In November 1989, the Large Electron-Positron Collider at CERN turned on. Its 27-kilometer ring of magnets and accelerator components, buried 100 meters under the French-Swiss countryside near Geneva, sent counter-rotating beams of electrons and positrons around the ring at energies tuned to produce the Z boson — the heavy neutral mediator of the weak force.
Four experiments — ALEPH, DELPHI, L3, and OPAL — recorded the collisions. Each was a giant cathedral of tracking chambers, calorimeters, and muon detectors arranged around an interaction point. By the end of 1995, the four experiments had collectively recorded about 17 million Z events — an enormous statistical sample that allowed them to measure the properties of the Z to a precision that had never been achieved before.
One of the most important measurements that came out of this program was the counting of light neutrino species. The Z boson decays into all the fermion-antifermion pairs that are kinematically allowed. Some of those decay channels produce visible particles like quarks or charged leptons. Others — those involving neutrino-antineutrino pairs — produce only invisible neutrinos that pass through the detectors unobserved.
The total decay rate of the Z, measured through its line shape, includes the contribution from these invisible channels. Subtracting the visible channels and dividing by the predicted rate per neutrino species gives the number of light neutrino flavors that couple to the Z. The answer, summed across all four LEP experiments, was N_ν = 2.984 ± 0.008 — consistent with exactly three.
Why the Z width is a neutrino counter
The Z boson is the neutral electroweak gauge boson. It has a mass of about 91.2 GeV and a total decay width of about 2.5 GeV — meaning that on average, after about 3 × 10⁻²⁵ seconds, every Z decays into a pair of Standard Model fermions.
The branching fractions are dictated by the Standard Model. About 70% of Z decays produce a quark-antiquark pair (which becomes a hadronic jet). About 10% produce a charged-lepton pair (electron, muon, or tau plus its antiparticle). The remaining roughly 20% produce a neutrino-antineutrino pair — invisible to ordinary detectors.
The “invisible” contribution to the Z width can be computed by subtraction: take the total width (measured from the energy dependence of the cross-section across the Z resonance), and subtract the visible contributions (measured directly from the rates of detected hadronic and charged-lepton events). The remaining width is the invisible part.
The Standard Model predicts each light neutrino species contributes about 167 MeV to the Z width. So dividing the measured invisible width by 167 MeV gives the number of active neutrino species:
N_ν = Γ_invisible / Γ(Z → νν̄ per species)
LEP’s combined result of 2.984 ± 0.008 is, to within experimental uncertainty, exactly 3.
How LEP achieved the precision
Two technical achievements made the precision possible.
Beam energy calibration to one part in 10⁵. The shape of the Z resonance is a Lorentzian peak with a width of about 2.5 GeV centered at 91.2 GeV. To extract the resonance parameters precisely, you have to know the collision energy of each event very accurately. LEP achieved this through resonant depolarization — a technique that uses the spin of the electrons in the beam to track changes in the orbital frequency.
A small oscillating magnetic field was applied transverse to the orbital plane. When the frequency of that field matched the natural precession frequency of the electron spins, the polarization of the beam dropped abruptly. By scanning the frequency and detecting where the depolarization occurred, the collaboration could determine the beam energy to a precision of about 1 MeV out of 45 GeV per beam — about 2 parts per million.
Even at this precision, the LEP team eventually discovered that several effects — the Moon’s tidal pull on the rock around the tunnel, the rising and falling water level of Lake Geneva, leakage currents from the TGV high-speed rail line — all moved the ring slightly and changed the orbital length by tiny amounts. Each effect had to be measured and corrected for.
Extraordinary statistics. The four experiments collectively recorded about 17 million Z events. The statistical uncertainty on the line shape scales as 1/√N, so 17 million events give a precision of about 0.02%. Combined with the energy calibration, this allowed the Z mass to be determined to about 2 parts per million and the total width to about 0.1%.
What the result rules out
The N_ν = 2.984 ± 0.008 result places a tight constraint on certain kinds of new physics.
No fourth active neutrino. A hypothetical fourth neutrino species that participated in the weak interaction the way the three known species do (and had a mass below about 45 GeV) would have increased N_ν to roughly 4. The LEP data exclude this at very high confidence. Any fourth neutrino must be either heavier than 45 GeV or sterile to the weak interaction.
No anomalous Z couplings. Some proposed extensions of the Standard Model predict modifications of the Z’s couplings to neutrinos that would change Γ_invisible without adding a new species. These are constrained to be small by the same data.
The Standard Model’s three-generation structure is correct. The known fermion content of the Standard Model fits in three generations. If there had been a fourth, the LEP measurement would have caught it.
What it doesn’t rule out
The LEP measurement is sometimes loosely described as “there are exactly three neutrino species.” This overstates the case. Several things are not ruled out.
Sterile neutrinos. Particles that mix with ordinary neutrinos via oscillation but do not interact via the weak force are invisible to the Z boson. Their existence is constrained by entirely different measurements — oscillation experiments, cosmology, beta decay endpoint experiments — but the LEP result says nothing about them. The LSND/MiniBooNE, reactor antineutrino, and gallium anomalies all hint at possible sterile states, and none is constrained by LEP.
Heavy active neutrinos. Particles that couple to the Z but are too heavy to be produced in Z decay (above about 45 GeV per particle in the pair) would not contribute to the LEP measurement. The see-saw mechanism predicts heavy right-handed neutrinos at much higher masses, and these are entirely consistent with the LEP result.
So the LEP result is more precisely stated as: “there are exactly three light active neutrino species.” Heavier active species or any number of sterile species are allowed.
How the measurement was combined
Each of the four LEP experiments measured its own value of N_ν, but the energy calibration was common to all four. The collaborations therefore agreed in advance to combine their data only after careful internal cross-checks. The combination procedure accounted for both statistical and systematic correlations between the four detectors.
The final published result, in a long 2006 paper jointly authored by all four collaborations and the LEP Electroweak Working Group, was N_ν = 2.9840 ± 0.0082. The central value differs from 3 by about 2σ — small enough that it is generally treated as consistent with 3, but precise enough that it has occasionally been cited as a mild hint of new physics. Subsequent re-analyses have shifted the central value closer to 3 as the QED radiative corrections used in the calculation have been improved.
LEP’s broader legacy
The LEP program ended in November 2000, when the machine was shut down to make way for the LHC. By that point it had measured every major property of the Z and W bosons to per-mille precision, set strong limits on the Higgs boson mass (later confirmed when ATLAS and CMS discovered the Higgs), and established the validity of the Standard Model in the electroweak sector to a degree no previous experiment had approached.
The N_ν measurement was one of many, but it was also one of the most consequential. By the early 1990s, the field was confident that the Standard Model’s three-generation structure was complete — that the three known neutrinos were not the tip of a longer series of weakly-interacting species. This shaped almost every subsequent direction of neutrino physics. Sterile-neutrino searches focused on hidden-sector models rather than additional Standard-Model neutrinos. Cosmological N_eff constraints became a precision test rather than a discovery channel. Theoretical work on neutrino mass took the three-generation structure as a given.
The number that stayed three
A precision measurement of the lifetime of one elementary particle — the Z boson — determined that the universe contains exactly three light, active neutrino species. The measurement is now thirty years old and has been refined slightly in successive years, but the central conclusion has stayed solid: three. Not two, not four, just three.
The simplicity of the result obscures how hard it was to extract. The LEP program required four detectors, the world’s best particle accelerators, and a generation of physicists who spent a decade pursuing precision. The result is one of the cleanest tests of the Standard Model anywhere, and the value of N_ν is one of the most stable numbers in particle physics: it does not move when better data come in. It is what nature picked.
For the related cosmological count of relativistic species, see Big Bang nucleosynthesis. For the sterile-neutrino question LEP does not address, see Sterile neutrinos: a stubborn maybe. For the broader Standard Model picture neutrino physics lives in, see How neutrino oscillation works.
Frequently asked
What was LEP?
LEP (Large Electron-Positron Collider) was a 27-kilometer circular accelerator at CERN that collided electron and positron beams at energies of 91 to 209 GeV. It operated from 1989 to 2000. Four large experiments — ALEPH, DELPHI, L3, and OPAL — recorded data from the collisions. LEP's tunnel is now used for the Large Hadron Collider.
How does LEP count neutrino species?
By measuring the lifetime of the Z boson, which decays into all kinematically accessible fermion pairs including neutrino-antineutrino pairs. Each light neutrino species — those with masses below half the Z mass, about 45 GeV — contributes equally to the Z's total decay width. The 'invisible' contribution from neutrinos divided by the contribution per species gives the number of light neutrino flavors.
What does the LEP measurement say?
The combined LEP result is N_ν = 2.984 ± 0.008. This is consistent with exactly 3 active neutrino species participating in the weak interaction, ruling out a fourth Standard-Model-like neutrino at high confidence.
Does this rule out the sterile neutrino?
No. The LEP measurement only counts neutrinos that couple to the Z boson via the standard weak interaction — that is, only 'active' neutrinos. Sterile neutrinos, by definition, do not interact via the weak force and would not show up in the Z width. The result is therefore a constraint on additional active neutrinos, not on sterile ones.
Why was LEP's measurement so precise?
Because the experiments collected enormous numbers of Z bosons (about 17 million across the four experiments) at extraordinarily well-calibrated energies. The energy of the LEP beams was determined to about one part in a hundred thousand using the technique of resonant depolarization. The combination of high statistics and excellent energy calibration made the Z line shape one of the best-measured curves in particle physics.
Cite this article 5 formats
APA
Neutrino Times Editorial Team. (2025, November 12). LEP: how a Geneva collider counted the neutrino species and got 3. Neutrino Times. https://neutrino-times.com/articles/lep-z-width-three-light-neutrino-species/
Chicago
Neutrino Times Editorial Team. "LEP: how a Geneva collider counted the neutrino species and got 3." Neutrino Times, November 12, 2025. https://neutrino-times.com/articles/lep-z-width-three-light-neutrino-species/.
MLA
Neutrino Times Editorial Team. "LEP: how a Geneva collider counted the neutrino species and got 3." Neutrino Times, 12 Nov. 2025, https://neutrino-times.com/articles/lep-z-width-three-light-neutrino-species/.
BibTeX
@misc{neutrino-times-lep-z-width-three-light-neutrino-species,
author = {Neutrino Times Editorial Team},
title = {LEP: how a Geneva collider counted the neutrino species and got 3},
howpublished = {Neutrino Times},
year = {2025},
month = {nov},
url = {https://neutrino-times.com/articles/lep-z-width-three-light-neutrino-species/},
note = {Accessed: 2025-11-12}
} RIS
TY - GEN TI - LEP: how a Geneva collider counted the neutrino species and got 3 AU - Neutrino Times Editorial Team PY - 2025 DA - 2025-11-12 PB - Neutrino Times UR - https://neutrino-times.com/articles/lep-z-width-three-light-neutrino-species/ ER -