JUNO: The World’s Largest Neutrino Detector, 700 Metres Under Kaiping

For more than a century, Kaiping (开平) was a place people left. Its sons and daughters boarded junks at the Haikoubu departure port, laid the transcontinental railroads, and worked the hand laundries of North America. Their remittances came home as fortified towers — the diaolou that now make the county a UNESCO World Heritage landscape. Today Kaiping sits above a different kind of frontier.

Seven hundred metres beneath Dashi Mountain (打石山) in Kaiping’s Jinji Town, the Jiangmen Underground Neutrino Observatory (江门中微子实验, JUNO) — the largest and most precise neutrino detector ever built — has been quietly reading the ghost particles that stream out of the region’s nuclear reactors. Since it began operating in August 2025, it has produced the world’s most precise measurements of how neutrinos change as they fly; in June 2026 it delivered the strongest hint yet to one of particle physics’ oldest open questions. The emigration country has become a science destination, and the world’s physicists now fly into Kaiping to use it. For a descendant of Wuyi, the question a century of letters home kept asking has turned inside out: not “what did we leave behind,” but “what has it become?”

A Particle That Passes Through Everything

Neutrinos (中微子, zhōngwēizǐ — “little neutral ones”) are the most abundant massive particles in the universe, and the hardest to catch. Every second, trillions of them pass through your body — and through the entire Earth — without touching a single atom. They carry no electric charge and almost no mass, so they barely interact with ordinary matter. To a neutrino, a planet is mostly empty space.

That elusiveness is precisely why they matter. Because neutrinos barely interact, they carry information out of places nothing else can escape: the cores of stars, the heart of a supernova, the interior of our own planet. And because they come in three “flavours” that continually morph into one another as they travel — a phenomenon called neutrino oscillation (中微子振荡) — they are one of the few known cracks in the Standard Model of particle physics, a hint of physics beyond what we currently understand.

Neutrinos come in three flavours, named for the charged particles they accompany: the electron neutrino, the muon neutrino and the tau neutrino. As a neutrino travels, it does not stay one flavour — it oscillates, cycling through all three, and the rate at which it does so is governed by six numbers: three mixing angles (θ₁₂, θ₁₃, θ₂₃) and three mass-squared differences. Measuring those six numbers to ever-finer precision is how physicists stress-test the Standard Model and hunt for whatever lies beyond it. JUNO was built to nail down the two hardest of them, and to settle, once and for all, the order of the three masses.

To study them, you need a detector of almost absurd scale and purity, buried deep enough that nothing else can reach it. JUNO is that detector.

Why Here, Why Kaiping

The first question people ask about JUNO is why a world-class physics experiment sits in a county better known for its emigrant villages than its laboratories. The answer is geometry.

Nuclear reactors produce a continuous flood of antineutrinos, and JUNO is built to watch them. It sits exactly 53 kilometres from both of the two nearest nuclear power stations — Yangjiang (阳江, six reactors) and Taishan (台山, two reactors) — a distance that is not arbitrary. At 53 kilometres, the neutrino oscillation signal — the tell-tale pattern of how the particles change flavour as they travel — reaches its strongest, cleanest point. Put the detector closer or farther, and the signal smears out.

This is not Guangdong’s first neutrino experiment, and the choice of province is no accident. JUNO’s predecessor, the Daya Bay Reactor Neutrino Experiment (大亚湾中微子实验), sat just across the estuary near Shenzhen, and in 2012 it made the landmark measurement of the mixing angle θ₁₃ that reoriented the entire field — work later recognised with the 2016 Breakthrough Prize in Fundamental Physics. That experiment proved Guangdong’s reactor coast, with its dense cluster of powerful plants, is one of the best places on Earth to do neutrino physics. JUNO is its direct heir: proposed in 2008, approved in 2013, under construction from 2015, and conceived from the start around the geometry of the Yangjiang and Taishan plants. The site search was a literal exercise in geometry — the ideal location was a rectangle two kilometres long and 200 metres wide, exactly 53 kilometres from each plant, and Kaiping’s Jinji Town fell inside it.

That delicate arithmetic placed the world’s most sensitive neutrino experiment on a hillside in Kaiping, roughly equidistant from the two plants. The site, at 700 metres of rock overburden, also shields the detector from the cosmic-ray noise that would otherwise drown out the signal. The Pearl River Delta’s geography — its granite, its reactor coast, its position on the map — is the reason the experiment is here, not anywhere else.

The Crystal Ball Under the Mountain

The detector itself is a sphere, and it looks like something between a cathedral and a machine. Inside a hall hollowed out of the granite, a transparent acrylic sphere 35.4 metres across holds 20,000 tonnes of liquid scintillator (液体闪烁体, yètǐ shǎnshuòtǐ) — a mineral-oil-like fluid that flashes with light the instant a neutrino collides with one of its protons. Lining the sphere are more than 40,000 photomultiplier tubes (光电倍增管), each a glass bulb that turns a single flash of light into an electrical signal, recording both the energy and the exact position of every neutrino event.

Around the sphere sits a pool of ultra-pure water that acts as a shield, and a set of coils that cancel the Earth’s magnetic field so it cannot bend the faint signals. On top, a “top tracker” catches cosmic muons that make it through the rock, so the experiment can tell genuine neutrinos from impostors.

The engineering is as extreme as the physics. The 20,000 tonnes of liquid scintillator had to be purified until almost nothing in it is radioactive, because a single stray impurity would drown the faint neutrino flashes in background noise. The acrylic sphere that holds it — the largest such vessel ever built — was assembled inside the cavern with millimetre precision and filled over months without the ultra-pure liquid ever touching open air. Even the surrounding water is ultra-pure, re-filtered so that it shields the detector while contributing nothing of its own. The result is an instrument that resolves a neutrino’s energy to within 3 percent — a figure the collaboration says the running detector has already met or beaten.

The scale is hard to hold in the mind, so the collaboration’s own renderings place tiny human figures beside the sphere — technicians in hard hats who come up to a fraction of one photomultiplier tube. The whole apparatus resolves the energy of a neutrino to within 3 percent. It is, the collaboration says, the most precise neutrino energy measurement on Earth.

First Light: The Nature Cover

JUNO did not wait long to deliver. On 19 November 2025, less than three months after it switched on, the collaboration announced its first physics result: precise measurements of two of the six numbers that govern how neutrinos oscillate — θ₁₂ (the “solar” mixing angle) and Δm²₂₁ (one of the two mass-squared differences).

The result was strong enough that, on 10 June 2026, the journal Nature published it as a cover article — a rare distinction for a facility that had been running for only 59 days. Analysing data from 26 August to 2 November 2025, JUNO pinned down sin²θ₁₂ = 0.3092 ± 0.0087 and Δm²₂₁ = (7.50 ± 0.12) × 10⁻⁵ eV², improving the precision of the previous world-leading combination by a factor of 1.6 — relative uncertainties of 2.81% and 1.55% respectively. One reviewer called it the opening of the era of precision neutrino-oscillation measurement.

It was the first time a large basic-physics facility led from China had claimed the cover of Nature, and the collaboration’s first data also re-confirmed a long-standing “solar tension” between solar and reactor measurements — a discrepancy that some physicists suspect points to genuinely new physics.

The two numbers carry real physical meaning. θ₁₂, the solar mixing angle, governs how the neutrinos streaming from the Sun’s core oscillate on their way to Earth — the oscillation that earlier experiments such as Canada’s SNO and Japan’s KamLAND first measured. Δm²₂₁, the smaller of the two mass-squared differences, is the parameter that separates the two lightest neutrino masses. To pin both down to 1.6 times better precision in a single paper, from only 59 days of data, is what persuaded Nature to give the collaboration its cover.

A Hint of an Answer

The big prize, though, is the question JUNO was built to answer: what order do the three neutrino masses come in? It is one of the outstanding puzzles in particle physics, because the answer shapes how we understand why there is more matter than antimatter in the universe — why we exist at all.

On 22 June 2026, at the International Neutrino Conference (Neutrino 2026) held at the University of California, Irvine, JUNO spokesperson Wang Yifang (王贻芳) reported the collaboration’s newest results. With the data accumulated since operations began, JUNO now finds that the masses are ordered the “normal” way (正序, zhèngxù — the lightest neutrino first), with a statistical significance of 2.3 standard deviations — about 97.5% confidence. It is the most precise mass-ordering determination made anywhere to date, and the collaboration measured the two mass-squared differences to about 1% precision and the mixing angles to about 2% — both world records.

“Normal” versus “inverted” ordering is not an academic distinction. The order of the three neutrino masses is written into how the early universe behaved, and it is entangled with the deepest open question in cosmology: why the universe contains matter at all, rather than having annihilated into pure radiation a fraction of a second after the Big Bang. Neutrinos — and whether they are their own antiparticles — sit at the heart of several proposed answers. So when JUNO reports that the masses appear to be ordered the normal way, at a confidence no experiment has matched, it is nudging an entire field toward an answer.

The same announcement included a quieter result that speaks directly to the ground beneath Kaiping: JUNO has now collected the largest sample of geoneutrinos (地球中微子, dìqiú zhōngwēizǐ) ever recorded — neutrinos born from the radioactive decay of uranium and thorium deep inside the Earth. Analysing them, the collaboration says it can already exclude several models of the Earth’s interior, and finds a hint that the geochemistry of southern China differs from the global average. A detector built to look outward at the cosmos is, at the same time, peering downward at the geology of the very province it stands in.

The Nuclear Coast That Powers It

There is a symmetry here that is hard to miss. JUNO does not make its own neutrinos; it borrows them from the two nuclear plants 53 kilometres away — the same stretch of Guangdong coast that once sent coolie ships out to the Americas. The coast of departure has become the coast that powers the world’s most precise physics experiment.

Even the detail work is regional. The Taishan Neutrino Experiment (台山中微子实验, TAO), JUNO’s smaller companion, sits just 44 metres from the core of the Taishan reactor itself. Where JUNO is a 20,000-tonne cathedral, TAO is a 2.8-tonne instrument that measures the reactor’s raw neutrino spectrum with about 2% energy resolution — the finest of its kind — so that JUNO’s oscillation signal can be read against a perfectly known reference. TAO began taking data in 2026 and its high-precision spectrum was built directly into JUNO’s June results. Two detectors, one far and one near, are jointly unpicking the neutrino’s deepest secret.

The reactors themselves are regional landmarks. Taishan’s two units are EPR reactors — among the largest single-unit power stations in the world — and together with Yangjiang’s six, they give the two plants the largest combined reactor-neutrino output anywhere on Earth. That is the quiet infrastructure fact underneath the science: the world’s most precise neutrino experiment runs on the same power stations that light southern China. A coast that was once a launch point for emigration is now an engine room of the modern economy.

From Departure Ports to a Science Destination

What does any of this have to do with a family-history site? Everything.

Kaiping is where a diaspora reader’s ancestors left from. The diaolou that stud its paddies were paid for by overseas remittances — concrete proof, in the landscape, of the two-way bond between the Wuyi counties and the world. The departure ports, the railroad camps, the laundries and the restaurants all belong to one story: a region that sent its people out into the world because it was too poor to keep them.

JUNO is the inverse of that story, and it is the part of Wuyi the diaspora has never been told. The county your family left now hosts the most advanced neutrino detector ever built. The physicists who fly into Kaiping to read the detector fly in on the very transport corridor that descendants take to reach their ancestral villages. The same remittance economy that raised the diaolou a century ago is, one generation of infrastructure later, the region that pours concrete for a 700-metre-deep laboratory and plays host to 700 researchers from 17 countries. Emigration country, turned science destination — that is what Wuyi is now, and no history of the region is complete without it.

There is a symmetry worth sitting with. The emigrant of 1900 travelled outward to find work; the physicist of 2026 travels inward to find knowledge. The first raised towers to defend a home he had left behind; the second builds instruments to read the universe from that same ground. Wuyi has, within a single century, been both the region that sent its people away and the region the world now travels to.

Where Wuyi Is Going

JUNO does not stand alone. It is one of the clearest signals that the Siyi region has been pulled into the orbit of the Greater Bay Area (粤港澳大湾区), the Pearl River Delta’s bid to weld Guangdong, Hong Kong and Macau into a single economic and scientific zone. The collaboration behind JUNO spans 17 countries and regions and around 75 institutions — a reminder that the international network the Wuyi diaspora built by emigration is now mirrored, a century later, by an international network built on science.

The inversion runs both ways. Once, Kaiping’s global connections ran through its people abroad. Today they run through its laboratories, its reactor coast and its transport links — while the villages the emigrants left behind, Chikan and the rest, are now heritage destinations that bring the descendants back. The region is learning to be known for more than the fact that it was once emptied of its people.

What Comes Next

JUNO is just getting started. Its design life is 30 years, and the numbers will only sharpen: the 2.3σ mass-ordering hint is expected to firm up into a definitive answer within a few years, settling the ordering question that has shaped the field for a generation. And because the detector’s cavern and its 20,000 tonnes of scintillator are built to be reconfigured, JUNO is designed to be upgraded into the world’s most sensitive neutrinoless double-beta-decay (无中微子双贝塔衰变) experiment — a search that, if it succeeds, would reveal whether the neutrino is its own antiparticle and explain why there is any matter in the universe at all. Answering that would take a detector even more sensitive than today’s JUNO, which is why the upgrade path is built in from the start: the cavern, the shielding and the 20,000 tonnes of scintillator are sized to be reconfigured rather than rebuilt. The experiment is designed to keep giving answers for three decades.

The collaboration has already announced that the next International Neutrino Conference will be held in Beijing in 2028. By then, the experiment buried under a Kaiping hill may have answered one of the deepest questions physics has ever asked — from a county whose name, for a diaspora reader, has always meant home.

Sources

1. Jiangmen Municipal Government — “江门中微子实验” (location, 700 m depth, 53 km geometry, 20,000-tonne detector): http://www.jiangmen.gov.cn/home/bmdt/content/post_3398028.html
2. Xinhua / 新华社 — “JUNO的首个物理成果” (Nature cover, 59 days of data, 1.6× precision, 40,000+ photomultiplier tubes): https://www.news.cn/liangzi/20260611/669163116a594522a1f96496b1bc7619/c.html
3. Guangming Daily / 光明网 — “江门中微子实验正式运行” (operation start 2025-08-26; 17 countries, 74 institutions, 700 researchers; 30-year life; 0νββ upgrade): https://m.gmw.cn/2025-08/27/content_38245879.htm
4. ScienceNet / 中新网 — “江门中微子实验发现中微子质量顺序倾向于正序的迹象” (Neutrino 2026, 22 June 2026, normal-ordering hint): https://news.sciencenet.cn/htmlnews/2026/6/567051.shtm
5. Caixin Global — “China’s Giant Neutrino Detector Delivers First Results With Record Precision” (first results 2025-11-19, 1.5–1.8× precision): https://www.caixinglobal.com/2025-11-20/chinas-giant-neutrino-detector-delivers-first-results-with-record-precision-102384735.html
6. 王贻芳 et al., JUNO Collaboration — “Reactor neutrino oscillation measurement based on JUNO’s first data”, Nature (cover article), 10 June 2026 (sin²θ₁₂ = 0.3092 ± 0.0087; Δm²₂₁ = (7.50 ± 0.12) × 10⁻⁵ eV²).
7. Institute of High Energy Physics, Chinese Academy of Sciences (中国科学院高能物理研究所) — JUNO project pages: http://www.ihep.ac.cn/