A Mile Under Ontario, 24 Crystals Colder Than Deep Space Just Started Hunting Light Dark Matter

More than a mile beneath the surface of Ontario, inside the Vale Creighton nickel mine near Sudbury, one of the most sensitive dark matter experiments ever built has started collecting data.

The Super Cryogenic Dark Matter Search at SNOLAB — SuperCDMS — has entered what the collaboration calls its early-science phase, running 24 ultra-pure silicon and germanium crystals, each roughly the size of a hockey puck, inside a refrigerator colder than the vacuum of interstellar space. Full-sensitivity operation is scheduled to begin in 2027.

"The search for dark matter at SuperCDMS SNOLAB is finally underway," said Tina Cartaro, the experiment's operations manager at the Department of Energy's SLAC National Accelerator Laboratory, which leads the 28-institution collaboration. "Even in this early phase, our most sensitive detectors have the potential to deliver breakthrough discoveries. At the same time, we're preparing and testing the entire system, learning how our detectors and cryogenic cooling perform together so we can unlock their design sensitivity."

The physics target is what the field calls light dark matter — hypothetical particles far less massive than the WIMPs that dominated searches for three decades. Dark matter accounts for about 85% of all matter in the universe, and the large liquid-xenon detectors that have led the hunt are best at catching heavier particles slamming into a nucleus. A very light particle carries too little momentum to make that kind of splash. SuperCDMS instead looks for phonons — quantized vibrations that ripple through the crystal lattice when something hits it — alongside a small electrical signal, a combination that lets the detectors register far smaller energy depositions.

Reading a signal that faint means the crystals are wired with superconducting sensors that only function at temperatures near absolute zero, and it means burying the whole apparatus under a mile of rock to block cosmic rays. Even that is not enough. The experiment sits inside nested layers of copper, polyethylene, ultrapure lead and a radon barrier. Part of that lead came from ancient ballast recovered from Mediterranean shipwrecks, prized because centuries under the sea have let its own natural radioactivity decay away — freshly smelted lead is too radioactive to shield a detector this sensitive.

The early-science run continues through fall 2026. After that, the collaboration plans to warm the experiment up to further optimize the cryogenic system and the noise environment, work expected to run into late 2026, followed by a full year of data collection at design sensitivity.

"Our detectors will explore, with unprecedented sensitivity, regions where the lightest-mass dark matter particles may be lurking," said SuperCDMS spokesperson Priscilla Cushman, a professor at the University of Minnesota School of Physics and Astronomy. "This opens up new avenues in the search for dark matter."

The stakes are less about confirming a favored candidate than about closing a door. Decades of null results from xenon experiments have pushed the community toward lower masses, and SuperCDMS is one of the few instruments built to look there. A signal would be the first direct laboratory detection of the substance that holds galaxies together. Silence, at this sensitivity, would itself be a significant constraint on what dark matter can be.