Physicists running the LUX-ZEPLIN experiment have reported a single particle interaction they cannot explain with known background signals. The result is the strongest hint of dark matter the project has produced, though it falls short of a discovery.
Dark matter makes up roughly 85% of the mass in the universe, but no one has detected it directly. LUX-ZEPLIN, known as LZ, is one of the world's most sensitive detectors built to search for weakly interacting massive particles, or WIMPs.
A single flash, one mile down
The detector holds seven tonnes of liquid xenon in a tank buried about a mile beneath the Black Hills of South Dakota, at the Sanford Underground Research Facility. Rock above shields it from cosmic rays, and the xenon is kept extremely quiet so a rare collision stands out.
The unexplained event was logged on 16 June 2023, at 39 seconds past 3:22 in the afternoon. A new analysis found it hard to attribute to ordinary matter.
If the event came from dark matter, the WIMP behind it would have a mass of at least 200 gigaelectronvolts, or more than 200 times the mass of a proton.
Why it is not a discovery yet
The analysis reaches 2.6 sigma. That means there is roughly a 0.5% chance the event could be explained by known backgrounds. Physicists require 5-sigma significance, about a one in 3.5 million chance of a false signal, before calling something a discovery.
Single events are also hard to interpret. A detector can register an unusual flash for reasons that have nothing to do with new physics, from trace contamination to a rare combination of ordinary interactions.
The team presented the result at the 2026 TeV Particle Astrophysics conference in Japan and submitted a paper to Physical Review Letters.
What more data would mean
LZ is still collecting. More exposure time raises the chance that a real signal would repeat, which is the cleanest way to separate dark matter from noise. Researchers say the next data release will decide whether this hint survives.
Other experiments are chasing the same target with different technologies, and a confirmed signal from more than one detector would be far harder to dismiss.