Tuesday, November 17, 2026

Who Pays To Keep The Electrons Cold? – Analysis



November 16, 2026
By Burak Oktenli


Key Takeaways:

China’s ENN Group reported proton–boron fusion reactions on its EXL-50U and aims for electricity by 2030, but detected reactions remain far from a net-power reactor.

The fuel’s appeal—mainly helium instead of neutrons—comes with severe losses from hard-to-fuse ions and electron bremsstrahlung, so keeping ions hot while electrons stay cool requires continuous, costly energy input.

Recent models show recirculating power can rise two- to threefold once full particle interactions are counted, meaning any claim of favorable gain must quantify the full energy ledger before electricity is promised.


Proton–boron fusion promises cleaner energy, but its most attractive operating condition raises an awkward question: how much power is needed to maintain it?

In September, China’s ENN Group reported proton–boron fusion reactions on its EXL-50U device and began building a larger research platform in Hebei Province. The company’s stated ambition is electricity from this fuel in 2030. That date is a target, not an achievement: a detected fusion reaction is still a long way from a power station supplying the grid.

The milestone adds to a genuine international research record. In 2023, a team from Japan’s National Institute for Fusion Science and the US company TAE Technologies measured proton–boron reactions in the Large Helical Device, using injected high-energy protons and boron powder. The researchers detected helium nuclei produced by the reaction. They did not demonstrate a self-sustaining reactor or net electrical output.

Those distinctions matter because proton–boron fusion has one of the most attractive propositions in energy science. Its main reaction combines a proton with boron-11 and releases roughly 8.7 million electronvolts of energy, mainly in three energetic helium nuclei, rather than the fast neutron that dominates the deuterium–tritium reaction. A future reactor could reduce some neutron-related engineering burdens. That does not mean a complete plant would be entirely neutron-free or without radiation and materials problems.

The trade-off is severe. Proton–boron fuel is much harder to fuse efficiently. The reacting ions need extraordinary energies, and the electrons accompanying them can shed considerable energy as X-rays through a process called bremsstrahlung, or braking radiation. Suppressing those losses is among the central challenges of the fuel cycle.

The trap of the two thermostats


One tempting answer is to make the ions very hot while keeping the electrons relatively cool. Hot ions improve the chance of fusion; cooler electrons can reduce X-ray losses. In a presentation, the two temperatures can look like independent settings on a pair of thermostats.

A plasma does not come with two independent thermostats. Ions and electrons continually exchange energy through collisions. Heat flows from the hotter population toward the cooler one. Keeping a large difference between them is not impossible, but it demands an explanation of the whole energy flow: how the ions are heated, what happens to the electrons, and where the energy ultimately goes.

That does not necessarily imply a literal refrigerator inside a reactor, powered from the wall. A particular plasma may maintain unequal temperatures through selective heating, radiation, transport and other effects. Some energy may be recovered or redirected. The requirement is more precise: any arrangement that depends on a sustained temperature difference must explain and quantify the physical processes maintaining it, including their losses and recirculating power.

The warning is not new. Todd Rider’s 1997 study of nonequilibrium fusion plasmas questioned the energy cost of maintaining particle populations away from equilibrium. A 2025 reassessment using particle simulations revisited part of that constraint for proton–boron conditions. In one model, an electron-to-ion temperature ratio near 0.4–0.6 was preferable; forcing the electrons still colder increased recirculating power and reduced calculated gain. These remain theoretical operating conditions, not demonstrated reactor performance.

The cost grows when the whole plasma is counted

A particularly useful 2026 result puts numbers on an adjacent problem. A study published in June in Plasma Physics and Controlled Fusion examined specially shaped proton-energy distributions intended to increase the probability of fusion. Its simplified kinetic model found potential improvements under favorable assumptions. But when collisions with electrons, boron ions and fusion-produced helium nuclei were included, the calculated recirculating power needed to maintain those distributions rose by roughly two to three times compared with an estimate based on proton–proton collisions alone.

That is not a measurement of the electricity required to cool electrons in a real plant, and it should not be presented as one. It is a model-dependent finding about maintaining a nonthermal proton distribution. Yet it illustrates the broader point: a promising plasma condition can be much more expensive once the other particles and their interactions enter the account.

The radiation problem itself is still being examined. In January, Princeton-led researchers modeled bremsstrahlung constraints on proton–boron inertial fusion. Their result concerned an imploding fuel target, not a magnetic-confinement reactor, and its numerical thresholds should not be transferred across those designs. The common lesson is that radiative losses and fusion-product behavior need to be counted before a favorable power balance is claimed.

Energy recycling is a research program, not a free arrow

There are scientifically serious proposals for changing that balance. Research on alpha channeling, for example, studies ways of redirecting energy from fusion-produced helium nuclei through plasma waves toward useful particle populations. Other work examines how to manage helium ash before it worsens pressure and radiation losses. These ideas could help, but none can be added to a reactor diagram as a lossless energy-recycling arrow.

A complete claim would identify the wave or other mechanism, the particles receiving the energy, the fraction transferred, the cost of generating and controlling that process, and its behavior in a real plasma with impurities and changing conditions. If it reduces the need for external heating, that saving belongs in the calculation. If it causes new losses, those belong there too.

The industrial stakes now reach well beyond one laboratory. China’s Helong-2 construction project is targeting a new experimental platform; the earlier Japanese–American result established a way to measure reactions in a magnetic device; and US and UK laboratories have announced plans to connect AI-enabled fusion research infrastructure. The last initiative concerns fusion research broadly, not a demonstrated proton–boron reactor. These are different milestones, but together they make a public standard for comparing energy claims increasingly valuable.

The experiment that should accompany the promise

For any future proton–boron proposal that relies on cool electrons and hot ions, the evidence should include both particle temperatures and distributions, their measurement uncertainties, radiation losses, collisional transfer, heating and energy-recovery pathways, and the stability of the operating regime.

 Researchers should report how long the condition persists, not only its best instant.

The power ledger should then distinguish fusion energy produced, energy circulating within the plasma, externally supplied heating and control power, and eventual electricity delivered after conversion and plant overhead. Scientific fusion gain is not the same as net electricity at the meter. Nor is the successful detection of a reaction proof of either.

There is no reason to demand a commercial plant from a laboratory experiment. Early research should be judged against the question it was designed to answer. But as a project moves from scientific feasibility toward claims about electricity, the accounting must expand with it. A reactor concept that relies on electrons staying cooler than the ions should state what maintains that difference and what the demonstrated power balance says about doing so.

Proton–boron fusion deserves ambitious research precisely because its possible benefits are substantial. The point is not to dismiss a difficult route to cleaner energy. It is to make its hardest assumptions testable before treating them as solved.

Every favorable assumption eventually receives an invoice. In proton–boron fusion, one of the most important invoices is still addressed to the electrons.


About Burak Oktenli
Burak Oktenli holds an MBA and a Master of Professional Studies in Applied Intelligence from Georgetown University. His research addresses the governance of authority in autonomous and AI-enabled systems, and his writing has appeared at the Modern War Institute at West Point, RUSI, RealClearDefense, RealClearMarkets, and Geopolitical Monitor. He is the author of Authority Architectures for Autonomous Systems, a ten-volume series on how authority in autonomous systems is delegated, monitored and recovered, at authority-architecture.me.
View all posts by Burak Oktenli →


No comments: