Most fusion headlines fixate on the reactor: the plasma, the magnets, the moment a machine supposedly beats breakeven. The less glamorous bottleneck sits in the power electronics that have to switch enormous currents on and off thousands of times a second without melting. That is the part Helion Energy is now being paid to solve, with the Everett, Washington company in line for a $175 million award to develop high-voltage semiconductor technology tied to its fusion work. The figure comes from regional business reporting, and the details of the award structure matter less than the signal: the U.S. is treating fusion as a supply-chain problem, not just a physics experiment.
This is a meaningful shift in how the fusion industry is being funded. For years, the assumption was that if you could get a plasma hot enough and confined long enough, the rest would follow. Reality has been messier. A working fusion plant is a brutal power-conversion system wrapped around a physics core. It needs components that can handle voltages and switching speeds well beyond what today’s grid-scale silicon was designed for. If those parts don’t exist, the reactor design has to compromise around them.
That’s where high-voltage semiconductors come in. Fusion machines, particularly the pulsed approaches Helion favors, depend on rapidly transferring huge amounts of energy into and out of magnetic systems. The switches that do this work, historically built on silicon, are increasingly being pushed toward wide-bandgap materials like silicon carbide and gallium nitride. These materials tolerate higher voltages, run cooler, and switch faster, which translates directly into smaller, cheaper, and more efficient power systems. The catch is that the most advanced versions aren’t cheap or mature, and the supply chain for the highest-performance devices is thin.
A government award of this size aimed at semiconductor R&D for fusion suggests Washington sees a chicken-and-egg problem worth breaking. Fusion developers can’t justify building custom chip fabs on their own; chipmakers can’t justify designing specialized high-voltage parts for a market that doesn’t exist yet. Public money can bridge that gap, in roughly the same way CHIPS-era funding has propped up other advanced semiconductor work. The recent news that Vertical Semiconductor received $73 million in CHIPS funding for related compound semiconductor work fits the same pattern.
There’s a strategic angle too. High-voltage power semiconductors are foundational to a lot of industries beyond fusion: electric grids, data centers, industrial motors, and electric vehicles. Building domestic capacity and know-how in this space has value even if a particular fusion company stumbles. That’s part of why these awards tend to be framed as industrial policy rather than pure science funding. The upside isn’t just a working reactor; it’s a stronger position in the power electronics supply chain at a moment when everyone is suddenly worried about where critical components come from.
For Helion specifically, the award is a validation of a strategy that has always been more engineering-heavy than its peers. The company has long argued that its approach, which uses pulsed magnetic fields and aims to recover energy directly from the fusion process, depends on power electronics as much as plasma physics. Critics have questioned whether the timelines its leadership has floated are realistic. A major federal award doesn’t settle that debate, but it does mean the company’s component-level bets are being taken seriously by people who evaluate supply chains for a living.
It’s worth keeping some perspective. Fusion remains a field where optimistic timelines have a long history of slipping, and a semiconductor award is not a demonstration of net energy gain. Announcements like this are easy to overread. The money is targeted at R&D, and R&D has a way of producing useful technology that doesn’t always look like the original goal. Even if Helion’s reactor plans change, better high-voltage switches would find buyers elsewhere.
The broader takeaway is that fusion’s future will be decided as much in semiconductor labs as in plasma chambers. The industry has spent a decade convincing investors it can build the physics. The next decade is about building the hardware around it, at scale, at a price the grid can absorb. Awards like this one are a quiet acknowledgment that the hard part isn’t only the star in the bottle. It’s everything that has to move the power in and out of it.