Helion Secures $225M to Build Custom Semiconductors for 2028 Fusion Plant

Key facts

Helion Energy closed a $225 million financing round dedicated to advancing two intertwined technical objectives: finishing its seventh-generation prototype and production-scale infrastructure, and developing proprietary high-power semiconductor devices. According to reporting by TechCrunch, the semiconductor initiative focuses on specialized solid-state switching components engineered to handle the extreme electrical currents and rapid pulse frequencies required by Helion's pulsed reactor architecture. If successful, the hardware modifications could double the effective power output of the company's planned commercial facilities without necessitating a proportional increase in reactor vessel size.

Helion's design relies on magneto-inertial fusion, a hybrid technique that departs fundamentally from the massive, steady-state magnetic confinement tokamaks pursued by international research consortia such as ITER in France. Instead, Helion forms two field-reversed configuration plasmoids at opposite ends of an elongated linear vacuum chamber, accelerates them toward each other at supersonic speeds using magnetic coils, and compresses them at the center using powerful magnetic fields to trigger fusion conditions at temperatures exceeding 100 million degrees Celsius.

Because the reaction produces charged particles that expand against the surrounding magnetic field, the process induces an electric current directly into the external magnetic coils. This mechanism bypasses the steam-cycle turbines and thermal cooling loops standard in traditional nuclear fission and coal-fired generation. However, the direct induction mechanism places extraordinary demands on the electrical power electronics that capture, regulate, and reinject the energy into the system. The newly announced semiconductor push addresses this electrical bottleneck by engineering solid-state switches capable of managing multi-megawatt pulses with minimal heat loss and microscopic response latencies.

Why it matters

Developing custom semiconductor hardware represents a high-stakes pivot from general systems assembly to component-level physics manufacturing. Power electronics have long represented one of the most stubborn friction points in high-repetition pulsed power engineering. Commercial off-the-shelf semiconductors, primarily designed for electric vehicle motor controllers, locomotive drives, and regional grid substations, are not rated to withstand the extreme voltage transients, high di/dt current slew rates, and cyclical thermal stresses native to pulsed fusion reactors.

If Helion succeeds in developing switches that double plant yield, the capital cost per kilowatt of installed capacity would fall dramatically. Nuclear facilities—both fission and theoretical fusion—suffer from high upfront capital expenditure relative to operational expenditure. A system that yields twice the saleable electricity per reaction chamber fundamentally alters the levelized cost of electricity (LCOE), making fusion far more competitive against utility-scale solar paired with long-duration battery storage and modern natural gas combined-cycle turbines.

Furthermore, the capital injection directly affects the broader commercial technology landscape. Hyper-scale cloud providers, led by Microsoft, Amazon, and Google, face an acute power supply crunch driven by the exponential energy appetites of generative artificial intelligence data centers. In May 2023, Microsoft signed a binding power purchase agreement with Helion to buy power from a 50-megawatt plant slated for 2028, with utility Constellation Energy acting as the power marketer. Commercial penalties and reputational credibility hang on meeting that deadline. The deployment of custom semiconductors indicates that Helion is aggressively optimizing the electrical back-end of its plants to meet contractual transmission quotas.

Helion was co-founded in 2013 by fusion scientists David Kirtley, John Slough, Chris Pihl, and George Votroubek, emerging from research conducted at the University of Washington and early development work supported by the United States Department of Energy's Advanced Research Projects Agency-Energy (ARPA-E). For roughly a decade, the venture operated in relative quiet compared to higher-profile magnetic confinement initiatives, steadily constructing a progression of experimental prototypes named Inductive Pulsed Compression devices.

The venture broke into broader public view following a $500 million Series E funding round in late 2021, led by Sam Altman, the chief executive of OpenAI, who contributed $375 million of his personal wealth and assumed the role of Helion's board chair. That funding round contained conditional commitments for an additional $1.7 billion tied to the achievement of key technical milestones, including net electricity generation and high-temperature fusion thresholds.

Helion's technical pathway is distinct in two crucial aspects: fuel choice and energy conversion. Most mainstream fusion projects pursue a deuterium-tritium (D-T) fuel cycle, which fuses at relatively lower temperatures (roughly 100 to 150 million degrees Celsius) but releases 80 percent of its energy in the form of energetic 14.1 MeV neutrons. These neutrons severely degrade reactor containment walls and induce secondary radioactivity in surrounding structural steel, while requiring thermal water blankets to absorb heat and drive conventional steam turbines.

Helion instead targets a deuterium-helium-3 (D-He3) fuel cycle. This reaction is predominantly aneutronic, releasing energy primarily through charged alpha particles and protons. The charged reaction products push outward against the magnetic containment field, enabling direct recovery of electrical energy via Faraday's law of induction. However, D-He3 requires ignition temperatures approaching 200 million degrees Celsius, and helium-3 is exceptionally rare on Earth. Helion produces helium-3 internally by capturing and decaying tritium generated from side-reactions in pure deuterium pulses. The engineering tradeoff shifts the primary technological challenge away from steam thermodynamics and toward solid-state power switching, high-voltage capacitor banks, and real-time magnetic field manipulation.

Reaction

Neither Microsoft nor Helion issued an immediate joint public statement elaborating on the specific contractual milestones associated with the fresh capital tranche beyond the technical details confirmed by TechCrunch. Industry analysts and fusion physics specialists are watching the development with a mixture of interest and technical skepticism.

Independent energy analysts have frequently noted that while Helion's direct inductive energy conversion is theoretically elegant, the durability of power electronics under continuous high-repetition pulsed magnetic fields remains unproven at commercial utility scale. Independent power producers and transmission system operators are expected to scrutinize Helion's interconnection filings as the 2028 timeline nears. In power generation markets, utilities must complete rigorous interconnection studies with regional transmission organizations (RTOs) years before injecting electricity into wholesale grids, a process that typically requires fully specified hardware and transient stability modeling.

Within the semiconductor sector, materials specialists anticipate that Helion's internal design work likely leverages wide-bandgap (WBG) or ultra-wide-bandgap materials, such as silicon carbide (SiC) or gallium nitride (GaN), which offer superior breakdown electric fields and thermal conductivity compared to standard silicon. Whether Helion intends to build an in-house fabrication facility or partner with specialized commercial pure-play semiconductor foundries to produce the chips remains an active question among industrial suppliers.

What we don't know yet

Crucial technical and corporate details remain undisclosed. TechCrunch's report confirmed the $225 million figure and the semiconductor focus, but the identity of the lead investors in this specific financing vehicle has not been made public, nor is it clear whether this sum represents a draw against the $1.7 billion milestone commitments established during the 2021 Series E round or entirely new equity.

Technically, Helion has not published peer-reviewed metrics demonstrating continuous net electrical gain (net energy out exceeding all system input energy) from its sixth-generation prototype, Trenta, or its seventh-generation machine, Polaris. While the company stated in 2021 that Trenta achieved bulk plasma temperatures of 100 million degrees Celsius, commercial viability requires sustained, high-repetition-rate operation over months without component degradation.

Furthermore, the precise technical specifications of the proprietary semiconductors—their material composition, switching frequencies, current-handling thresholds, and manufacturing supply chains—have not been detailed. It is also unclear where the 2028 Microsoft-tied power plant will be physically sited, what municipal permits have been obtained, or how the facility will handle grid interconnection and environmental reviews.

What to watch

Progress over the next 18 to 24 months will provide definitive signals regarding whether Helion can meet its 2028 commercial target. Observers should watch for operational milestones from the Polaris prototype, which was engineered to demonstrate net electricity recovery. The performance of Polaris's internal solid-state pulse-power switches will validate whether the new semiconductor architecture operates as claimed.

Second, regulatory filings in Washington state or adjacent power markets will reveal the plant's intended physical location. Because Helion avoids the large tritium inventories and neutron activation profiles of deuterium-tritium reactors, it falls under an alternative regulatory framework; the Nuclear Regulatory Commission (NRC) voted unanimously in April 2023 to regulate fusion systems under its 10 CFR Part 30 materials licensing framework rather than the burdensome Part 50/52 rules applied to commercial fission reactors. Observers should track state-level licensing applications under this framework to determine if physical construction has begun.

Finally, procurement milestones will indicate manufacturing maturity. If Helion signs advanced packaging or foundry agreements with established semiconductor fabricators, or begins taking delivery of specialized power-conditioning hardware at scale, it will indicate that the newly funded chip initiative is moving from computational modeling into physical deployment.

This report incorporates original reporting published by TechCrunch and author Tim De Chant.