and is most likely to have become a tau-neutrino or, with a very low probability, an electron-neutrino.
End of Cycle at 616 miles: the beam has completed a full 360 degree phase alignment cycle and is most likely to have become a muon-neutrino once more.
The Long Base Experiment
Catching a muon-neutrino oscillating into a tau-neutrino is a very specific event, although it could also have oscillated into an electron-neutrino. However, the oscillation to an electron-neutrino is rarer but it also occurs at a precise distance from the source. The result is that at the half way distance, there will be a mixture of all three flavours, the most dominant being the tau flavour. Remember that the calculation to reveal the approximate oscillation length was performed using the simple two flavour model which provided a starting point.
So, as already described, two detectors are employed, one close to the source of pure muon-neutrinos and one at the calculated distance from the source where it is expected to see tau-neutrinos dominate the mix. For a 1GeV muon-neutrino this corresponds to 308 miles from the source. This detector does need to be large as the muon beam will have diverged somewhat despite the time of flight being measured in milliseconds at relativistic speeds.
Result: the remote detector shows a mixture of neutrino flavours, dominated by tau-neutrinos and demonstrates oscillation.
The precise 1 GeV Long Base muon neutrino experiment is the T2K (Tokai-to-Kamioka) experiment in Japan. The neutrinos travel through the earth's crust across the width of Japan's main island. The muon beam is generated at the J-PARC accelerator complex, where high-energy protons are fired at a fixed graphite target to generate pions, which quickly decay into anti-muons and muon-neutrinos. All charged particles are removed from the beam using magnets and beam dumps, leaving a pure beam of muon-neutrinos.
Experiments to detect oscillation require up to three years continuous running to collect statistically significant data. Experiments to investigate deeper into oscillations may require up to ten years of continuous running.
Conclusion
Originally, according to the standard model of particle physics, neutrinos had no mass and travelled at the velocity of light in a vacuum from a theoretical point of view.
Proof that neutrinos oscillate implies that they “evolve” over time and anything that evolves cannot travel at the velocity of light in a vacuum and therefore, must have mass. This changed the standard model and for the first time, confirmed that there was physics beyond the standard model.
Final Take Aways
Neutrinos are created in a pure flavour state (electron-, muon- or tau-neutrino) but in a superposition of all three mass states due to neutrino mixing.
While they travel, as a superposition of all three mass states, they maintain a state of coherence allowing their quantum amplitudes to interfere with each other. This results in flavour oscillation.
Due to the different speeds of the mass components, flavour oscillation may be weakened or disappear all together over cosmic distances.
Using a simplistic two flavour model, there is an approximate relationship between the distance over which one complete oscillation occurs for a neutrino at a given energy and the difference between the squares of the two mass values involved. This equation provides the oscillation length. At half this distance, the two flavour model, suggests that the initial neutrino has changed to the second type. This is the baseline distance at which the remote detector should be built.
For example; a) use an accelerator to produce predominately muon-neutrinos, proved by a detector close by. b) Allow the flux of neutrinos to travel to the remote detector, placed at half the calculated oscillation length. c) measure the neutrinos types reaching the remote detector.
If the remote detector shows that the dominant measurement indicates tau-neutrinos, then oscillation has been detected.
Base-line experiments confirm that neutrinos evolve over time and therefore cannot travel at the speed of light and must have mass. To produce statistically significant data, takes up to three years of continuous running.
The detection of neutrino oscillations has been regarded by physicists as one of the most profound discoveries in recent years, leading to a change to the standard model of particle physics and to a Nobel prize in physics in 2015.
--
© 2026 Kieron Conway - All rights reserved.
–
There are now three articles on neutrinos on this blog:
Why are neutrinos so difficult to detect?
How do you detect neutrinos?
How do you detect neutrino oscillations?
The final article;
What is the ICECUBE Experiment?
is coming soon, to commemorate the award of the 2026 Nobel prize in physics.
--
Liked this article? Check out:
where you can read all about an exciting new science series: A Journey into Modern Physics by Kieron Conway, available in three parts, from Amazon on-line shops.
Also, the web-site has an index of all the blog articles published to date, for easy access to an article that might interest you. You can access the index from the web-site's main-menu, using the link above.
Also, the first year's worth of blog articles have been published in an E-book and a paper-back, available from your Amazon store. You can still continue to access all articles from the web site's blog, but why not have them all at you finger tips in book form?
----
Comments