To build a fusion power plant, engineers must make some difficult choices. Do we go with a simpler design and force the plasma to not die out during operation, or do we choose a more complex design that is harder to build but leads to a happier plasma? ?
Or what if there was a way to do both?
Thea Energy hopes “both” is the correct answer. The startup is betting that software will replace manufacturing precision in its goal of providing reliable and cheap fusion power. The company recently raised $20 million in Series A funding, according to information obtained exclusively by TechCrunch. Prelude Ventures led the round, with participation from 11.2 Capital, Anglo American, Hitachi Ventures, Lowercarbon Capital, Mercator Partners, Orion Industrial Ventures, and Starlight Ventures.
There are two main approaches to fusion power generation: inertial confinement and magnetic confinement. The former made headlines in late 2022 for proving that net-positive fusion power generation is more than just science fiction by using giant lasers to vaporize fusion fuel pellets.
However, many startups use a variation of the former. In magnetic confinement, a burning plasma is confined by a strong magnetic field generated by a high-temperature superconductor. Tokamaks, a donut-shaped design used in many large-scale nuclear reactor projects, require magnets to be constructed with incredible precision to contain the plasma and keep it at the right temperature.
In stellarators, the magnets need to be more precise, but several startups are favoring this design because it's easier to achieve a stable plasma. Tokamaks are often compared to classic raised donuts. I like to think of Stellarator as an old-fashioned donut. Although it has an irregular shape, it is essentially a donut.
The entire stellarator twists and rotates according to the pre-calculated demands of the plasma. This shape comes from intentionally distorted magnets, and each magnet requires a great deal of engineering and manufacturing know-how to properly manufacture, all of which drives up costs.
The team at Thea Energy wanted to build a Stellarator, but didn't want to go through all that hassle. Instead, they are using an approach developed at the Princeton Plasma Physics Laboratory that consists of a series of high-temperature superconducting magnets arranged in a doughnut-shaped reactor, each controlled by software. By stretching and contracting the magnetic fields of the various magnets, this array allows the plasma to behave as if it were inside a more complex stellarator.
Of course, none of this is simple. There is nothing simple about fusion power generation. “We haven't eliminated complexity. We haven't eliminated precision,” Brian Berzin, co-founder and CEO of Thea Energy, told TechCrunch. “But what we've done is we've taken as much information as possible from the hardware and pushed it into the control system.”
Berzin likens the planar coil stellarator design to a computer display. Each magnet is like a pixel that can be controlled individually. They're creating stellarator shapes with inherent stability, so the computers controlling them don't have to be anything special. “What we're talking about is not even having to run a server cluster,” he said. “The near real-time computing we need doesn't exist.”
Thea believes its approach is better at confining plasma than competing designs. “Confinement has improved by an order of magnitude,” Bergin said. “We can create more precise stellarators than we can with modular squiggly coils.”
The modular approach should also speed up system development. The company is currently producing full-scale magnets at its research facility in Jersey City. By comparison, the magnets that form ITER's 64-foot tokamak are assembled in a vast warehouse in the French countryside. Thea's tiny magnets can be tested individually in the same lab or in small arrays that mimic parts of the final design.
“You can iterate multiple generations within a year without spending an exorbitant amount of money on a single piece of hardware,” Berzin said.
Thea plans to build a pilot-scale reactor later this decade, followed by a larger 350-megawatt demonstration plant in the 2030s. By the time a commercial product is connected to the grid, it hopes to be able to produce $50 per megawatt hour of electricity. Lazard said this represents the lower end of the current level of combined solar and battery power. Slightly more expensive than current combined cycle gas power plants, but slightly cheaper than coal. In other words, if Thea achieves its goals, it will have a competitive product.
As with all fusion power startups, the same caveats apply. This technology is incredibly difficult to master, so difficult that no one has yet achieved it on a commercial scale. Once that happens, the challenge will be to keep costs down so that nuclear reactors can compete with renewable energy and battery storage, which continue to fall in price. There are several ways to do that, but Thea's approach is smart enough that it might work. Software has successfully taken over many other industries. Why not try fusion too?