LUX-ZEPLIN in South Dakota Detects Dark Matter Hint
A collaboration of 250 scientists says the LUX-ZEPLIN detector, buried deep underground in South Dakota, has recorded an event it cannot explain with known sources of radiation.
The international LUX-ZEPLIN (LZ) collaboration, roughly 250 scientists and engineers from 39 institutions, presented a result on September 1 at the TeV Particle Astrophysics 2026 conference in Japan that researchers cannot yet explain through known sources of background radiation. The finding is a single event recorded by a detector operating deep underground at the Sanford Underground Research Facility in South Dakota, a former gold mine. The result falls short of the statistical threshold needed to claim a discovery, but the collaboration says it is the most compelling hint of dark matter the experiment has produced so far. The paper has been posted as a preprint on arXiv and submitted to Physical Review Letters for peer review.
Under the current scientific consensus, dark matter cannot be observed directly, yet it is believed to make up about 85 percent of the mass in the universe, with the remaining 27 percent of the universe’s mass and energy split between ordinary visible matter and dark energy. Its existence is inferred from gravitational effects, chiefly from galaxies rotating faster than their visible mass would allow. LZ is built to search for so-called WIMPs, weakly interacting massive particles, one of the candidates that have been studied for decades alongside axions and MACHOs such as rogue planets and burned-out stars. Earlier astronomical observations have shown there simply aren’t enough of those objects to account for the missing mass, which keeps WIMPs among the leading candidates. LZ had already combed through this same dataset once before, looking for fainter signals from the simplest types of collisions between WIMPs and xenon nuclei. The new analysis widened the search to more energetic interactions, and it was in that broader range that the anomalous event turned up.
The analysis covered 220 days of detector operation, with data collected between March 2023 and April 2024. At the heart of the detector sits a cylindrical chamber holding 10 tons of ultra-pure liquid xenon, surrounded by a water tank and outer detectors that screen out background neutrons, while nearly a mile and a half of rock above shields the chamber from cosmic rays. Over that entire period, in the region where background radiation is lowest and where a dark matter signal would be expected, researchers recorded just one event whose nuclear recoil energy spectrum they could not attribute to any known source. The finding carries a statistical strength of 2.6 sigma, well below the 5 sigma threshold physics requires to claim a discovery. The estimated probability that the event is nonetheless the product of known background is about 0.5 percent. If it were caused by dark matter, calculations suggest the particle would have a mass above 200 giga-electron-volts, more than 200 times the mass of a proton, pointing to a type of interaction that falls outside the simplest model.
Rick Gaitskell, a professor at Brown University and spokesperson for the LUX-ZEPLIN collaboration, described the finding cautiously:
With just one event, we don’t want to get ahead of ourselves. We are not claiming that we’ve seen dark matter.
The University at Albany also helped build the tools used to analyze the event. A team there, including associate professors Cecilia Levy and Matthew Szydagis, developed the NEST simulation software that the LZ collaboration uses to separate possible background signals from dark matter candidates. Szydagis said in a University at Albany statement:
University at Albany scientists led the development of the simulation software known as NEST, which is used in the LZ experiment.
The detector is operated by Lawrence Berkeley National Laboratory, part of the U.S. Department of Energy. The collaboration’s next step is to analyze additional data collected since, which will show whether the signal grows stronger over time or fades into statistical noise.
What This Finding Does Not Prove
A single event is not a discovery, and the collaboration itself stresses that point. A statistical strength of 2.6 sigma leaves a real, if small, chance that this is a rare background occurrence researchers haven’t yet recognized as such, an estimated 0.5 percent by their own reckoning. LZ has not previously announced any finding reaching the 5 sigma threshold required to claim a discovery in particle physics, and even this result, as the collaboration’s own statement notes, is only now headed to peer review at Physical Review Letters. For the finding to be taken seriously as a dark matter candidate, standard scientific practice would require it to be reproduced, ideally in a different type of detector, such as one using liquid argon, and for the particle to show up under laboratory conditions rather than as a single trace in an underground detector. Until then, this remains exactly what the collaboration itself calls it: a hint worth watching, not proof.
