Why can a neutrino change flavor?

Flavor is a name for how a neutrino interacts

Physicists distinguish electron, muon and tau neutrinos. The names refer to the charged lepton each one is paired with in charged-current weak interactions. When there is enough energy, an electron-neutrino interaction can produce an electron, and a muon-neutrino interaction can produce a muon.

A detector identifies flavor from what an interaction leaves behind. It reconstructs the products and asks which processes could have made them. Its energy threshold and its ability to tell those products apart determine which flavors it can identify.

Neutrino oscillation means that a source of one flavor can produce a different flavor at a distant detector. The evidence is statistical, built from many events. Measuring a single particle again and again along its path would be a different experiment.

Sources: [1] · [2] · [3]

The same state can also be described by mass

States with definite mass are different from states with definite flavor. A neutrino created with electron flavor is a quantum combination of mass states, and the coefficients of that combination describe mixing.

To see the idea without a large matrix, keep only two flavors and two mass states. A single mixing angle then sets how much amplitude each mass state contributes to the electron flavor. The real world has three active flavors; the two-state picture is a teaching model.

The plus sign in the equation describes one coherent quantum state, not a neutrino split into two small particles and not a source that secretly emits two classical types. When we calculate what a detector will see, the component amplitudes interfere.

Sources: [1] · [4]

The phase difference between mass components sets the flavor

Each mass component builds up a phase as it travels. The phase marks where an amplitude is in its cycle, as it does for a familiar wave. For very energetic neutrinos, different masses advance the phase by slightly different amounts.

Flavor depends on the difference between those phases. At production, the amplitudes combine to give the original flavor. Farther along, the same components add and cancel differently when tested against the possible flavors.

In vacuum this happens without any collision. The quantum state evolves, and with it the probability of detecting each flavor. Oscillation experiments do not establish the effect by directly timing the tiny speed differences.

Sources: [4] · [1]

Distance and energy set the pattern

In the simple vacuum model, the survival probability is the chance of detecting the original flavor. It oscillates with the travel distance divided by the neutrino energy, together with a difference between squared masses. The mixing angle sets how deep the oscillation can be.

That is why experiments use different baselines, meaning source-to-detector distances. At fixed energy, moving the detector changes the phase difference. At fixed distance, sorting events by energy can reveal a changing survival probability.

The formula has simple checks. At zero distance the survival probability is one. With no mixing, or with equal masses, this model gives no oscillation. Several names for neutrinos are therefore not enough: oscillation needs both mixing and a nonzero mass-squared difference.

Sources: [1] · [4]

Why real data do not look like a perfect sine wave

A real source emits a range of energies and may be spread over a region instead of sitting at one point. A detector measures energy imperfectly and records some interactions more efficiently than others. Averaging over these effects can soften or wash out the pattern of the simple formula.

Matter can also change how flavor evolves. Solar neutrinos cross the Sun’s changing density, so the vacuum equation alone is not a complete solar-neutrino calculation. Three-flavor mixing adds more parameters and more than one oscillation scale.

Experiments test this fuller picture from several directions: fewer events of the original flavor, events of a different flavor, and changes with energy or travel distance. Super-Kamiokande, SNO and KamLAND measured patterns of interactions, and one propagation theory had to explain them all together.

Sources: [2] · [3] · [5] · [1]

Try it in the Lab

Primary sources & revision

1. Particle Data Group · Neutrino Masses, Mixing, and Oscillations (2024)
2. Super-Kamiokande Collaboration · Evidence for Oscillation of Atmospheric Neutrinos (1998)
3. SNO Collaboration · Direct Evidence for Neutrino Flavor Transformation (2002)
4. Giunti & Laveder · Neutrino Mixing (2004)
5. KamLAND Collaboration · Evidence of Spectral Distortion (2004/2005)

First published 2026-10-09; last revised 2026-10-09. Original explanatory text and diagrams by Sandbox Physics. Illustrations are schematic; no experimental event records are reproduced here. This is an independent educational publication, not an official Nobel or experiment collaboration publication.