In Richmond, British Columbia, pistons squeezed a lithium shell around plasma until it passed 12 million degrees, and the machine cleared fusion’s oldest temperature bar without a single magnet or laser
A ring of pistons slams inward at once.
For a few thousandths of a second, a shell of liquid metal collapses around a cloud of hydrogen gas inside a steel chamber in Richmond, British Columbia.
No laser fires.
Engineers have run this cycle enough times that the geometry and timing have become routine, even if each shot still demands a long list of calibrations beforehand.
No superconducting magnet hums to life.
Just metal, motion, and pressure.
What made this squeeze different from six decades of failed attempts is the question worth answering first.
How do you heat plasma without a laser or a giant magnet
Most fusion machines reach extreme temperatures by either firing powerful lasers at a fuel pellet or trapping plasma inside a cage of superconducting magnetic fields twisted into a donut shape. Both approaches work, but both demand exotic hardware and rare materials that are expensive to build at scale, which has kept the field tied to a narrow set of suppliers for decades.
The Richmond machine, called LM26, takes a blunter route borrowed from old ideas about mechanical compression. Instead of magnets holding the plasma in place for a long, slow burn, it forms a short lived magnetized plasma and then physically crushes it with brute mechanical force.
A ring of pistons drives a liquid lithium liner inward around the plasma in a sphere shaped chamber, squeezing it the way a fist tightens around a balloon. Because the method avoids delicate superconducting coils and intense optics, the company behind it has argued the approach could eventually be built from conventional industrial metals rather than specialized components. The October test pushed electron temperatures past the 1 keV mark, roughly 12.6 million degrees Celsius.
Inside the chamber where the squeeze actually happens
Picture a spherical steel vessel about the size of a small delivery truck, plumbed with ports for diagnostic instruments and ringed by a battery of pneumatic pistons. Each piston carries enormous force, and when they fire in near perfect unison, they drive a thin liner of molten lithium metal inward in milliseconds, crushing the trapped plasma to a fraction of its starting volume.
Instruments capture what they can in the narrow window before the chamber cools, then the machine resets for the next shot. Engineers have run this cycle enough times that the geometry and timing have become routine, even if each shot still demands a long list of calibrations beforehand.
The scale matters here too, since LM26 is built close to what a future commercial device would need, not a tabletop proof of concept.
What the instruments actually measured and who checked the numbers
The headline number, an electron temperature exceeding 1 keV, did not come from a single company claim. It was measured using a diagnostic called Thomson scattering, done in collaboration with the United Kingdom’s national fusion research authority, which independently verified the readings before any public announcement went out.
That outside check matters because 1 keV is not an arbitrary round number. The threshold is regarded as a key milestone because it demonstrates the ability to produce and control fusion relevant plasma temperatures, first reported on the Soviet T3 tokamak, announced in 1968 and independently verified by U.K. physicists using Thomson scattering the following year.
Reaching it again, nearly six decades later, with a mechanical squeeze instead of a magnetic cage, is what makes this result distinct. “Reaching more than 1 keV on LM26, alongside confirmed ion heating and corroborating AXUV data, is an important step forward for General Fusion and a meaningful validation of the approach we have spent more than two decades developing,” said General Fusion CEO Greg Twinney. The data has been submitted for peer review, a step still pending as of this fall.
Where the method still has to prove itself
One keV is a long way from the roughly 10 keV needed to approach the Lawson criterion, the combination of temperature, density, and confinement time that marks the edge of a reaction producing more energy than it consumes. While not the first fusion company to achieve this threshold, it is a first for General Fusion’s unique magnetized target fusion approach, which uses low-speed compression.
Rivals using lasers or superconducting magnets passed similar temperature marks years ago, so the achievement validates a different road rather than setting a new record outright. Even so, the mechanical approach carries its own tradeoffs worth watching, since piston timing and liner uniformity are notoriously hard to perfect at commercial scale, and each compression cycle currently produces only a brief pulse rather than sustained output.
The next milestone will test whether the same hardware, tuned further, can survive the far more violent squeeze that 10 keV demands.
What a passed threshold means for the decade ahead
For an industry used to incremental magnet upgrades and laser recalibrations, a result built on off the shelf engineering is notable mostly for what it implies about cost and manufacturability down the line. If a commercially relevant machine can be driven by pistons and lithium rather than cryogenic magnets, the bill of materials for a future plant could look very different from the one most of the field has assumed.
That same appetite for unlikely engineering solutions runs through other corners of the energy world right now. A suspended ceramic chip has multiplied a temperature gradient a hundredfold, while a retrofitted gas platform keeps splitting seawater into hydrogen without ever shutting down. None of these projects promise a finished product tomorrow.
Upgrades to LM26 are already underway for the next test, aimed at the far steeper climb to 10 keV. Whether pistons and molten metal can carry fusion that much further remains an open question, but for one week in October, in a steel chamber in Richmond, the squeeze worked exactly as the roadmap said it should.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
