Jiangmen Underground Neutrino Observatory

Objective

Determine the neutrino mass ordering using precision spectroscopy of reactor antineutrinos at a 53-km baseline, and provide the most precise measurement of Δm²₂₁ and θ₁₂.

Method

A 20-kiloton liquid-scintillator detector viewed by 17,612 large-area photomultipliers and 25,600 small-area photomultipliers, giving 78% photocathode coverage. The detector is placed 53 km from the Yangjiang and Taishan nuclear power complexes, a distance chosen to sit near the first Δm²₂₁ oscillation maximum, where the fast Δm²₃₁ modulation interferes coherently with the slow Δm²₂₁ modulation to produce a mass-ordering-dependent spectral shape.

Key results

• 2024: detector filled with ~20 kt of linear-alkylbenzene scintillator; first science data.
• 2025: first oscillation spectrum measurements, consistent with standard three-flavor framework.
• Expected mass-ordering determination at 3–4σ after 6 years of running.
• Ongoing supporting experiments: TAO (Taishan Antineutrino Observatory) near one reactor for independent spectral reference.

Significance

JUNO provides the first definitive mass-ordering measurement independent of matter effects — a clean, vacuum-oscillation argument that complements the matter-effect-based determinations of DUNE and atmospheric experiments. Its combined precision on Δm²₂₁, Δm²₃₁, and θ₁₂ will be sub-percent, providing the most accurate oscillation parameter anchor available.

Why 53 km

The two reactor complexes at Yangjiang and Taishan, with combined thermal power of 35 GW, lie at 53 km and 53 km respectively from the JUNO site. This distance places JUNO near the first oscillation maximum of the $\Delta m_{21}^2$-driven solar-sector oscillation — the same oscillation previously observed by KamLAND. At the same distance the much faster $\Delta m_{31}^2$ oscillation produces rapid spectral modulation.

The two frequencies interfere in the energy spectrum. Critically, the sign of $\Delta m_{31}^2$ shifts the pattern in a way that is independent of any matter effects and independent of ${\delta }_{CP}$. The mass ordering thus emerges as a spectral-shape analysis rather than an event-count comparison.

Detector design

JUNO is the largest liquid-scintillator detector ever built — 20 kilotons of ultra-pure linear alkylbenzene (LAB) in a 35.4 m acrylic sphere, viewed by 17,612 20-inch PMTs and 25,600 3-inch PMTs providing 78% photo-coverage. The sphere sits inside a stainless-steel support structure, submerged in an ultra-pure water Cherenkov veto detector 43.5 m in diameter, instrumented for outer muon tagging.

Photon statistics at the target energy of a few MeV are sufficient to achieve 3% energy resolution — a demanding specification requiring optimized scintillator, maximal photocoverage, and 1 atm of ultra-pure conditions.

Physics capabilities

• Mass ordering: primary goal, ~3σ in 6 years through spectral analysis
• $\Delta m_{21}^2$ and ${\sin }^2{\theta }_{12}$: sub-1% precision, an order of magnitude improvement over current values
• $\Delta m_{31}^2$: improved precision through the reactor channel, cross-checking long-baseline accelerator measurements
• Solar neutrinos: ${}^8$B and ${}^7$Be below 1 MeV at higher statistics than Borexino
• Supernova neutrinos: ~5000 events from a galactic core collapse at 10 kpc
• Diffuse supernova background: complementary to Super-K-Gd
• Geoneutrinos: ~400 events per year, improving continental-crust vs mantle discrimination
• Solar and atmospheric neutrinos at low energy
• Nucleon decay, proton lifetime limits extending current bounds

TAO: the satellite experiment

A 2.8-ton gadolinium-loaded liquid-scintillator detector, TAO, is deployed near one of the Taishan reactors at 30 m baseline. Its goal is to provide a model-independent reference spectrum of reactor antineutrinos at higher statistics and better energy resolution than JUNO can achieve at its core-far-detector separation.

TAO serves two purposes: cancel uncertainties in the reactor fuel evolution model when combined with JUNO, and provide a benchmark for the still-not-fully-understood “5-MeV bump” in reactor antineutrino spectra.

Current status

Filling of the central detector completed in 2024. First data has been accumulating since. The 2025 public results include the first oscillation-spectrum measurements consistent with the standard three-flavor framework, demonstrating that the detector performance meets specification. The mass-ordering determination, the primary goal, is expected to mature through the second half of the 2020s.