Particle Physics

Neutrino Oscillations

ν_e, ν_μ, ν_τ are superpositions of mass eigenstates ν₁, ν₂, ν₃ — flavor oscillates with L/E

Neutrinos come in three flavors (electron, muon, tau) and three mass eigenstates (ν₁, ν₂, ν₃) — these bases differ. Neutrinos produced as one flavor (e.g. ν_e from the sun) propagate as superpositions of mass eigenstates with different phases (because masses differ), causing flavor to oscillate with travel distance L and energy E. The probability of detecting a different flavor is P(ν_α → ν_β) = sin²(2θ) sin²(Δm² L/(4E)). Discovery: Super-Kamiokande (1998, atmospheric ν oscillations) and SNO (2002, solar ν flavor change) — proved neutrinos have mass, contrary to the original Standard Model assumption (Nobel 2015 to Kajita and McDonald). Mass differences known: Δm²₂₁ ≈ 7.4 × 10⁻⁵ eV², |Δm²₃₂| ≈ 2.5 × 10⁻³ eV². Absolute masses < 0.8 eV (KATRIN). Open: ν mass ordering (normal vs inverted), CP violation phase δ_CP, sterile neutrinos, Dirac vs Majorana nature.

  • Three flavorsThree mass states
  • ProbabilityP(α→β) = sin²(2θ) sin²(Δm²L/(4E))
  • DiscoverySuper-K 1998, SNO 2002
  • Solar Δm²Δm²₂₁ ≈ 7.4×10⁻⁵ eV²
  • Nobel 2015Kajita & McDonald
  • KATRIN boundm_ν < 0.8 eV

Interactive visualization

Press play, or step through manually. The visualization is yours to drive — try it before reading on.

Open visualization fullscreen ↗

Watch the 60-second explainer

A condensed visual walkthrough — narrated, captioned, under a minute.

Why neutrino oscillations matter

  • Beyond the Standard Model. The original Standard Model assumed massless neutrinos. Oscillations require nonzero mass differences and therefore nonzero masses — the first demonstrated departure from minimal Standard Model predictions.
  • Leptogenesis. Heavy right-handed Majorana neutrinos in seesaw models could decay CP-asymmetrically in the early universe and seed the observed matter-antimatter asymmetry. Light-neutrino CP measurements probe the structure indirectly.
  • Supernova probes. A core-collapse supernova emits about 10⁵⁸ neutrinos in seconds — their flavors, energies, and arrival times encode the explosion mechanism, the proto-neutron-star equation of state, and any oscillation effects in extreme matter.
  • Solar physics. The MSW effect (matter-enhanced oscillation) explains why low-energy solar neutrinos oscillate differently than high-energy ones — it lets us probe the Sun's interior density profile.
  • Cosmology. Neutrino masses contribute to the universe's matter budget. Cosmic microwave background plus large-scale-structure data bound the sum of neutrino masses to under about 0.12 eV.
  • Reactor & accelerator neutrino experiments. KamLAND, Daya Bay, T2K, NOvA, DUNE, JUNO, and Hyper-K each pin down a different combination of oscillation parameters using nuclear-reactor or accelerator-produced neutrino beams.
  • Sterile neutrinos. Anomalies in short-baseline experiments (LSND, MiniBooNE, gallium) hint at additional sterile neutrino states that mix with active flavors. MicroBooNE and SBN are testing the hypothesis.

The oscillation formula

For two-flavor mixing the survival probability is P(α → α) = 1 − sin²(2θ) sin²(1.27 Δm² [eV²] L [km] / E [GeV]). The 1.27 absorbs the constants (ℏ, c, GeV/eV conversions). Two parameters control the curve: the mixing angle θ sets the amplitude (sin²(2θ)), and the squared mass difference sets the L/E scale of oscillation. Choosing experiment baselines and energies tunes sensitivity to particular Δm² values.

Three-flavor oscillations involve the full PMNS matrix; transitions between any pair of flavors mix two squared mass differences and three angles. CP violation appears as a difference between P(ν_α → ν_β) and P(̅ν_α → ̅ν_β) for α ≠ β.

Landmark experiments

  • Super-Kamiokande (1998). Observed a deficit in upward-going atmospheric muon neutrinos versus downward, with the right energy dependence. First definitive evidence for oscillations; led to the 2015 Nobel.
  • SNO (2002). Used D₂O to measure both electron-neutrino and total-flavor solar fluxes. Deficit of nu_e plus correct total flux confirmed flavor change in flight.
  • KamLAND (2003+). Long-baseline reactor antineutrino experiment in Japan. Pinned down Δm²₂₁ with high precision.
  • Daya Bay (2012). Reactor experiment that first measured the small angle θ₁₃ with high significance, opening the door to CP-violation searches.
  • T2K and NOvA (ongoing). Long-baseline accelerator beams testing δ_CP and the mass ordering.
  • KATRIN (2019+). Direct kinematic mass measurement of tritium beta decay, current bound m_ν < 0.8 eV.

Common misconceptions

  • "Neutrinos are massless." Disproven 1998. Oscillations require nonzero mass differences and therefore nonzero masses for at least two of the three states.
  • "All three flavors are detected easily." Tau neutrinos are hardest. Identification needs a high-energy tau lepton produced in the detector. OPERA observed only a handful of tau-neutrino events from a CERN-to-Gran-Sasso beam over years.
  • "Oscillation only happens in vacuum." Vacuum oscillation is one mode. Dense matter (Sun, Earth, supernova) modifies effective oscillations via the MSW resonance, where electron neutrinos pick up an extra effective potential from forward scattering on electrons.
  • "Neutrinos travel at light speed exactly." Massive neutrinos travel at slightly less than c. The OPERA 2011 superluminal claim was a measurement error (loose fiber-optic cable).
  • "We measure absolute masses from oscillations." Oscillations only give squared mass differences. Absolute scale requires kinematic experiments (KATRIN), neutrinoless double beta decay (sensitive to a Majorana mass combination), or cosmology.
  • "Three is enough." Possible additional sterile neutrinos mixing with active flavors are the subject of ongoing investigation; short-baseline anomalies hint at Δm² near 1 eV².

Frequently asked questions

Why does neutrino flavor oscillate during travel?

Flavor eigenstates (electron, muon, tau neutrinos) and mass eigenstates (nu_1, nu_2, nu_3) are different superpositions of the same three states. A neutrino produced in a definite flavor (for instance an electron neutrino from the Sun) is a coherent superposition of mass eigenstates that propagate with slightly different phases because their masses differ. The mass eigenstates accumulate different phases over a distance L, so the recombined wavefunction at detection has a different flavor mix than at production. The result is sinusoidal oscillation with L/E.

What was the solar neutrino problem and how did SNO solve it?

Starting in the 1960s Ray Davis's chlorine experiment counted only about a third of the predicted Sun-produced electron neutrinos. The deficit puzzled physicists for thirty years. The Sudbury Neutrino Observatory (SNO) used heavy water to count neutrinos two ways: charged-current reactions sensitive only to electron neutrinos, and neutral-current reactions sensitive to all three flavors. The total all-flavor flux matched solar models exactly while the electron-only flux was suppressed, proving electron neutrinos had oscillated into other flavors during their flight from the solar core.

What is the PMNS matrix?

The Pontecorvo-Maki-Nakagawa-Sakata (PMNS) matrix is the unitary three-by-three matrix that connects flavor eigenstates to mass eigenstates: nu_alpha = sum over i of U_alpha_i nu_i. It is parameterized by three mixing angles (theta_12, theta_23, theta_13) and one CP-violating phase delta_CP (with possibly two extra Majorana phases). Measured values: theta_12 about 33 degrees, theta_23 about 49 degrees, theta_13 about 8.5 degrees. PMNS is the lepton-sector analog of the CKM matrix in the quark sector.

What is the difference between Dirac and Majorana neutrinos?

A Dirac neutrino is its own distinct antiparticle (like the electron). A Majorana neutrino is its own antiparticle. The distinction matters because the two cases have different mass-generation mechanisms and different lepton-number conservation behavior. Determining the answer requires observing neutrinoless double beta decay: if it occurs, neutrinos are Majorana. Experiments such as KamLAND-Zen, GERDA, and LEGEND-200 push limits on the half-life. As of now no signal has been found, but the next generation may settle the question.

Why is the neutrino mass hierarchy unknown?

Oscillation experiments measure squared mass differences Delta-m-squared, not the masses themselves or their absolute ordering. We know |Delta-m-squared-32| but not its sign. Two scenarios remain: normal ordering (m_1 < m_2 < m_3, with two light states and one heavier) or inverted ordering (m_3 < m_1 < m_2). DUNE, JUNO, and Hyper-Kamiokande aim to resolve the hierarchy by measuring matter effects in long-baseline experiments and accurate vacuum oscillations near a reactor.

How could neutrino CP violation explain matter-antimatter asymmetry?

Leptogenesis is a candidate mechanism for the universe's baryon asymmetry. If heavy right-handed neutrinos exist (a natural feature of seesaw mass models), they decay in the early universe with a CP-asymmetric rate, producing more leptons than antileptons. Sphaleron processes in the electroweak phase transition partially convert this lepton asymmetry to a baryon asymmetry. Detecting CP violation in light-neutrino oscillations (the delta_CP phase in PMNS) supports the existence of high-scale CP violation needed for leptogenesis.