PORRITT INC. — ENERGY INTELLIGENCE BRIEF
Advanced Fission Analysis Series — April 2026
The Reactor That Can’t Melt Down: And That’s Not Marketing — It’s Physics
Every nuclear accident in history — Three Mile Island, Chernobyl, Fukushima — shares the same failure mode: loss of coolant leading to fuel overheating, cladding failure, and fission product release. The entire regulatory architecture of the NRC, the design basis of every light-water reactor, and the public’s fear of nuclear power all trace back to this single thermodynamic vulnerability.
X-energy’s Xe-100 eliminates it. Not with better safety systems. Not with redundant cooling loops. With physics.
The Xe-100 is a high-temperature gas-cooled reactor (HTGR) using TRISO fuel in a pebble-bed configuration, cooled by helium gas. The fuel itself is the safety system — and understanding why requires looking at what TRISO actually is.
TRISO: The Fuel That Is Its Own Containment
TRISO (Tri-structural Isotropic) fuel particles are uranium kernels approximately 0.5 mm in diameter, encased in four concentric layers: a porous carbon buffer that absorbs fission gas swelling, an inner pyrolytic carbon layer, a silicon carbide ceramic shell, and an outer pyrolytic carbon layer. Each particle is a self-contained micro-reactor with its own pressure vessel and fission product barrier.
The silicon carbide layer is the critical engineering achievement. SiC maintains structural integrity to approximately 1,600°C. At atmospheric pressure. Without active cooling. The melting point of SiC is 2,730°C — well above any credible loss-of-coolant temperature scenario in a pebble-bed configuration.
In the Xe-100, these TRISO particles are embedded in graphite spheres approximately 60 mm in diameter — the “pebbles.” Each pebble contains roughly 18,000 TRISO particles. The pebbles are buoyant in helium and circulate through the reactor core continuously, enabling online refueling without shutdown.
The physics case for “cannot melt down” is straightforward: in a total loss of coolant and loss of all active safety systems, the maximum fuel temperature in a Xe-100 core is bounded by the geometry and thermal conductivity of the graphite pebble bed. The reactor passively sheds decay heat through conduction and radiation to the reactor vessel and surrounding structures. At no point does fuel temperature approach the 1,600°C SiC integrity limit.
This is not a probabilistic safety argument (“the odds of failure are 10⁻⁶”). It is a deterministic one. The laws of thermodynamics prevent fuel failure. Full stop.
76 MWe Per Module — And the Real Value Is the Heat
Each Xe-100 module produces approximately 200 MWt (megawatts thermal) and 76 MWe (megawatts electric) through a conventional Brayton cycle using the helium coolant directly. Modules are designed to be factory-fabricated and site-assembled — a modular construction model that reduces on-site construction time and enables multi-unit scaling.
But here is where the Xe-100 separates from every other SMR on the market: the helium outlet temperature is 750°C.
Light-water reactor steam loops operate at ~300°C. Sodium-cooled fast reactors like TerraPower’s Natrium reach ~500°C. The Xe-100’s 750°C helium outlet opens a thermodynamic regime that no other commercial reactor design can access: industrial process heat at steam-equivalent temperatures of 565°C — directly usable in petrochemical processing, hydrogen production via high-temperature steam electrolysis, ammonia synthesis, and desalination.
This is the dual-use case that makes the Xe-100 relevant beyond electricity. A refinery needs two things: electricity for rotating equipment and controls, and process heat for distillation, cracking, and reforming. The Xe-100 supplies both from a single thermal source. No natural gas boilers. No Scope 1 emissions. No carbon tax exposure.
For data centers, the electricity is the product. For heavy industry, the heat is the product. The Xe-100 serves both markets from the same reactor design. No other SMR can make that claim at 750°C.
The Dow Partnership: Industrial Nuclear at Scale
In 2023, Dow Chemical and X-energy announced plans to deploy Xe-100 reactors at Dow’s Seadrift, Texas operations — one of the largest integrated chemical manufacturing sites in the United States. X-energy submitted the construction permit application to the NRC in March 2025, with the NRC issuing an 18-month review schedule. Construction is planned to begin in 2026.
This is not a demonstration project. Dow’s Seadrift facility is an operating petrochemical complex that consumes massive amounts of electricity and process steam. The Xe-100 deployment will provide both — replacing natural gas–fired cogeneration with nuclear cogeneration. DOE’s Advanced Reactor Demonstration Program (ARDP) is backing the project with a cooperative agreement worth approximately $2.5 billion in combined public-private investment.
The Amazon factor adds another dimension: Amazon led X-energy’s $700 million Series C-1 funding round in February 2025. Amazon has separately committed to developing SMRs with Energy Northwest (four reactors, 320 MWe initial capacity) and Dominion Energy (300 MWe at the North Anna nuclear station site). Amazon’s investment in X-energy signals that the hyperscaler views HTGR technology specifically — not just “nuclear” generically — as a strategic power source for its data center expansion.
The Fuel Supply Chain: TRISO-X and the May 2026 Milestone
A reactor is only as deployable as its fuel supply. X-energy anticipated this constraint and built TRISO-X, a dedicated fuel fabrication subsidiary. The TRISO-X facility (TX-1) in Oak Ridge, Tennessee, is designed to produce 5 metric tons of TRISO pebble fuel per year — enough to fuel up to 11 Xe-100 reactors.
NRC regulatory approval for TX-1 is anticipated by May 2026. When approved, it will be the first commercial TRISO fuel fabrication facility in the United States. This is a structural advantage that no other advanced reactor company has replicated: X-energy controls both the reactor design and the fuel supply chain. Kairos Power uses TRISO fuel but does not manufacture it. TerraPower’s Natrium uses HALEU metallic fuel from a different supply chain entirely.
Vertical integration of the fuel cycle is the kind of unsexy infrastructure investment that separates companies that build demonstration reactors from companies that deploy fleets.
The Competitive Landscape: Kairos, TerraPower, NuScale
Kairos Power is the closest competitor in advanced fission. Their Hermes demonstration reactor (fluoride salt–cooled, TRISO-fueled) received the first NRC construction permit for a non-light-water reactor in over 50 years (December 2023). Nuclear construction began May 2025; operational target is 2027. Hermes is a low-power test reactor, not a commercial unit — but the NRC precedent it sets is invaluable for the entire advanced reactor industry. Google signed a 500 MW multi-unit PPA with Kairos. The molten salt coolant enables passive safety similar to the Xe-100, but at lower outlet temperatures (~650°C vs. 750°C).
TerraPower broke ground on the Natrium reactor in Kemmerer, Wyoming in June 2024 — a 345 MWe sodium-cooled fast reactor with integrated molten salt energy storage. Bill Gates-backed, DOE ARDP-funded, and targeting operation by 2030. Natrium’s value proposition is load-following capability (the salt storage can ramp from 345 MWe to 500 MWe during peak demand). But sodium coolant introduces its own engineering challenges — sodium burns on contact with air and reacts violently with water — requiring specialized containment infrastructure that helium-cooled designs avoid entirely.
NuScale was the first SMR to receive NRC design certification (2023) for its VOYGR light-water SMR (77 MWe per module). However, the Carbon Free Power Project at Idaho National Lab was cancelled in November 2023 when subscription costs exceeded projections. NuScale’s stock has been volatile, and no U.S. construction project is currently underway, though the company has international agreements (Romania, South Korea). The VOYGR design is the most conventional advanced reactor — light-water cooled, evolutionary rather than revolutionary — and its cancellation exposed the economic vulnerability of any SMR that doesn’t offer something beyond “smaller version of existing technology.”
Why X-energy Wins the Fission Race for Industrial Power
The leaderboard for advanced fission isn’t one-dimensional. Different reactors win on different metrics:
• Fastest to NRC construction permit: Kairos (Hermes, Dec 2023)
• Largest single-unit output: TerraPower Natrium (345 MWe)
• First NRC design certification: NuScale VOYGR (2023)
• Highest outlet temperature / industrial dual-use: X-energy Xe-100 (750°C)
• Integrated fuel supply chain: X-energy (TRISO-X, May 2026 approval)
• Largest DOE + private investment committed: X-energy ($2.5B ARDP + $700M Amazon)
X-energy doesn’t win every category. But it wins the two that matter most for mass deployment: fuel supply chain control and dual-use thermal output. You cannot deploy a fleet of reactors if you can’t feed them. And you cannot serve both the data center market and the industrial decarbonization market from the same reactor unless your outlet temperature is high enough to produce process-grade steam.
The Xe-100 does both. No other advanced fission design on the market can say the same.
The Implications for AI Infrastructure and Industrial Decarbonization
The AI power crisis and the industrial decarbonization mandate are converging on the same solution. Data centers need carbon-free baseload electricity at the gigawatt scale. Refineries and chemical plants need carbon-free process heat to eliminate Scope 1 emissions without shutting down operations. Both markets are worth trillions of dollars over the next two decades.
X-energy’s Xe-100 is the only reactor design in active NRC licensing that can serve both markets from a single platform. The Dow Seadrift deployment will be the proof point: if a four-pack of Xe-100 modules can replace natural gas cogeneration at an operating petrochemical complex, the blueprint scales to every refinery, ammonia plant, and steel mill in the country.
And if Amazon’s investment thesis is correct, the same reactor design powers the data center campuses that are running the AI models that are accelerating the engineering work that designs the next generation of reactors.
That’s not a metaphor. That’s a feedback loop. And it starts in Seadrift, Texas, in 2026.
Timothy Porritt is founder of Porritt Inc., building AI-powered tools for heavy industry including NEXUS CAD and NORMEX Standards AI. A petroleum engineer by training, Timothy writes about the intersection of industrial engineering, AI, energy, and entrepreneurship.