An experiment buried beneath a Canadian mine has just begun watching for signals so faint that they could reveal the famous matter never directly observed.
Its name is SuperCDMS SNOLAB. The facility lies more than 1.6 km beneath the surface, in the Creighton mine near Sudbury, Canada. This depth shields the detectors from much of the particle radiation arriving from space, which could produce false signals.
A SuperCDMS detector equipped with its electrical connections inside a copper housing.
Credit: SNOLAB
Researchers are searching there for dark matter. This invisible matter is inferred from its gravitational effects on galaxies and other celestial objects. It would account for about 85% of all the matter in the Universe, but its nature remains unknown.
SuperCDMS is primarily interested in hypothetical, very light dark matter particles. If one of them strikes the material in a detector, it should transfer only a tiny amount of energy to it. The instruments must therefore be able to detect minute disturbances.
To do this, the experiment uses 24 extremely pure silicon and germanium crystals, each about the size of a hockey puck. An expected impact would produce a very small vibration in the crystal, called a phonon, as well as a weak electrical signal. Superconducting sensors are tasked with measuring them.
These sensors must operate at an extremely low temperature, close to absolute zero. At this temperature, thermal disturbances become weak enough to search for the expected traces. The crystals are therefore placed in a cooling system colder than the surrounding space.
The device must also be protected from natural radioactivity. Several layers of copper, polyethylene, and lead surround the detectors. A barrier also limits the entry of radon, a naturally occurring radioactive gas. Even deep underground, these precautions remain necessary to distinguish a possible dark matter signal from ambient noise.
The data collection that has just begun is still a preliminary phase. It will make it possible to test the cooling system, the detectors, and the facility’s noise level. This first campaign will continue through fall 2026 and could already produce scientific results.
The experiment will then be temporarily warmed up for adjustments and maintenance operations. A new campaign lasting about a year is scheduled to begin in 2027, this time with the detectors operating at their intended sensitivity.