After three decades of expensive government-funded research that has failed to produce tangible breakthroughs, nuclear fusion has gone from a promising source of effectively limitless power to something more like a punch line.

In the past year, that has started to change, however. Several privately funded companies and small university-based research groups pursuing novel fusion reactor designs have delivered promising results that could shorten the timeline for producing a prototype machine from decades to several years. Commercial power generation from fusion is still a long way off, but the outlines of such a reactor can now be perceived.

Traditional fusion research has centered on large, doughnut-shaped machines called tokamaks, which exert powerful magnetic fields to compress high-temperature plasma—roiling balls of charged particles that fuse to form helium, releasing large amounts of energy in the process. The challenge is to contain the hot plasma and keep it stable; the fusion reactors of today, such as the one at the International Thermonuclear Experimental Reactor (ITER) project in southern France, use giant coils of electromagnets that consume much more energy than the machine actually produces. ITER (pronounced “eater”), which draws scientists and funding from China, the European Union, India, Russia, Japan, South Korea, and the United States, is projected to cost dozens of billions of dollars to produce a working reactor sometime in the 2030s. Maybe.

Two recent developments, offering new and faster pathways to energy-producing fusion reactors, have galvanized the fusion community. Tri Alpha Energy, based in Foothill Ranch, California, said in early August that it has succeeded in keeping a high-energy plasma stable for five milliseconds—much less than the blink of an eye, but “half an eternity” on the scale of fusion reactions, according to chief technology officer Michl Binderbauer.

Tri Alpha, says Binderbauer, is bringing the principles of high-energy particle accelerators, such as the Large Hadron Collider, to bear on the problems of fusion reactors. Specifically, the team has built a device, 23 meters long, that fires two clouds of plasma at each other to form a ring of plasma. The magnetic field that holds the ring together is generated by the plasma itself—a technique known as a field-reversed configuration. The plasma is sustained by the injection of high-energy particles from accelerators.

The challenge for Tri Alpha’s design, says Binderbauer, is “hot enough and long enough”—keeping the plasma stable at a high-enough temperature to achieve energy-positive fusion. The recent experiment indicated that the company—which has attracted millions of dollars in funding from investors including Goldman Sachs and Vulcan, the investment fund of Microsoft cofounder Paul Allen—has solved the long-enough problem. Making the plasma hot enough is the next key challenge. Next year, Tri Alpha will begin building a new and more powerful version of its experimental device to test the process at higher temperatures.

At MIT’s Plasma Science and Fusion Center, a group headed by Dennis Whyte, a professor of nuclear science and engineering and the center’s director, and graduate student Brandon Sorbom published a conceptual design in July for a machine called the ARC reactor (“affordable, robust, compact”). The novelty of the ARC design is the nature of the electromagnets that confine the plasma. Using recently developed, flexible superconducting tapes made of rare-earth barium copper oxide, the ARC reactor can achieve magnetic fields with much higher amplitude—thus enabling a reactor design much smaller than other tokamak-based machines. The researchers also envision a liquid “blanket” surrounding the plasma that will absorb neutrons without damage and provide an efficient heat-exchange medium to produce electricity.

Increasing the amplitude of the surrounding magnetic field raises the amount of fusion power produced in the plasma to the fourth power—a dramatic increase that could lead to a commercial prototype in a matter of years, according to Whyte.

“It’s well known that you can make very compact devices if you raise the magnetic field to very high levels,” he says, “but the electromagnets had to be copper—no superconductor could tolerate that magnetic field.” Now the advent of advanced superconductor tapes could enable a compact reactor that produces fusion continuously.

Published in Fusion Engineering and Design, the ARC reactor paper stresses that, for the moment, it’s a conceptual design only. Whyte is hoping to attract funding to build an experimental machine over the next few years. Meanwhile a clutch of private companies, including not only Tri Alpha, but also Tokamak Energy, based in England, and Vancouver-based General Fusion, are working on related but different designs to bring fusion to the prototype stage (see “A New Approach to Fusion”).

“We are getting closer to working machines,” says Michel Laberge, the founder and chief scientist at General Fusion. “For many years, fusion research was the realm of big government labs that did great work and established the basis for fusion to work. But there was not a great sense of urgency.”

Now the urgency has risen, and these companies are testing new ideas and new approaches—and attracting the investment to do so. General Fusion recently landed $27 million in new funding from a group of investors led by the sovereign wealth fund of Malaysia.

“Right now what’s happening is a rethinking,” says Burton Richter, who won the Nobel Prize in physics in 1976 and is an advisor to Tri Alpha. Budget cuts in the 1990s forced the shutdown of alternative approaches outside of ITER and the U.S. Department of Energy’s National Ignition Facility. Companies like Tri Alpha offer a path to fusion paved not with taxpayer dollars but with private-sector money—which ultimately is the only way to actually get something built.

Jonathan Menard, a plasma physicist at the Princeton Plasma Physics Laboratory, directs the National Spherical Torus Experiment, which is pursuing a tokamak shaped like a beach ball instead of a doughnut. Menard, whose own program recently completed a $94 million upgrade of its experimental machine, has closely followed developments with the Tri Alpha and ARC efforts and believes that these innovations should be pursued further.

With the wariness of a veteran fusion scientist, though, he advises caution: “Till you build it, you don’t know for sure.”