A Canadian company is testing the world's first 'on-demand' muon beam

Ideon Technologies is part of an international consortium attempting, for the first time, to generate an artificial muon beam with a laser-plasma accelerator — an alternative to the cosmic-ray muons used today to map the subsurface.

Ideon Technologies, a Vancouver-based company specialized in muon tomography for mineral exploration, joined an international research consortium attempting, for the first time, high-resolution muon imaging using a laser-plasma particle accelerator. The experiment began in August 2026 and runs for five weeks at the ELI-NP accelerator in Magurele, near Bucharest, Romania.

Muon tomography works like an X-ray of the subsurface: muons are subatomic particles that pass through rock and other dense materials, losing energy in a predictable way, which allows density maps to be reconstructed at depth — including more than a kilometer down. The problem is the source. Until now, this entire technology has depended on muons produced naturally when cosmic rays collide with the Earth's atmosphere: a constant but weak, dispersed flow, outside the control of whoever is measuring it. That's why traditional muon tomography can take weeks or months to produce a useful image of a geological structure. The consortium Ideon is part of attacks exactly that bottleneck. The goal is to generate multi-GeV electron beams with a laser-plasma accelerator — equipment far more compact than a conventional particle accelerator — and convert them into directed muon beams, produced on demand. If it works, there's no longer a need to passively wait for the cosmic flow: the muon source can be switched on, aimed and intensified like any other laboratory instrument. Ideon has already received funding from ARPA-E, the U.S. Department of Energy's advanced research projects agency, for a related project targeting a staged laser-plasma accelerator with electron and muon beams of 20 to 25 GeV.

For mining, faster and more controllable muon tomography means shortening the time between 'where minerals might be' and 'where minerals actually are' — today one of the most expensive bottlenecks in mineral exploration, since drilling is costly and often guided by incomplete geological models. The technology is already being tested at operations such as Rio Tinto's Kennecott copper mine in Utah, to refine existing geological models, and also to locate hidden underground hazards, such as pockets of water or abandoned tunnels that don't appear on old maps. An artificial, on-demand muon beam points to a second generation of this technology — faster, more predictable and potentially more accessible as laser-plasma accelerators become more compact and cheaper than conventional accelerators. It's still experimental-stage research, not a tool ready for day-to-day use in a mine, but it's the kind of advance in applied physics that usually takes five to ten years to become commercial equipment — and it's worth watching from now.