05/09/2026
On September 1, 2026, at the TeV Particle Astrophysics conference in Tendo, Japan, the LUX-ZEPLIN (LZ) collaboration presented a result that has the dark matter community talking without quite allowing itself to celebrate: a single event, buried in 2.84 tonne-years of data, that is difficult to explain with the known background processes considered by the experiment. It isn't a discovery. The background-only hypothesis is disfavored at a global significance of 2.6 sigma after accounting for the look-elsewhere effect, while the maximum local significance reaches 3.4 sigma among the dark-matter models tested — well short of the 5-sigma threshold conventionally used in particle physics before claiming a discovery. A significance around 3 sigma is often described as evidence, while 5 sigma is the conventional discovery threshold. This result sits between those levels, which is exactly why the LZ collaboration is treating it as an intriguing event rather than a discovery. The important distinction is that 2.6 sigma is the global significance, while 3.4 sigma is the maximum local significance found across the models tested. LZ is not claiming to have detected dark matter.
Here's what actually happened. LZ — one of the world's leading dark-matter detectors — sits nearly a mile underground at the Sanford Underground Research Facility in Lead, South Dakota, where roughly 10 tonnes of ultra-pure liquid xenon are used to search for extremely rare particle interactions. The detector's active target contains about seven tonnes of liquid xenon. Combing through 220 live days of data collected between March 2023 and April 2024, the collaboration searched an extended, higher-energy nuclear-recoil region reaching approximately 270 keV, designed in part to probe dark-matter models in which high-energy recoils are more prominent than in the simplest spin-independent WIMP scenario. Buried in that dataset was a single event recorded on June 16, 2023, at 21:22:39 UTC. It was reconstructed as an elastic nuclear recoil with an energy of 248 ± 23 (stat) ± 23 (sys) keV, in a region where the expected known background is low. The event's properties are consistent with a nuclear recoil, although the event-by-event information does not by itself establish that it was caused by dark matter. The LZ analysis finds a global 2.6-sigma tension with the background-only hypothesis after the look-elsewhere effect is taken into account, while the maximum local significance reaches 3.4 sigma in the models tested.
So what is dark matter, and why does one strange event matter this much? Every star, planet, and person we can see is made of ordinary matter, and ordinary matter accounts for only about 15% of the matter in the universe, with dark matter making up roughly the remaining 85%. More broadly, when dark energy is included in the universe's total energy budget, ordinary matter is about 5%, dark matter about 26%, and dark energy about 69%. Dark matter does not appear to emit, absorb, or reflect electromagnetic radiation at detectable levels. We know it exists because of its gravitational effects: it provides the unseen mass needed to explain, among other observations, why galaxies rotate much faster at their outskirts than their visible matter alone would allow, and why galaxy clusters exhibit gravitational effects that cannot be accounted for by their luminous matter alone. No dark-matter particle has yet been directly detected. That is exactly what experiments such as LZ are trying to accomplish.
The idea has a longer and stranger history than most people realize. Fritz Zwicky was not the first person to propose unseen matter, nor was he the first to use the idea of "dark matter." In 1906, French mathematician Henri Poincaré used the expression matière obscure — "dark matter" — while discussing the possibility of unseen matter in the Milky Way. Dutch astronomer Jacobus Kapteyn used the term "dark matter" in 1922, and in 1932 Jan Oort produced an important estimate of unseen matter in the vicinity of the Milky Way based on stellar motions. But in 1933, Swiss astronomer Fritz Zwicky made one of the most influential early cases for unseen mass when he studied the Coma Cluster and found that its galaxies were moving far too rapidly to be gravitationally bound by the visible matter alone. Zwicky inferred that the cluster contained much more mass than could be accounted for by what could be seen. He used the German expression dunkle Materie, or dark matter. His result did not immediately become the accepted explanation for the missing mass, and the scientific significance of dark matter developed only gradually as additional observations and theoretical developments accumulated.
It took Vera Rubin and her collaborators to make the missing-mass problem impossible to ignore. Working at the Carnegie Institution with instrument-builder W. Kent Ford Jr., Rubin measured how quickly stars and gas rotate within galaxies. Newtonian gravity predicts that, once most of a galaxy's visible mass has been enclosed, orbital speeds should generally decline with increasing distance from the center, much as planets farther from the Sun move more slowly than planets closer in. But Rubin and Ford's measurements of the Andromeda Galaxy, published in 1970, showed that the outer regions were rotating much faster than expected from the visible matter alone. During the 1970s, Rubin, Ford and collaborators extended these observations to many more spiral galaxies, and the same broad pattern appeared repeatedly: rotation curves remained approximately flat rather than falling as expected from the visible matter. The observations provided powerful evidence that galaxies contain large amounts of unseen mass extending beyond their luminous regions. Together with other independent evidence accumulated over subsequent decades — including gravitational lensing, galaxy clusters, the cosmic microwave background, and the growth of cosmic structure — this helped establish dark matter as a central component of modern cosmology.
None of that told anyone what dark matter actually is, and that's the question LZ is built around. The idea of a weakly interacting massive relic particle predates supersymmetry: in the late 1970s, particle physicists were already studying heavy weakly interacting particles as possible relics from the early universe. By the early 1980s, supersymmetric particles such as the neutralino had emerged as particularly attractive WIMP candidates. The basic idea behind the so-called WIMP miracle is that a stable particle with roughly weak-scale mass and interaction strength could have undergone thermal freeze-out in the early universe and naturally been left with a relic abundance of roughly the right order of magnitude to account for dark matter. That coincidence helped motivate decades of searches for WIMPs. It is also why this new LZ event is interesting to theorists rather than decisive: a roughly 248-keV recoil lies in a high-energy region that can be especially informative for models involving momentum-dependent or inelastic dark-matter interactions. If interpreted as a WIMP interaction, the event would favor a particle with a mass of at least roughly 200 GeV/c² and an interaction structure more complicated than the simplest spin-independent WIMP scenario. One possibility is inelastic dark matter, in which the incoming dark-matter particle can transition between internal states during the interaction, allowing a higher-energy recoil than would be typical in the simplest elastic-scattering picture. But that is a possible interpretation, not an established explanation of the event.
Direct detection has broken hearts before. In 2020, the XENON1T experiment reported an unexplained excess of low-energy electronic-recoil events that attracted enormous interest because several forms of new physics could potentially have produced such a signal. Its successor, XENONnT, subsequently performed a more sensitive search and did not reproduce the excess; the absence of the excess pointed toward trace amounts of tritium in the xenon as a likely explanation of the XENON1T result. That history is exactly why the LZ collaboration is being cautious now and why one unusual event is being treated as a hint rather than a discovery. The crucial difference is that the new LZ event occurs in a much higher-energy nuclear-recoil region, rather than being the same type of low-energy electronic-recoil excess seen by XENON1T.
There's also a remarkable bit of continuity to where LZ sits. The detector operates at the Sanford Underground Research Facility in the former Homestake mine in South Dakota, the same underground complex that became famous for Ray Davis's solar-neutrino experiment. Davis spent decades detecting neutrinos using a huge tank of perchloroethylene, a chlorine-rich liquid also used in dry cleaning. His pioneering solar-neutrino experiment successfully detected neutrinos from the Sun but found far fewer than expected. The mystery ultimately helped lead to the discovery of neutrino oscillations, showing that neutrinos change flavor and have mass. Davis received one quarter of the 2002 Nobel Prize in Physics, jointly with Masatoshi Koshiba, "for pioneering contributions to astrophysics, in particular for the detection of cosmic neutrinos." Today, the same deep South Dakota underground complex that once housed Davis's neutrino experiment helps shield LZ from cosmic-ray backgrounds as it searches for the vastly rarer interactions expected from dark matter.
Which raises the obvious question: has dark matter ever won a Nobel Prize? Not directly, because no dark-matter particle has yet been conclusively detected. But dark matter's fingerprints run through the Nobel Prize's history all the same. In 2019, James Peebles won half of the Nobel Prize in Physics "for theoretical discoveries in physical cosmology" — work that helped establish the modern cosmological framework describing the evolution of the universe, including the role of dark matter and dark energy. The Nobel Committee's presentation describes a universe in which ordinary matter accounts for about 5% of the total cosmic energy content, dark matter about 26%, and dark energy about 69%. Peebles's work was therefore deeply connected to the framework in which dark matter is a major component of the universe, even though it was not a prize for detecting a dark-matter particle. And then there is the prize that never came. Vera Rubin, whose observations of galaxy rotation provided some of the most influential evidence for dark matter, died in 2016 without receiving a Nobel Prize. Many scientists have argued that her foundational contribution deserved Nobel recognition, although the Nobel Foundation does not award prizes posthumously.
More than nine decades after Zwicky's famous Coma Cluster result, we still do not know what dark matter is. LZ's single event will not settle that question, and it may ultimately turn out to be a rare background event or an unrecognized detector effect rather than new physics. The collaboration itself is not claiming otherwise. But the event is scientifically interesting because it appeared in a region where the expected background is low, it has characteristics consistent with a high-energy nuclear recoil, and several classes of dark-matter models can produce spectra in which events like this become more plausible. LZ has already accumulated substantially more data than the 2.84 tonne-years used in this analysis, providing an important opportunity for future analyses to test whether this event remains unique or whether similar events appear. If additional events with compatible properties emerge, the statistical case could become much stronger; if they do not, the anomaly may fade away. For now, the scientifically accurate description is not "LZ found dark matter," but something more intriguing and more honest: LZ found one unusual event that does not yet have a convincing explanation, and the next data may tell us whether it was the first whisper of new physics or simply an exceptionally rare background.
(Photo credit: LUX-ZEPLIN (LZ) Collaboration)
