Companies

Commonwealth Fusion Systems raised another $1 billion of equity. Still no public information on what the total company valuation is. Oddly, this was the first round where specific investors were not named. It was mentioned that the round included pension funds, sovereign wealth funds, infrastructure investors, and industrial corporate partners, which are classes of entities that have participated in previous rounds. While the round raised last year was explicitly described as funding to complete SPARC, this round has been identified as for ARC development, site prep, and supplier development.

Tokamak Energy publishes its fusion pilot plant pre-conceptual design as part of the US Department of Energy Milestone Program. First of all, I congratulate the team on pulling this all together; it was no small feat. For those not in the know, Tokamak Energy was founded many years ago on the thesis that HTS was a game changer (same as CFS) and that a low-aspect-ratio tokamak was advantageous over a high-aspect-ratio tokamak. Not a big surprise to me, but incorporating all of the power plant considerations into the design has increased its size. The company’s first patent had a tokamak with a major radius ≤1.5 m, with 1.35 m as the main first design point. The major radius of the pre-conceptual design, ST-E1, has increased to 5 m. The diameter of the largest poloidal field coils for ST-E1 is about 30 m; for comparison, the largest ones on ITER are about 24 m. The increase in size is in no small part a result of (1) a design choice to have horizontal maintenance of machine wedges between TF coils and (2) the need to have sufficient nuclear shielding of the inner legs of the toroidal field magnets to allow them to be lifetime components. These, along with all of the integrated physics, engineering, and economic requirements, made for a big machine. Also interesting is that aspect ratio is not a big controller of the final economics of the machine: with an increase from 1.9 to 2.3, the overnight cost only increases by 5%. To me, this, along with other projects like STEP, demonstrates that going to low-aspect-ratio is not a strong lever for making tokamaks smaller. One of the potential advantages of low-aspect-ratio tokamaks is that much of the plasma current is self-driven by the plasma, although CFS’ design is a potential counter to this in showing that pulsed operation with a central solenoid could be viable as well. The final point to note is that Tokamak Energy is relying on a high normalized confinement (H98 for those in the know) compared to the empirical scaling to achieve good economics, which is an optimistic assumption.

Commonwealth Fusion Systems passed the Department of Energy Milestone Program milestones on the pre-conceptual design and technology roadmap for ARC. No info on the details yet.

Inertia, the fusion company founded by the owner of The Onion, warns about extrapolation in the fusion industry, unveils Inertia’s plans for extrapolation. The chest-thumping of NIF performance over others is a bit heavy-handed, but the assessment of the 10 miracles to demonstrate in “phase 1” for them is relatively on target. No fusion concept it without its challenges.

Fuse claims a record neutron yield of 1.27×10¹² in a single shot for a fusion company, which is false: SHINE has demonstrated 3.7×10¹³ neutrons per second effectively in steady state.

Fuse signed a five-year Cooperative Research and Development Agreement with the Nevada National Security Site. It’s a framework for collaboration.

Terra Fusion reported to be raising a $10 million seed round.

Zap reports its liquid metal loop ran for 1,000 hours and updates its website. I cannot stand all the moving stuff. Is this my “I’m an old man” moment or is that just bad design?

Starlight Engine opened a nationwide Japanese search for the site of its FAST superconducting-tokamak fusion demonstration project and stated that it expects its reactor project will cost 1 trillion yen ($6.3B).

MIFTI expands LLNL collaboration. Includes alpha particle modeling and neutron diagnostics for MIFTI’s z-pinch.

Thea won a $20 million ARPA-E SCALEUP award to scale up its superconducting magnet manufacturing.

TAE Technologies signed an agreement with Black Moon Energy Corporation for helium-3 supply. Fantasy meets fantasy.

Oak Ridge National Laboratory licensed pellet-fueling technology to Type One Energy. If I’m in the market for a frozen pellet shooter, ORNL is the first place I’m shopping.

American Fusion disclosed nineteen patent applications and six patent applications. Surprised they have such a large IP budget for a penny stock company.

American Fusion appointed Texas Tech researcher Noah D’Amico as a strategic advisor. Noah, according to American Fusion, has expertise in nuclear measurements and a 4.0 GPA.

American Fusion releases its SEC Form 10-Q. I find this whole thing fascinating and am a bit disappointed I don’t have the time to fully understand what’s going on here. Some of you have reached out to me with rumors, but I won’t pass them along because they are not (yet?) supported by public evidence. In this 10-Q, they claim $58,000 of revenue, which was probably the previously reported phase-noise analyzers sold to Canada’s Department of National Defence. For the first half of 2026, AMFN spent $598,748 on corporate communications and marketing, $573,233 on consulting, $423,938 on professional (patent?) fees, and just $131,342 on R&D!

General Fusion filed a couple of SEC Schedule 13Gs and Form 3s that report on large ownership and option stakes, including: PenderFund Capital Management, a Vancouver investment manager that’s owned General Fusion shares since 2021 and has a board seat; Meteora Capital, an event-driven investment firm that has exited its position entirely; BDC Capital (Canadian Government); Norman Harrison (Board Director); Michel Laberge (founder and CSO); and Robert Crystal (VP Finance).

TAE+TMTG had an earnings call. The audit is the reason the merger S-4 has not yet been filed; it still targets closing by the end of 2026. TMTG emphasized that TAE and fusion will be the combined company’s main long-term business and capital priority.

TAE+TMTG filed SEC Form 10-Q. TMTG is now lending crypto to third parties. The Truth Social spin-off is abandoned for now. Q2 revenue was $1.67 million, up from $0.87 million in Q1. Its directly held Bitcoin and Cronos had a cost basis of $1.12 billion but a present value of only $597.7 million.

TAE+TMTG filed a couple of SEC Form 8-Ks: announced earnings call; management priorities: close TAE merger in Q4, tighten treasury management, grow media and monetize data; sees fusion as the largest component of the company after the deal; management thinks digital-asset treasury companies and prediction markets have become crowded; abandons Crypto.com prediction-market integration; and termination of the Crypto.com/Yorkville CRO treasury SPAC.

People

Dennis Whyte appointed UKAEA CEO. This is a big win for the UKAEA. Dennis has great, practical (as practical as one can be in fusion) vision and leadership. This signals to me that the UK’s present and near-term vision and funding are stronger than the US’s.

With Dennis moving to the UK, Guinevere Shaw steps into the CEO role at Rutherford Energy Ventures. The co-founders (Dennis, Andrew, and Maria) transition into Strategic Advisor roles.

Proxima Fusion appointed Ahmed Diallo as Head of Global Partnerships. Ahmed is great for this role. He’s done a lot of work at PPPL and the US Department of Energy connecting public and private fusion.

First Light Fusion promoted experimental physicist Jonathan Skidmore to Chief Scientist. Jonathan has spent the last 10 years climbing the ranks at First Light.

Government

Germany announces three national fusion hubs. Three hubs; I suppose this makes it the tricycle strategy for fusion? Marvel Fusion and Focused Energy work on the VEGA laser fusion hub. Proxima Fusion, Gauss Fusion, and the Max Planck Institute for Plasma Physics work on the STRIDE magnetic fusion hub. Karlsruhe Institute of Technology works on the MAT-TRIX fusion materials and fuel cycles hub. Starting support is 125 million euros from the Federal Ministry for Research, Technology and Space.

DOE announces awardees for the Genesis Mission. Looks like a lot of projects funded for the application of AI/ML to specific topics.

The White House’s new science policy highlights fusion as a high-level mission: “Demonstration of commercial fusion power in the United States by the mid-2030s, following the Department of Energy’s Fusion Science & Technology Roadmap.” Haven’t had the chance to read over the whole policy yet.

Fusion Industry Association and Clean Air Task Force release their principles for codes and standards for fusion. Little I disagree with in there.

ORNL and Rutherford Energy Ventures officially begin their work together. They’re starting on an 18-month project with REV helping ORNL that will result in a “technical blueprint, financial framework and operating model for a pilot Fulcrum at ORNL,” which aims to be a facility that helps address a cross-cutting issue for the fusion industry on DOE’s roadmap.

ITER announced installation of its sixth vacuum-vessel sector. The vacuum vessel is now two-thirds in place.

The UK Parliament opened an inquiry into the proposed national planning framework for fusion power infrastructure. This would give fusion plants national-priority status and broad siting freedom. Interesting to watch this as an outsider to how things like this work in the UK.

UKAEA announced £40 million for more fusion education at British universities. Good workforce development.

UKAEA set revised priorities for the government-funded JET decommissioning and repurposing program.

China’s National Natural Science Foundation held a strategic forum on fusion-reactor materials that should inform the work that gets done in the next five-year plan.

General Fusion stock starts trading. Realta announces move to former hot dog factory. F4E publishes Global Fusion Energy Trends report.

Companies

General Fusion completes deSPAC with Spring Valley Acquisition Corp. III. General Fusion began trading on Nasdaq on July 13 under GFUZ.

General Fusion files SEC Form 20-F with details on how the deSPAC closed. The main news is that 91.6% of the SPAC investors chose to take their money back instead of keeping shares, redeeming about $217m.

General Fusion still ended up with about $170m in cash, most of which came from private investors and money the company already had.

The private investors got a very favorable deal: a 12% annual return that can be added to what General Fusion owes them, plus warrants that let them buy more shares at $12 if the stock’s value goes up. If General Fusion’s stock falls or the company later raises money at a lower price, those investors can receive more shares, diluting ordinary shareholders.

The lead investor also received 3.5m shares for almost nothing. After five years, these investors can ask General Fusion to repay their preferred shares in cash, which could exceed $225m. The deSPAC bought General Fusion time, but at a price: future shareholders face substantial dilution, and the company may need another major financing or refinancing by July 2031.

Meteora Capital, a specialist in event-driven SPAC trades, did some large trades around the General Fusion deSPAC, as noted in this SEC Schedule 13G. SEC Form 25 indicates the removal of Spring Valley from Nasdaq, part of the process of listing General Fusion. SEC Form 4 indicates that the merger was completed on the previously negotiated terms. An SEC Schedule 13G indicates that RichRich Capital sold its entire position in General Fusion.

SEC Forms 3 were filed for General Fusion’s leadership, indicating their stock and option interests and compensation in the company, including Gregory Twinney, Wendy Kei, Thomas Boehlert, Klaas de Boer, Jan Laishley, Mark Little, Ross Donaldson, and Megan Wilson.

The Financial Times reports on General Fusion going public, including some comments from me and Tony Donné.

TAE+TMTG have released a series of SEC Forms 8-K; nothing relates directly to the fusion half of the business: George Holding has resigned from the TMTG Board of Directors; TMTG announced a new API that allows paid, quicker access to information; and TMTG has resolved all claims with Patrick Orlando and ARC Global Investments II LLC, the former CEO and the SPAC that took TMTG public.

Realta Fusion announced that it is moving its headquarters to the former Oscar Mayer plant in Madison. The city and state have committed up to $55M to support this move, including tax exemptions and credits. Next week, CFS will announce its relocation to the Dunkin’ Donuts bakery and Proxima to the Rischart Bavarian bakery. Who’s next to move to a facility that made food in the shape of its plasma?

Commonwealth Fusion Systems posted a SPARC project update. CFS said its cryogenic plant is operating and circulating cryogenic fluid through magnet-cooling subsystems, and that the upcoming “dry dress rehearsal” will run the plant as if it were powering a fusion pulse while tokamak assembly continues in parallel. Standard plant commissioning work: bringing systems online for operation.

StandardX comes out of stealth. From Kyoto Fusioneering co-founder Richard Pearson and nuclear physicist Ross Allen comes “the world’s first isotope refinery: a single accelerator platform that produces the full portfolio of isotopes from one machine.” They plan on using a fixed-field alternating-gradient accelerator. I am seeing a lot of companies planning on isotope production. See the StandardX website for the menu of isotopes they would like to sell to you, including tritium.

Plasma-control startup Fusionality comes out of stealth. It was co-founded by Federico Felici, a longtime plasma-control expert, and Jonas Buchli, a robotics expert, both of whom are research scientists at DeepMind.

The Onion owner’s fusion company, Inertia, opens its headquarters in Livermore, CA. The six-month conversion of the facility marks a major advance in humanity’s decades-long effort to construct the building where fusion might happen.

Larmor mmWave Systems receives $50k to commercialize gyrotrons for fusion, co-founded by Massimo Carpita and Mack Van Rossem of EPFL.

Pacific Fusion’s pulsed-power prototype completes 3,000 shots. An important achievement in scaling up its system. At about 1 Hz for its power plant, 3,000 shots would take about an hour, so another factor of 10,000 is needed to reach a one-year-equivalent shot lifetime. I wonder whether reaching this rep rate was the major hurdle or whether a lot of effort will be needed to progress through the next couple of decades of shots.

Kyoto Fusioneering announced a strategic partnership with NGK to jointly develop and commercialize FLiBe. NGK, a leading producer of beryllium copper, should help a lot with the beryllium-handling side of things.

Type One Energy hosted an open house near its planned Tennessee facility. More than 150 people attended.

Starlight Engine has raised $37m to lead the Japanese FAST project.

Barclays joins Type One Energy, Tokamak Energy, and AECOM in the UK Infinity Fusion Consortium.

Eternafusion gets $1.6m to develop tokamak current-injection technology.

Avalanche announces the availability of a 300 kV feedthrough product based on its high-voltage R&D.

American Fusion moves its HQ to North Texas and announces patent filings, states that its CTO is presenting its fusion concept at the APS Division of Particles and Fields meeting (the standard APS meeting for this would be the Division of Plasma Physics meeting), where his abstract states that he will talk about black holes, reports that it has received a Certificate for Industrial Radiation Machines from the Texas Department of State Health Services, and has up to 1 GW covered for research purposes.

Government

F4E publishes the first Global Fusion Energy Trends report. This is a really nice compilation of the state of fusion development internationally, covering China, the EU, Japan, the UK, and the US. There are really useful figures placing machines under construction in the context of what has already been achieved and showing timelines for planned facilities. It also shows data on things like publications and patents (China is patenting a lot, but only domestically!). Finally, there’s a comparison of regulatory timelines in the major regions. We’re seeing different strategies unfold in each region. Which one will be the most successful at developing fusion?

China requests public comment on the draft “Safety Analysis Work Guideline for Deuterium-Tritium Fusion Experimental Devices”.

New facility in Denmark to test remote-handling robotics for fusion.

New research consortium tests materials for fusion power plants. It combines expertise and experimental capability in high heat flux, neutron irradiation, and material characterization.

UKAEA published its procurement pipeline for those wanting a window into what it plans to procure and do.

F4E opened applications for its Supply Chain Excellence Award.

People

Nancy Lipson has joined Zap Energy as Chief Legal Officer. Nancy spent most of her career as Deputy Counsel and then Chief Legal Officer of Newmont Corporation, a publicly traded mining company and the “world’s leading gold company.” Perhaps this is a sign that Zap is thinking about going public.

Derek Sutherland has been promoted to Chief Science Officer at Realta. Derek has one of the most diverse backgrounds in private fusion.

Gauss Fusion appoints Dr. Uwe Bau as its new CEO. Uwe has a PhD in nonlinear control, worked at McKinsey, and was the head of the semiconductor equipment business section at RI Research Instruments.

Verdict: Did they directly convert plasma kinetic energy into electricity? Probably. Were they the first private fusion company to do so? Probably not. Was it at a power level relevant to fusion reactors? Not at all. Rather than being a significant technical achievement, what Realta has done here is highlight the arbitrariness of some milestones.

What happened

On June 30, Realta announced that it had demonstrated direct energy conversion on the Wisconsin HTS Axisymmetric Mirror (WHAM) during experiments on June 19. Charged particles escaping through the mirror ends passed through two grids and reached a positively biased ion collector. Slowing and collecting the ions produced multiple amps at around 100 volts.

Realta was clear that this was not net fusion electricity. Nearly all the recovered energy had first been supplied by WHAM’s plasma-heating systems, not by fusion reactions.

Why it matters

Mirrors are well suited to direct conversion because they naturally channel escaping charged particles toward their ends, which are much freer than toroidal systems to expand as needed for direct energy conversion.

At low fusion gain, a converter could recycle some externally supplied heating power. At high gain, it could recover part of the alpha-particle energy. But in D-T fusion, approximately 80% of fusion energy appears in neutrons and only 20% in charged alpha particles. Most of the plant must still use a blanket and thermal cycle.

What is actually new

Very little.

Direct energy conversion has been demonstrated in magnetic mirrors. Lawrence Livermore began work on plasma direct converters in the 1970s, with tests reaching 100 kilovolts and 6 kilowatts. In 1983, a converter on the Tandem Mirror Experiment recovered 67 watts from approximately 140 watts of incident ion power, for a measured efficiency of 48%. A separate experiment with a 22-stage converter had previously reported approximately 86% electrostatic recovery.

Realta’s claim is narrower: that no private fusion company had previously converted plasma kinetic energy into electricity. Putting on my pedantic physicist hat, that is false as written.

The Migmacell II counterexample

While there may be earlier counterexamples, this was the best evidence I could find against the claim that Realta was “the first private fusion company in history to successfully demonstrate direct energy conversion (DEC) of plasma kinetic energy into electricity.”

In 1977, the Migma Institute and Fusion Energy Corporation published “Fusion products detection system in Migmacell II”. They used silicon detectors to measure protons and tritons from D-D reactions involving 0.55 MeV deuterons. The charged fusion products deposited kinetic energy in the silicon, creating electron-hole pairs and electrical charge pulses. Fusion Energy Corporation had therefore converted charged-particle kinetic energy into electricity 49 years before Realta. The output was microscopic and diagnostic. Nonetheless, it meets the criteria of Realta’s claim.

What the numbers imply

The following examines how direct energy conversion affects the ultimate plant power output.

Let $\eta_{\mathrm{DC}}$ be the direct-conversion efficiency and $\eta_{\mathrm{th}}$ the thermal-cycle efficiency. If the power not recovered directly becomes useful heat, the effective efficiency of the converter stream is:

$\eta_{\mathrm{stream}}=\eta_{\mathrm{DC}}+\left(1-\eta_{\mathrm{DC}}\right)\eta_{\mathrm{th}}$

The first term is electricity recovered directly. The second is electricity produced from the remaining heat.

If $f_{\mathrm{DC}}$ is the fraction of total fusion power delivered to the converter, gross plant efficiency is:

$\eta_{\mathrm{gross}}=\left(1-f_{\mathrm{DC}}\right)\eta_{\mathrm{th}}+f_{\mathrm{DC}}\eta_{\mathrm{stream}}$

Substituting the expression for $\eta_{\mathrm{stream}}$ gives:

$\eta_{\mathrm{gross}}=\eta_{\mathrm{th}}+f_{\mathrm{DC}}\eta_{\mathrm{DC}}\left(1-\eta_{\mathrm{th}}\right)$

The improvement over an all-thermal plant is therefore:

$\Delta\eta_{\mathrm{gross}}=f_{\mathrm{DC}}\eta_{\mathrm{DC}}\left(1-\eta_{\mathrm{th}}\right)$

This shows why a high converter efficiency does not produce an equally large increase in total plant efficiency. The benefit is multiplied by the fraction of reactor power that actually reaches the converter.

Using Realta’s optimistic assumptions:

$f_{\mathrm{DC}}=0.20$

$\eta_{\mathrm{th}}=0.45$

$\eta_{\mathrm{DC}}=0.90$

The effective efficiency of the converter stream is:

$\eta_{\mathrm{stream}}=0.90+\left(1-0.90\right)(0.45)=0.945$

Whole-plant gross efficiency becomes:

$\eta_{\mathrm{gross}}=0.80(0.45)+0.20(0.945)=0.549$

Thus, 90% direct conversion of the entire alpha-particle energy stream raises gross efficiency from 45% to 54.9%, not to 90%.

What would have to be true

For direct energy conversion to be useful in a magnetic mirror:

• A large fraction of alpha and injected heating power must leave axially as collectable charged particles.
• The converter must remain efficient across broad particle energies and pitch angles.
• Grids, collectors, and insulators must survive high voltage, particle flux, radiation, and harsh plasma conditions.
• The converter must not compete with the plasma potentials and expander geometry needed for confinement and stability.
• The value of the recovered electricity must justify the additional system’s capital cost and maintenance burden.

What is still missing

Realta has not released any substantial information on the plasma conditions under which this energy conversion was achieved. The fusion contribution was presumably negligible. This is only their first step along the technology roadmap.

To put it into perspectiv: External heating power is about 1 MW on WHAM. The results presented here are about 500 W. Scaling to WHAM-relevant conversion of plasma kinetic energy into electricity would therefore require an increase of about a factor of 1,000, followed by another factor of about 100 to reach reactor-level conditions. These are very large leaps in plasma conditions and power density, and achieving them will be no small technological feat.

This achievement is on par with the high power plasma diagnostics. For example, the Langmuir probes I used on the Alcator C-Mod tokamak operated at up to a few amps and a few hundred volts, the same regime as Realta’s first generation energy conversion system.

The claim beneath the claim

Good milestones matter.

As I stated in last week’s Fusion Fortnightly, I see this as an attempt by Realta to scoop Helion’s milestone of demonstrating the first direct conversion of plasma energy into electricity. I find this announcement a good-natured way of taking the wind out of Helion’s sails over a milestone that makes a good headline but might not be a substantial technical achievement.

This is an opportunity to raise a broader point about fusion-industry milestones: Many are set, achieved, and communicated in ways that make them sound bigger than they really are. This is a problem. Fusion energy is a complicated field, and those outside it have a hard time separating meaningful progress from modest milestones.

What Realta has done here is demonstrate that an arbitrary milestone such as “the first X to do Y” can sometimes be achieved relatively easily in a way that represents only a small step toward what a fusion reactor actually requires. Here, I have taken that a step further and shown that the supposed “first” had already occurred decades earlier in an even more trivial fashion.

Bottom line

Realta performed a real and worthwhile experiment, but wrote the historical claim too broadly.

Proxima raises €411M. Realta claims first direct energy conversion. Shareholders approve General Fusion deSPAC.

Companies

Proxima Fusion raises €411M ($468M) round. They were kind enough to release the news just before the Fusion Fortnightly publication deadline. The company is now valued at €2.4B ($2.7B). This is complemented by matching funds of €400M from the state of Bavaria and is expected to be supported by more from the German federal government. Funds will be used to finish their stellarator model coil as well as build up their internal capabilities in preparation for Alpha, their €2B net-energy stellarator (I predict it will cost more).

Realta claims first direct energy conversion. The system gathered “multiple amps of current at around 100 volts.” This feels to me like Realta did this entirely to beat Helion to be the first, not because it was a critical demonstration; and I kind of like that. Direct energy conversion is not new in mirrors, having been considered for many decades and first demonstrated at least 45 years ago. As I stated last time this came up with Helion: basically every fusion experiment has had some sort of “direct energy” conversion from the input and fusion energy in the plasma to electricity. That’s what most (all?) measurement systems do. The largest demonstrated conversion of energy in a plasma to electricity is likely with the central solenoid in tokamaks when the plasma current is ramped down at JET with a 7 MA plasma, which is about 50 MJ of stored inductive energy at probably about 1 MW.

Shareholders approve General Fusion deSPAC. Votes are tallied here. A yes vote is more of a formality these days, as only low-single-digit percentages don’t make it through this stage. The real test of SPAC shareholder confidence will be seen on redemption, at which $234M is at stake. The redemption numbers will be in the closing 8-K. Anything redeemed will no longer be available for General Fusion to use. The shareholders can vote yes on the merger and still redeem their investment plus initial interest and warrants, effectively shielding them from the downside and keeping the upside. After this vote, the company has to get through some regulatory approvals and then trades as GFUZ (shares) and GFUZW (warrants) on July 10. Call up your broker!

Chinese fusion company Xingneng Xuanguang (Xeonova) raises $73.8M Series A. Sun Xuan, the company’s founder, previously worked at TAE. It’s similar to Helion in that it’s using pulsed FRC compression but also has an external mirror field.

Honghu Fusion tests a full-scale, double-pancake HTS stellarator magnet. Nice curves.

Startorus Fusion tests a 2 × 4 m HTS magnet to 11.1 T peak field.

Astral Systems raises £23M to advance radioisotope production by fusion. Astral is similar to SHINE in building compact fusion neutron sources for non-energy fusion applications.

CFS named first international partner in UKAEA’s LIBRTI programme. LIthium BReeding Tritium Innovation programme (LIBRTI) is a high-flux D-T neutron source to test tritium breeding in different blanket materials. Ironic that a US company announces a collaboration with the UK on an experiment with a name that sounds like liberty nearly 250 years to the day after the signing of the Declaration of Independence of the US from the UK.

UKAEA and Eni launch RH3OVA, a collaboration on the fusion fuel cycle. RH3OVA is a joint venture between the D-T experience on JET at UKAEA and gas-processing codes at Eni that aims to provide tritium system engineering services.

General Fusion and Renexia announce deployment of fusion power in Italy. Renexia is a renewable energy developer, mostly in offshore wind.

General Fusion releases a slick new promo video and an updated website. It is gearing up for that deSPAC.

nT-Tao releases their quarterly update. They’re rolling their own HTS magnets and planning fusion on a barge.

Government

First sector of the BEST vacuum vessel is being prepared for installation.

A full-sized HTS central solenoid coil has been tested for the BEST tokamak.

UKAEA and PPPL sign a memorandum of understanding to collaborate. Seems like they should’ve already been working closely together.

Fusion for Energy releases its 2025 highlights report. Lots of nice pictures and progress on ITER, JT-60SA, and IFMIF-DONES in there.

UKAEA opens a Spinout Accelerator at Culham.

ITER’s first remote handling tool is complete.

People

CFS hires new CFO Lorence H. Kim. Kim was the CFO at Moderna, where he helped raise $4.4B of capital, including during its IPO in 2018. Like fusion energy today, Moderna was a pre-revenue hard-tech company (including through its IPO).

Maura Burns joins Fuse’s advisory board. Maura was the Assistant Director of the CIA’s Weapons and Counterproliferation Mission Center, which fits in well with Fuse’s defense posture.

SRNL has an opening for Division Director, Energy Sciences and Engineering, a new research division that includes fusion energy.

Journal Club

STEP Fusion: Overview of the Concept Design for the Spherical Tokamak for Energy Production and Lessons Learnt published in Journal of Fusion Energy. STEP still aims for a spherical tokamak (it’s baked in the name) by 2040 that can produce at least 100 MWe net and be tritium-fuel self-sufficient. The major updates from SPP-1 to SPP-2 are due to center-column magnet lifetime and blanket feasibility. The major radius increased from 3.6 m to 4.275 m, raising the projected center-column lifetime from 0.22 to 0.6 full-power years. This is a common trend for fusion: as design fidelity increases, important issues like neutron shielding reveal that early designs were too optimistic (more holes!). The breeder blanket changed from liquid lithium to a solid lithium oxide.

New ambient-pressure superconductor record temperature set to 151 K. Probably not practical for fusion but neat technique to achieve it.

Siemens patents a thermal system for a molten salt laser fusion device.

Paper published on Tritium supply and demand for fusion: Socioeconomic scenario modelling. I was initially interested in reading through this to see what present credible tritium projections are. However, just seeing their tritium demand estimates scaled from EU-DEMO based on fusion power (e.g., they calculate 750 g for SPARC, which in fact is 10 g), I couldn’t bring myself to read or believe the rest of the paper.

Misc

FIA releases 2026 Fusion Industry Supply Chain Report. Fusion supply-chain spending reported by surveyed companies rose 24% in 2025 to $538M and is expected to rise another 27% in 2026 to $681M. Many established suppliers are moving toward fusion and expanding capacity. But there are challenges because the market lacks clear, long-term demand.

Tungsten prices have increased nearly 10x since the start of the year. Of course, 80% of the present tungsten supply comes from China (another 5% from Russia and North Korea). Tungsten is useful for fusion as a plasma-facing component because of the excellent thermal properties, low transmutation, low tritium retention, and low sputtering. Tungsten is useful in war because it is dense, hard, and heat resistant. Tungsten used in fusion can likely be recycled; tungsten used in war is likely lost. Recent conflicts have increased tungsten consumption. Taking the early ARC design as an example, it estimated that tungsten was about 0.025% of the total cost, so a 10x increase does not strongly affect the bottom line for fusion.

Quaise Energy raises $134M to build a geothermal plant with fusion technology. It all started with a fusion gyrotron that was gathering dust in a back room at MIT.

Steering Committee formed for fusion materials database (MatDB4Fusion).

Verdict: Incremental progress, but still a sub-keV plasma. LM26 showed compression and electron heating, but nowhere near Lawson-relevant performance.

Read my How General Fusion’s reactor will work (or won’t) article for background on what General Fusion is trying to do and how it has performed to date.

What happened

On June 22, 2026, General Fusion announced that its LM26 magnetized target fusion machine had achieved compressional plasma heating.

The pre-peer-review technical paper presents results from the first 11 LM26 compression shots. The best shots showed more than a 3× rise in electron temperature, about a 10× rise in density, and about a 10× rise in poloidal magnetic field during roughly 3× radial compression.

Thomson scattering, the gold standard for electron-temperature measurements, was used, with the best results on shot LMC-11: 0.72 keV; another independent measurement (UV photodiodes) gave 0.67 keV. The ion-temperature result was lower. Neutron-inferred deuterium ion temperatures during compression were about 0.41 keV for LMC-11, with a modest rise near the end of the compression.

What is actually new

This is the first detailed public data from LM26 lithium-liner compression shots: measured liner compression, magnetic-field amplification, density rise, electron-temperature rise, neutron and gamma signals, MHD behavior, and transport modeling. It’s a well-diagnosed and analyzed set of experiments.

The data support the narrow claim that compressional heating occurred, especially for electrons. Importantly, they also show that LM26 came far short of reaching fusion-relevant ion temperatures.

The claim beneath the claim

The claim beneath the claim is that LM26 has validated the central Magnetized Target Fusion bet: form a magnetized spherical tokamak plasma at moderate temperature, mechanically compress it fast enough with a lithium liner, conserve particles and magnetic flux well enough, and let adiabatic compression carry the plasma to fusion-relevant conditions.

If compression is self-similar and particle-conserving, then density should scale like inverse volume:

where $C_R$ is the radial-equivalent compression ratio (i.e., the ratio of the initial plasma radius to the final plasma radius).

For plasma heated adiabatically (without energy loss),

For a successful adiabatic compression, the 3× radial compression General Fusion achieved in its best LM26 experiments should ideally give:

and

General Fusion demonstrated about a 10× rise in density, 3× rise in electron temperature, and practically no rise in ion temperature. This is far short of adiabatic compression.

What would have to be true

Achieving closer to adiabatic compression requires some combination of better confinement, higher starting temperature, larger compression, and faster compression. The paper points to higher magnetic field and improved shaping, including negative triangularity, as routes to lower ion heat transport and better performance.

Ion heating must become much closer to adiabatic

The paper’s ion-temperature model has the form

The first term is the desired adiabatic heating. The exponential term is the loss penalty from ion heat transport. The paper finds that an effective ion heat transport of $\hat{\chi}_i \simeq 4\ \mathrm{m^2/s}$ is enough to prevent significant ion heating, while $\hat{\chi}_i \simeq 2\ \mathrm{m^2/s}$ would have produced a stronger ion-temperature rise. In plain English: the plasma leaked heat too fast.

The compression ratio must be judged against the starting temperature

For ideal adiabatic heating,

and

Using the reported starting electron temperatures, LM26 had sufficient compression to achieve 1 keV if it were adiabatic:

or about 10× volume compression.

Reaching 10 keV is much harder

From a 0.21 keV starting plasma, adiabatic heating to 10 keV requires

or about 330× volume compression.

The confinement time must improve relative to the compression time

General Fusion has previously pointed to energy-confinement times exceeding 10 ms as the target-plasma lifetime needed for LM26’s compression program.

In LMC-11, the inferred ion energy confinement time was about 5.3 ms, while the compression time was about 3.6 ms, giving

That is enough to see some late ion reheating, but not enough for clean adiabatic ion heating to keV-scale ion temperatures.

What is still missing

The missing result is convincing, fusion-relevant plasma performance. General Fusion needs to show ion heating or it does not have a chance at a fusion energy machine.

Why it matters

After two and a half decades of work, General Fusion has not yet demonstrated compressive heating of ion temperatures above 1 keV, let alone the 10 keV needed for fusion energy. This suggests fundamental flaws in what they are doing and/or how they are doing it. The work presented in this paper is a small step forward from where General Fusion was before LM26. General Fusion is going to need to take a relatively huge leap in plasma performance with the funding from this deSPAC to survive long in the public markets.

Bottom line

LM26 has demonstrated compressional heating, but the published result is still a sub-keV electron-temperature result with small ion heating. It failed to make the plasma behave adiabatically enough for fusion, especially for the ions.

General Fusion publishes analysis of plasma compression in LM26. The United States Department of Energy released its finalized Fusion Science and Technology Roadmap. Xcimer preconceptual design approved by DoE.

Companies

General Fusion claims compressional heating of plasmas in LM26. It is great General Fusion is doing this level of analysis on their experiments. If you dig into the paper submitted for peer review, you see that only the electrons were substaintially heated. The ions, which matter for fusion, remained largely unchanged (typically cooling early in the compression only to recover to near the starting temperature at the end). This is consistent with my previous analysis of the progress, or lack thereof, the company has made toward its fusion target over the years. It can densify the plasma through compression but not get the ion heating needed to reach fusion conditions. It also reinforces my hypothesis that underperformance of the ill-named Lawson Machine 26 (they predicted the machine would exceed the Lawson parameter in 2026) caused the present inside investors to not want to continue to invest in the company, forcing it to de-SPAC to get money to continue to operate.

OpenStar publishes its concept paper on Deuterium–tritium levitated dipole fusion power plants. It has a good analysis of the “snowball-in-hell” problem of the levitated dipole magnet. There is still a lot of work to be done experimentally testing the assumptions on the physics basis of levitated-dipole fusion plasmas to make this concept viable.

Astral has raised £23 million to accelerate the development of fusion-produced medical isotopes.

Helion announced that it had received two Washington Department of Health licenses for Orion. The licenses are for Radioactive Materials and Radioactive Air Emissions.

General Fusion was named one of Time’s Top GreenTech Companies of 2026. Although it does not appear to be pay-to-play to get on the list, it is pay-to-use the award announcement. So maybe Time ranks higher the companies most likely to want the publicity and willing to pay, as it surely could not be for fusion performance, strong finances, or actual environmental impact just yet.

CFS announces investment by Abu Dhabi-based Plynth Energy. Plynth is a fusion-focused fund. This investment announcement came nearly a year after CFS announced its Series B2. Axios claims it was nine figures.

Focused Energy plans to spin off SourceLight. This is to make a clean separation between Focused Energy’s fusion-energy focus and the other technologies that have been in development at the company, including SourceLight’s laser-driven radiation sources for materials inspection.

Tokamak Energy and Furukawa Electric are collaborating on UK high-temperature-superconducting (HTS) tape capability. The effort has letters of support from the UK government and UK Industrial Fusion Solutions. Furukawa, through its U.S. subsidiary SuperPower, has been working on HTS. While it has developed good-performance HTS, it has lagged in scaling up commercial production. Given the rise of industry leaders Shanghai Superconductor in China and the kind-of-Russian-associated Faraday Factory in Japan, this collaboration between Tokamak Energy and Furukawa is probably meant to shore up a domestic UK HTS supply.

Xcimer’s preconceptual design has been approved by the U.S. Department of Energy as part of the Milestone Program.

Avalanche Energy claims that its Jyn device has reached apparent ion temperatures above 1 keV. Avalanche’s own memo indicates that it understands the shortcomings of its measurements. What it has done is measure a broad, rotating Hα charge-exchange spectrum and fit it with “plausible” models. But the width is a mixture of charge-exchange weighting, line-of-sight integration, bulk rotation, cycloidal orbit structure, assumed emissivity, and time averaging over a decaying discharge. Until the analysis resolves all of those issues, the “1.4-1.6 keV apparent ion temperature” claim should be considered “keV-scale fitted velocity variance,” not a measured plasma ion temperature.

SHINE partners with newcleo on nuclear waste reprocessing. newcleo is a European fission company founded in 2021 that is developing lead-cooled fast reactors. As such, it does not yet have spent fuel that needs reprocessing, but it is interested in working with SHINE on doing so, as well as using material that SHINE has reprocessed from others using neutrons from its DT fusion accelerators. In related news, SHINE was one of five companies selected by the DOE to talk about transforming Cold War-era plutonium into nuclear reactor fuel.

First Light Fusion is now using its fusion equipment for orbital impact studies. Its VIPER velocity amplifier can achieve projectile velocities above 12 km/s.

General Fusion filed its SEC Form 424B3. It describes all of the financial details of what the shareholders are voting on for the merger. It also amended the SEC Form 8-K, giving General Fusion SAFE holders the right to vote on the transaction.

TAE and TMTG file SEC Form 8-K. They restate their goal of completing the merger by the end of the year. Interestingly, the spin-off of Truth Social is no longer being pursued.

American Fusion files SEC Schedule 13D. It indicates that the CEO, Richard C. Hawkins, owns 8.59% of the common shares in the company.

Pinnacle Consulting Services files Schedule 13G for American Fusion. Pinnacle Consulting Services, controlled by Robert L. Hymers III, disclosed beneficial ownership of 145.1 million AMFN shares, or 8.99%.

People

Inertia, the fusion company founded by the owner of The Onion, forms a Science and Technology Advisory Board. Marv Adams, former Deputy Administrator for Defense Programs at the National Nuclear Security Administration, is the chair. The board is rounded out with ICF physicist Brian MacGowan, retired leader of the stealth bomber division at Northrop Grumman Doug Young, former NIF Director Jeff Atherton, ICF diagnostic expert Joe Kilkenny, fusion nuclear-materials professor and card counter Brian Wirth, high-energy-laser and optics expert Jon Zuegel, ICF target expert Larry Suter, laser-plasma physicist Bill Kruer, and the current Guinness World Records holder of the record for highest fusion areal density by grip strength alone, Chuck Norris.

Avalanche Energy appointed Mustally Hussain as chief financial officer. He will oversee finance, capital planning, corporate development, and investor relations as Avalanche scales commercialization of its compact fusion systems. Mustally has a background in banking, moved into finance and strategy at National Grid, led finance in auto lending at Hyundai America and Herc Rentals, and then joined Lucid Motors a few months after it de-SPACed, followed by a short stint at a humanoid-robotics company. The energy-plus-auto background makes sense based on Avalanche’s plans for impossible small-scale mobile fusion platforms. And the post-de-SPAC experience may be useful for the public rumors of Avalanche’s IPO aspirations.

TAE Technologies hired Jeff Woodbury as president and chief operating officer. TAE said the hire expands leadership as it advances commercial development, including construction of its first utility-scale fusion power plant. Jeff had 35 years of service at Exxon Mobil. TAE was able to entice him back to work after eight years of retirement.

Bernard Looney has joined the Type One Energy board of directors. Bernard worked his way up from drilling engineer to CEO at BP.

American Fusion adds Robert V. Duncan as an independent scientific and strategic advisor. This is telling, as Professor Duncan’s first research interest is listed as “Low-Temperature, Condensed Matter Physics,” indicating that American Fusion is not likely to achieve fusion with its present device or plans.

ORNL is hiring directors of the Fusion Engineering and Applied Materials, Fuel Cycle Science and Technology, and Nuclear Science & Engineering divisions.

Government

The United States Department of Energy released its finalized Fusion Science and Technology Roadmap. I started to write my thoughts on this, but it was turning into a rant that I will probably make into its own post. Overall, it is a fairly nice plan. But the second page tells the whole story: “Disclaimer → The activities outlined in the Fusion Science and Technology Roadmap are focused on prioritizing strategic directions for the United States (U.S.) Department of Energy to further collaborate with the U.S. Fusion Industry. This Roadmap does not commit the Department of Energy to specific funding levels. Future funding is subject to Congressional appropriations.” Without funding, it is just another nice plan.

Canada releases a nuclear energy strategy. Naturally, it focuses on being a leader in the fusion fuel cycle due to the extensive tritium production from its CANDU reactors.

Japan plans ¥3.1 trillion ($19 billion) in public-private fusion investment by 2040. Japanese media reported that a draft Takaichi growth strategy would set a ¥3.1 trillion public-private fusion investment target by FY2040, with a 2030s power-generation demonstration goal and a 30% global-market-share ambition. The number is still a draft public-private investment target, not yet an appropriated government budget.

ORNL is collaborating with Singapore on fusion.

China expands IPO support to “future industries,” including fusion. China has a “fifth listing standard” for strategic technology firms that may not yet be profitable to go public. They have now explicitly expanded it to include the future industries in their five-year plans.

The NRC finalized a fee rule aimed at increasing cost certainty and lowering barriers for nuclear energy projects. The rule sets fixed caps on many licensing and service fees and lowers some costs for applicants.

UK Atomic Energy Authority committed £20 million more to the UK Innovation & Science Seed Fund’s fusion portfolio.

United Kingdom and Japan (UKAEA and QST) signed a Memorandum of Cooperation covering breeding blankets, plasma science, materials, plant systems engineering, robotics, tritium technologies, regulation, standards, skills, affordability, and commercial viability. This creates a mechanism for cooperation between two of the leading government programs in fusion engineering technology.

General Atomics is collaborating with the U.S. DOE and Idaho National Laboratory on a Fusion Blanket Component Test Facility. This is the first step in which they will work to develop design concepts for a facility dedicated to testing full-scale fusion power-plant blankets, potentially using the same facility where GA assembled the ITER central solenoid magnets.

Project WISER, a “wind-tunnel” stellarator project, is launched. WISER, while not a small machine (R = 5 m, B = 1.9 T), is smaller and lower-field than what is projected for commercial-scale stellarators. Centered on the Spanish fusion ecosystem, €500 million is earmarked to support the project through construction.

Other

Financial Times writes a story on the fusion supply chain. There is not much new in there for those in the industry. There will be lots of demand if and when fusion ramps up, but there is still uncertainty about timing and technology specifics.

The folks at 1c Fusion Energy have released their technoeconomic model, including the code. Go play with it, find the holes, and help them improve it.

Helion, Focused Energy, and Thea Energy raise big rounds. Pacific Fusion passes a pulsed-power system milestone. CFS releases ARC physics basis.

Fusion sublimes addendum

It seems some people did not get an important point in my The sublimes of specious fusion concepts article last week. The sublimes are (1) not limited to specious fusion concepts and (2) not all negative in and of themselves. In fact, many fusion companies that I consider credible have some number of the sublimes and that is a fine thing. The major point missed was that one should not be distracted by the sublimes in considering fusion companies/concepts, and instead should look at the whole, detailed picture of what is being proposed.

Also, I was disappointed no one got the image reference for that article. I suppose none of my readers spent their summers as I did, listening to their local alt rock radio station while assembling thousands of dinning room chairs. Sublime is a ska punk band with a drugged-up sun logo. I had AI clean it up, as this is a professional newsletter.

Companies

Helion raises $465M Series G. This brings Helion’s total raised to $1.5B, half of what CFS has raised. The valuation puts Helion in the same category as Rippling, Magic Leap, Discord, and Mistral AI and ~50% higher than comparable companies in the energy space like CFS, Fervo, and Oklo, the latter two of which are public. Feels a bit high to me, but what do I know.

Focused Energy raises $240M Series A. It was supported by Rothschild & Co. They’re emphasizing that it is “the largest fully secured Series A financing in the global fusion industry” and, confusingly, sometimes leaving out the “fully secured” part. That bit is in there because both the Inertia and Pacific Fusion Series A rounds were much larger, but they were tranched based on milestones.

Thea Energy raises $100M Series B. Funding will help expand magnet manufacturing, as well as siting and construction of their Eos stellarator.

Avalanche Energy plans to IPO. Given the massive private rounds raised by the other three fusion companies above, this couldn’t be because there isn’t any private capital left for fusion. I suspect the story is similar to TAE and General Fusion: a fusion company that’s overpromised on its ability to deliver fusion power to the private markets driving them into the hot public markets.

Pacific Fusion reaches major pulsed-power milestone that unlocks the next tranche of its Series A. This is an important demonstration of the basic building blocks of its pulsed-power innovation. It produced 440 GW in 80 ns. The prototype pulser module is about a third the size of the full module, of which 156 would be needed for its full-size demonstration machine.

CFS publishes ARC Fusion Power Plant Physics Basis. Naturally, the machine has evolved from the original ARC: the major parameters that having changed include: increasing the fusion power from 525 MW to 1130 MW, gain from 13.6 to 51, major radius from 3.3 m to 4.62 m, and toroidal magnetic field from 9.2 T to 11.4 T. The increases stem from the fact that the mission of the original ARC was to see what a class of graduate students could design to deliver an ITER-level of fusion power (500 MW) with HTS magnets, and the mission of the present ARC is to be a first-of-a-kind fusion power plant with stringent economic and performance requirements, as well as a much higher level of design fidelity.

General Fusion and General Atomics to collaborate on advanced diagnostic systems. The title says it all: the press release is devoid of content on what they will work on together.

Rich Huang of RichRich Capital LLC buys 5.14% of Spring Valley Acquisition Corp. III, the General Fusion SPAC-to-be. It is not clear who this Rich is from public information, but could he be “the world’s richest boy” all grown up?

General Fusion amends its SEC Form F-4 with some financial updates.

Realta announced DOE’s approval of its preconceptual design magnetic mirror Hammir. The design point has 20+ T mirror coils (enabled by HTS), 500 MW fusion power, and Q_sci > 5.

Government

JP Allain steps down from Department of Energy leadership. I find it strange that it is characterized as “after completing his tenure,” as he had been appointed inaugural director of the new Office of Fusion only two months earlier. He was appointed DOE Fusion Energy Sciences Associate Director in July 2023. JP has done a great job leading fusion at the DOE, and I appreciate his service!

Tennessee will become the first state in the nation with its own regulatory framework for nuclear fusion machines. Tennessee is a Nuclear Regulatory Commission Agreement State, meaning that it regulates nuclear machines at the state level, rather than the federal level. This is common among states with extensive nuclear facilities. Now in Tennessee, “A fusion machine is defined as any machine capable of: Transforming atomic nuclei, through fusion processes, into different elements, isotopes, or other particles; and directly capturing and using the resultant products, including particles, heat, or other electromagnetic radiation.”

UK publishes Draft National Policy Statement for fusion energy generation. This puts fusion in its own distinct development track, rather than shoehorning it into another category, such as fission.

Fusion for Energy announced that Europe’s first four ITER divertor cassette bodies were complete and ready to ship. That’s a big hunk of metal!

European Commission announced the inaugural meeting of the new ENSREG Task Force on Fusion Energy.

People

General Fusion names Joanna Cameron as General Counsel and Corporate Secretary. Joanna has mostly worked within law firms during her 30-year career, but did spend nearly three years at NexGen Energy, a uranium exploration and development company. She “focused primarily on capital markets transactions and business law matters, including equity and debt financings, initial public offerings, and public and private mergers and acquisitions,” so she is an appropriate hire at this juncture for General Fusion as they work to de-SPAC.

General Fusion alumni Michael Delage and Alex Mossman have launched Basilisk Industries to “change the way the world cuts rock”. Michael was a VP and CTO at General Fusion for 11 years, and Alex was the Senior Director of Applied Physics and Plasma Experiment Operator at General Fusion for 12 years.

Other

U.S. Department of Energy-backed Antares advanced reactor at Idaho National Laboratory reached first criticality!

Hinetics completed the build of CRUISE, its 5 MW HTS electric motor prototype. The non-fusion HTS magnet applications are starting to open up.

The sublimes of specious fusion concepts

sublime /səˈblaɪm/ adjective

tending to inspire awe usually because of elevated quality

specious /ˈspiːʃəs/ adjective

having a false look of truth or genuineness; having deceptive attraction or allure

Bent Flyvbjerg, an academic expert on megaprojects, wrote about the four sublimes that explain the common story arcs of megaprojects and “seduce decision makers to underestimate hidden risks, and to overestimate benefits”:

• The “technological sublime” describes the excitement engineers and technologists get in pushing the envelope for what is possible in “longest-tallest-fastest” types of projects.
• The “political sublime” describes the tendency politicians have for constructing monumental infrastructure to benefit themselves and their causes.
• The “economic sublime” describes the delight business people and trade unions get from making lots of money and jobs from megaprojects.
• Finally, the “aesthetic sublime” is the pleasure designers and people who appreciate good design derive from building and using iconic and beautiful megaprojects.

In reviewing fusion companies and reactor concepts, I have found a similar pattern. There is a set of common sublimes that arise again and again: ideas that seem so attractive they can make other hard problems all but disappear.

The compactness sublime

A compact fusion device sounds like the clean answer to the giant fusion projects typified by ITER, NIF, and others. Make it smaller and it will cost less. Make it simple and the timelines shrink. Make it look like a product instead of infrastructure.

No one is building a fusion device bigger than it needs to be for their purposes. The problem is that fusion has fundamental limits to scale. Neutrons need a meter or two of shielding, even for aneutronic fuels. Thermal plasmas need distance for insulation. Materials have limits to power density, stresses, etc.

The non-thermal sublime

There are some who believe the laws of nature are like the laws of man, that they can pick which ones to follow and choose not to follow the laws of thermodynamics. They want to play Maxwell’s demon. They think they can reduce energy losses from ions to electrons. They think they can keep ions non-thermal with minimal recirculating power.

But physics sets the house rules and the house always wins. The Rider Thesis is the boogeyman to Maxwell’s demon in fusion.

The aneutronic sublime

Aneutronic fusion is the clean-fusion dream: no tritium, fewer neutrons, less activation, simpler licensing, and direct energy conversion. Some even go so far as to falsely claim it has zero radiation production.

It is easy to see the appeal. Tritium breeding is hard. Neutron damage is hard. Remote maintenance is hard.

But an aneutronic fuel does not make the system easier. Aneutronic fuels come with much harder plasma requirements: higher temperatures, better confinement, and stringent impurity control. These fuels always produce high energy radiation via secondary reactions which make truly aneutronic fusion reactors a fiction.

The direct-conversion sublime

Fusion is an advanced technology; why shouldn’t it have an advanced energy conversion system, one that is more efficient than just boiling water?

Much of the energy is released in charged particles, which can then be converted by electric and/or magnetic fields directly into electricity.

The problem is that it is hardly as simple as the cartoon picture. Real particles scatter energy. Real plasmas radiate photons. Real collectors need to survive harsh environments. While direct conversion could increase the conversion efficiency of fusion energy to electricity, the realities of physics and engineering constraints greatly reduce the benefits from what is purported.

The simulation sublime

The plots are beautiful. The fields look clean. The particles stay confined. Net energy is one experiment away.

All too often, simplistic models are used to demonstrate that a clever new fusion concept is the next ideal confinement device. In the compromises to build a solvable model, important physics is thrown out. The device never performs nearly as well as the initial models indicated.

All models are wrong. Some are useful. The least useful model is the one that works because it excludes the physics that would kill the concept.

The founder sublime

Investors are vulnerable to a good founder story. They love the outsider genius, the person who sees what the establishment missed.

Fusion is one of the most complex endeavors humanity has undertaken. There is not a single person on Earth who has a complete picture of how fusion energy will turn out. The romantic idea of an outsider displacing the field is unlikely; they all come in missing some of the important fundamentals, many of which are summarized in other sublimes here.

The useful question is whether the concept survives review after the founder leaves the room.

The secrecy sublime

A company says the key breakthrough is proprietary. They cannot share the information with reviewers, even under NDA. They are scared that it will be stolen.

Fusion companies should protect real intellectual property. But if the central physics claim cannot be checked under NDA by competent reviewers, secrecy is no longer protection; it is insulation from reality.

The incumbent-failure sublime

Tokamaks are too big. ITER is too slow. NIF is not a power plant. Stellarators are too complex. Magnets are expensive. Lasers are inefficient.

A critique of the incumbent is not validation of the alternative. The alternative still has to close on its own merits.

The single-bottleneck sublime

The magnets are the issue. The laser is the issue. The fuel is the issue. The materials are the issue.

Fusion is a coupled system. Higher field helps confinement but increases stress and shielding challenges. Smaller size reduces capital cost but increases wall loading.

There are no isolated issues or miracles in fusion.

The cheap-option sublime

Some concepts survive because the next experiment is cheap enough to fund.

For an investor, that may be reasonable. A $10M experiment can be a cheap way to buy optionality on a huge outcome. It is also a form of Pascal’s wager, a bet on a low-probability, high-impact outcome.

While “expensive to test” does mean “less likely to become a power plant,” the converse of “cheap to test” does not mean “more likely to become a power plant.” It means cheap to test.

The conclusion

Every specious fusion concept has its own selection of sublimes. The machine is small. The ions are non-thermal. The fuel is clean. The conversion is direct. The simulation works. The founder saw what others missed. The details are secret. The next experiment is cheap.

All too often, these sublimes seduce investors into funding concepts that will fail. The sublimes cannot replace the integrated and complicated physics and engineering needed to make fusion work. Fusion is hard. That is why the field needs discipline.

Ultimately, fusion energy is its own sublime. Once realized, it will be one of the great achievements of humanity. Fusion should inspire awe; but don’t let its aura blind you into making a financial decision based on one of these sublimes without understanding the whole picture.

Helion moves the goalposts on electricity milestone. U.S. DOE approves Zap’s preconceptual design milestone. Thea operates its first full-size and current HTS magnet.

Companies

Helion updates its website, including a new FAQ. Most interestingly, they moved the goalposts on their “net electricity” framing: “In 2021, we used ‘net electricity’ in a headline when discussing Polaris. In hindsight we realized that the term was not well defined within the fusion industry and created more confusion than clarity. Today we prefer to focus on a clearer goal that no one else in the industry has touched: demonstrating electricity from fusion.” In principle this would allow them to create a tiny amount of fusion power and have their magnet system capture a tiny fraction of that and call it success that they “demonstrated electricity from fusion.” This is a baloney metric. By that definition I “demonstrated electricity from fusion” in my PhD research where I fielded thermocouples on Alcator C-Mod, as did just about anyone else who fielded a diagnostic measurement system on a fusion experiment.

U.S. Department of Energy approves Zap’s fusion pilot plant preconceptual design. This is one of the standard milestones as part of this program. Thea had theirs approved earlier this year and more will come. To me, this only says that they have a relatively credible design if they are able to push the z-pinch plasma performance up to where it needs to be.

Thea Energy successfully operated its first full-size, full-current, and full-field planar shaping coil. The company said the magnet reached the fields (6 T) and currents required for its upcoming Eos stellarator.

Marathon releases a new paper on The Value and Cost of Fusion Neutrons. I have yet to read over it, so I cannot comment on the details, but I do agree with the overall framing.

General Fusion appointed Thomas Boehlert to its board. Thomas has a long career as a CFO in various energy and, more recently, mineral companies. He will be the chair of the nominating and governance committees at General Fusion. More shoring up of their board and leadership as General Fusion works to go public.

General Fusion is exploring listing on the Toronto Stock Exchange in addition to the Nasdaq.

General Fusion files Amendment No. 3 to the SEC Form F-4, a 10-Q, and an 8-K. They have some new info on legal and capital structure but nothing on how they are going to adiabatically compress their plasmas.

American Fusion filed its Form 10-Q for the quarter ended March 31, 2026. While nothing of consequence as progress toward fusion energy, reading through these is a fascinating look into the world of penny stocks and companies that may not be what they seem to be. It disclosed less than $100k of cash at quarter-end and said substantial doubt exists about the company’s ability to continue as a going concern while management pursues a 2026 capital raise. The chairman, Brent Nelson, has a 60% controlling share. The consulting and independent director advisory agreements provide total compensation of $1.68 million in common stock. The share count is determined by market price: if the stock is lower one year later, the company must issue more shares so the recipients still receive the agreed value; if the stock rises, there is no clawback.

Allegations of Sam Altman self-dealing with Helion. Was anyone surprised?

First Light Fusion has some new filings related to its recent round. Adrian Gareth Jones and Theodore Frederic Florent Marie Mollinger were appointed as directors. The documents imply a post-money valuation of about £65M.

TAE and UKAEA announced that TAE Beam UK had been formally established and was fully funded. They’re working together, using their experience building and operating neutral beams to make them smaller and cheaper for fusion energy. To me, this is one of the most worthwhile things TAE is doing to advance fusion energy. Neutral beams have been the workhorse of heating for many fusion experiments but, like much of fusion, have been too big and expensive to be practical for economic fusion applications. Case in point: F4E and ALSYMEX celebrate the completion of the MITICA beam source.

Commonwealth Fusion Systems announced that the second half of SPARC’s vacuum vessel had arrived and that fit-out work had begun. Brandon gives a nice video tour at the end of the article.

A*STAR and Commonwealth Fusion Systems sign collaborative research agreement. The five-year agreement is to help develop fusion technologies for CFS’ ARC and Singapore’s building of a domestic industry to supply the fusion industry. Reminder that Temasek, a Singapore sovereign wealth fund, is a large investor in CFS.

Proxima Fusion created an Industrial Development Board. The new body brings in senior industrial and energy figures including former EDF chief Luc Rémont, former Bosch CTO Michael Bolle, former VP of Breakthrough Energy Ventures, Europe Ann Mettler, and E.ON chair Erich Clementi. Proxima continues to demonstrate its skill in coalition building.

Avalanche announced its new machine, Lando. They proudly announce it is licensed for up to 10¹⁰ neutrons per second, which puts it on the same scale as neutron generators that are already available for commercial purchase. This is a very low rate as far as neutron generation goes; a 100 MW D+T fusion plant would be ~3.6 × 10¹⁹ neutrons per second. It will need 10,000 of these overlapping in space to meet the requirements of DOE’s volumetric neutron source.

Thea Energy adds Paul Wilkins as vice president of policy. Paul was Chief of Staff to a U.S. senator as well as VP of policy at Bloom Energy and Electric Hydrogen.

Kyoto Fusioneering awarded contract by QST to develop tritium extraction and accountancy system prototypes for ITER test blanket system. These are critical technologies that need to be developed to get the tritium out of where it’s bred.

Astral Systems, together with Promation and McMaster University, was awarded CMIE funding through the Scale-67 project. McMaster said the program will receive $500k to develop a fusion reaction-based process for Copper-67, which is used for diagnostic imaging and cancer treatment.

Pacific Fusion moves co-founder Leland Ellison into the vice president of technology role. Leland is a career computational physicist.

Type One Energy brings former ORNL fusion section head Larry Baylor in as engineering fellow. Larry has expertise across a variety of fusion energy systems but he’s most well known in the community as the go-to expert for systems to fuel the plasma with frozen pellets. And apparently, from his LinkedIn profile, he’s good at threading golf balls through a flock of Canada geese.

Government

The DOE is interested in applications for preliminary design studies for a volumetric neutron source. Such a facility would be used to study fusion nuclear materials, which is, in my opinion, the critical technology bottleneck for economic fusion energy.

FIA submits response to NRC’s proposed rulemaking and draft guidance for U.S. fusion regulations. The FIA requests a couple of changes (e.g., clarify what “lost” tritium means) and comments on areas directly requested by the NRC. Axios reports on the NRC fusion review as well.

The recent news of the DOE announcing more than $94 million of Gen III+ SMR pathway awards to eight companies, DOE’s Office of Critical Minerals and Energy Innovation announcing $45.7 million for 19 critical-material projects, and the Department of Commerce announcing letters of intent with 9 companies for $2B to accelerate U.S. leadership in quantum computing, as well as the $800M in federal cost-shared funding for fission announced at the end of last year, has me wondering what is in store for fusion…

The National Institutes for Quantum Science and Technology has now contracted with six Japanese suppliers to execute an additional 20-unit ITER gyrotron package.

China and Russia sign nuclear-related MoUs during Xi-Putin talks. Although details are light, it almost certainly includes collaboration on China’s new tokamak, BEST.

ITER Magnet Cold Test Facility operational. They used to say they didn’t have the time to test the magnets after they are built but before they are assembled with the rest of the machine. Now that the schedule has slipped so much, they have the luxury of time to do so to “strengthen ITER’s risk mitigation and readiness.”

Media

MarketWatch quotes me in an article on TAE-TMTG merger. The author of the article was fishing for evidence that the administration is supporting fusion in a way that benefits TAE and therefore TMTG financially. For the record, I’ve seen no evidence of this and will call it out if I do. The DOE and NRC support of fusion has been fair across the industry as far as I have seen and has been the best support the industry has ever seen.

Zap is doing fusion, fission, and fusion-fission. Type One Energy partners with Tokamak Energy and AECOM to build a UK stellarator power plant. Helion builds a new, smaller machine to learn stuff.

Companies

Zap Energy expands from fusion to fission and fission-fusion. It had to happen eventually. Increasing the capability of a fusion device from a good producer of neutrons to an economic source of energy is really hard and may be impossible for many of the fusion reactor concepts. Ideas have been thrown around for decades on fusion-fission hybrids where a fusion neutron source that’s not good enough on its own to make economic fusion energy is combined with a fission fuel that’s not good enough on its own to make a self-sustaining fission reactor. Hybrids are a highly debated space in the nuclear energy field: some consider them the best of both worlds, others consider them the worst of both worlds. Zap is framing this pivot as leveraging their liquid metal technology developed for fusion applications to be used on similar-scale (~50 MW) fission systems, fusion systems, and fusion-fission hybrid systems. Which begs the question: What do they have that is fundamentally better than the many fission companies out there at that scale? I admit to not knowing enough about the fission companies to make a clear statement. To me this pivot signals that Zap was not getting the performance out of their shear-flow-stabilized z-pinch as they had predicted and they needed to pivot. It’s an attempt to create a business model that justifies their previous valuations in the context of the technology they have developed and the demands of the outside world.

Zap Energy has appointed Zabrina Johal as CEO. Aligned with Zap’s pivot to include fission in their technology pathway, Zabrina spent nearly two decades at General Atomics in roles that “Led strategy, program development, and external engagement across General Atomics’ nuclear and defense portfolio.” This may be a great hire for Zap, depending on how well Zabrina can adjust to the organizational culture differences, which is no small feat.

Type One Energy, Tokamak Energy, and AECOM form the UK Infinity Fusion Consortium. This is a distinct effort from the UK’s STEP fusion power plant prototype. Type One Energy is an American stellarator company. Tokamak Energy is a British company that was a spherical tokamak company but has recently pivoted to being a superconducting magnet manufacturer and supplier. AECOM is an American multinational infrastructure consulting firm that knows how to build big stuff. I find it interesting that Tokamak Energy is the magnet supplier for Type One Energy in this consortium: just over a year ago it was announced that “…Type One Energy [gets] an exclusive license to use CFS’ groundbreaking high-temperature superconducting (HTS) cable technology in the development of its own proprietary stellarator fusion magnets. The agreement also gives Type One Energy the benefit of the experience CFS has in manufacturing the world’s most powerful HTS fusion magnets.”

Startorus Fusion closes a 500M RMB (74M USD) A+ round, becoming a unicorn (total valuation >1B USD). Startorus is working on a negative triangularity spherical tokamak. Based on this, it seems there is a disconnect between US and China private fusion company valuations, with China’s being on the high side.

Helion Energy is building “Tiny Merge,” a new plasma physics experiment. The smaller device will enable more physics learning alongside their larger experiment Polaris. Maybe they aren’t quite so ready for their 2028 power supply to Microsoft…

General Fusion announced a DOE-funded peer-reviewed fuel-cycle study with Savannah River National Laboratory. The short story is: having such a full and thick blanket allows for a high tritium breeding ratio and quicker ramp-up of fusion, assuming everything else works. They also found that a pure-lithium blanket trapped more tritium than a lead-lithium blanket (but we already knew that).

nT-Tao has signed a memorandum of understanding with Israel’s national water company to explore fusion energy for critical water infrastructure. Don’t know why a study would be needed: the answer is “yes” if they can build a fusion energy source. But I don’t think nT-Tao can.

Xcimer Energy hired Rachel Konrad as SVP of Strategic Communications. Rachel is an interesting choice given her background in mostly consumer-facing communications roles at places like Tesla (electric cars) and Impossible Foods (plant-based “meat”).

Xcimer Energy hired Douglas Kunzman as VP of Defense. Douglas has a long career in the US Navy and will help Xcimer sell to customers who want to shoot their lasers at non-fusion targets.

SHINE announces changes to leadership team and board, including new Chief Financial, Strategy, and Legal Officers.

General Fusion shares their Investor Day presentation as an SEC Form 8-K. Nothing new in there but a good resource to see what and how they are selling.

TMTG files SEC Forms 10-K and 10-K/A. No new information on the TAE merger. Q1 revenue was only $871,000, while the company reported a $405.9 million net loss. Also given is executive compensation, if you are into that type of voyeurism. The Wall Street Journal reports on the recent leadership shake-up within TMTG.

American Fusion, the first public fusion company, files SEC Form 8-K. It sold two phase noise analyzers to the Canadian Department of National Defence for $58,000. A single institutional accredited investor funded $793,000 through a stock warrant.

Government

South Korea launches its fifth basic fusion-energy plan, with a 56-member committee and a stated goal of demonstrating fusion power generation in the 2030s.

UKAEA releases its Fusion Futures Industry Capability Report and Fusion Futures End of Year Report 2024–2026. These are reports on progress in the UK to develop a domestic fusion industry.

UK Fusion Energy and Korea partner to scale superconducting cable manufacturing. I wonder how this works with Tokamak Energy’s STEP partnership.