Read our COVID-19 research and news.

Updated: Are old secrets behind Lockheed's new fusion machine?

The defense firm Lockheed Martin sent tech geeks into a frenzy yesterday when it revealed a few scant details of a “compact fusion reactor” (CFR) that a small team has been working on at the company’s secretive Skunk Works in Palmdale, California. The company says that its innovative method for confining the superhot ionized gas, or plasma, necessary for fusion means that it can make a working reactor 1/10 the size of current efforts, such as the international ITER fusion project under construction in France.

Being able to build such a small and presumably cheap reactor would be world-changing—ITER will cost at least $20 billion to build and will only prove the principle, not generate any electricity. But with little real information, no one is prepared to say that Lockheed’s approach is going to spark a revolution. “You can’t conclude anything from this,” says Steven Cowley, director of the Culham Centre for Fusion Energy in Abingdon, U.K. “If it wasn’t Lockheed Martin, you’d say it was probably a bunch of crazies.”

The Lockheed team predicts that it will take 5 years to prove the concept for the new reactor. After that, they estimate it would take another 5 years to build a prototype that would produce 100 megawatts (MW) of electricity—enough for a small city—and fit on the back of a truck. A Web page with video on the Lockheed site even talks of powering ships and aircraft with a CFR.

Lockheed statements reveal little about the nature of the reactor. Aviation Week yesterday carried the most detailed account having interviewed the team leader, Thomas McGuire.

Fusion seeks to release energy from inside atomic nuclei by getting light nuclei, usually isotopes of hydrogen, to fuse together to form helium. The problem is that nuclei are all positively charged and so repel each other. To get them close enough to fuse it is necessary to heat a plasma—a gas of nuclei and electrons—to more than 100 million degrees Celsius so that the nuclei travel at high enough speeds to fuse when they collide with each other. The challenge in building a fusion reactor is to confine the plasma such that it does not touch the sides of its container, because its temperature would melt any metal. Most reactors, such as tokamaks like ITER, use powerful magnetic fields for confinement.

From Lockheed photographs of the CFR, it shows similarities to a magnetic configuration known as a cusp geometry, perhaps one known as a “picket fence.” The images show a series of ring-shaped electromagnets arranged in a row, like curtain rings on a rail. If it is a picket fence, then plasma would be confined along the axis running down the middle of the rings and the electromagnets produce a series of magnetic fields that bulge out toward the central plasma—a series of cusps. The effect of this is that if a charged particle near the axis moves outwards it starts to experience a magnetic field pushing it back. This is gentle at first but the farther the particle strays from the axis, the more strongly it is pushed back toward the center. This makes the confined plasma less prone to instabilities that plague other types of fusion containment.

Cusp geometries were first proposed in the 1950s by Harold Grad of New York University but were abandoned because experiments showed such machines would be leaky: Particles could escape through the gaps between one electromagnet and the next. Some cusp ideas have been revived in more recent devices such as the Polywell, which creates a 3D rather than linear cusp geometry. According to Aviation Week, the CFR would use superconductors in its electromagnets—not available to researchers in the 1950s—which would provide stronger magnetic fields and so improve confinement. Lockheed statements refer to combining the best parts of several confinement approaches. Cowley thinks they may also be using a technique called a field-reversed configuration (FRC), in which helical magnetic fields are induced in the plasma so that it confines itself. FRCs again date back to the late 1950s and 1960s but tend to be very short-lived, lasting on the order of a millisecond. “They’re probably trying to create an FRC inside a picket fence,” Cowley speculates.

*Update, 17 October, 10:53 a.m.: Three U.S. patent applications filed on 9 October by McGuire reveal more details about the reactor. It does appear to be some sort of cusp geometry device but more complicated than a picket fence. It also appears to have a structure known as a magnetic mirror at either end. This acts as a magnetic plug to stop particles from escaping along the axis of the device.

One potential problem with the device that has been pointed out by scientists who have spoken with ScienceInsider is that it appears to have electromagnet coils made from superconductor inside the reaction vessel. If they were in that position in a working fusion reactor, the superconductor would be destroyed by the high-energy neutrons that are a product of fusion reactions. Other designs that use high-temperature superconductors have more than a meter of shielding to protect magnets from neutrons, although researchers at the Massachusetts Institute of Technology believe this could be reduced to as low as 77 centimeters. Even if it was possible to reduce this to 70 cm and such shielding was added to Lockheed’s current design, researchers say it would make the device 18 meters across, not the 7 meters that the company is claiming.