X-Energy Has Finally Ordered Something That Matters | Nuclear Now #67
The nuclear revival does not need another memorandum of understanding. It needs factories to start making the components that determine whether a reactor can ever be built. X-Energy’s first order for steam-generator tubing is therefore real progress, not another press-release milestone, although calling it a construction breakthrough would be premature.
The order, announced November 4, covers safety-significant, long-lead tubing for the first Xe-100 deployment, a proposed four-unit plant for Dow in Texas. Alleima will supply the advanced stainless-steel material and Doosan Enerbility will participate in the manufacturing chain. That matters because steam generators sit in the part of a nuclear plant where design, metallurgy, quality assurance, welding, inspection, and nuclear-code qualification all collide. Ordering the tubing means X-Energy is converting a paper design into a controlled industrial supply chain.
It does not mean X-Energy has a commercial reactor under construction. The Dow project remains dependent on licensing, financing, final engineering, and a supply chain capable of repeating the work at scale. The company is also advancing larger ambitions, including projects with Amazon, Energy Northwest, AEP, and Centrica, but gigawatts announced are not gigawatts financed. The Indiana proposal, for example, is still a feasibility-stage concept targeting a possible final investment decision late in the decade.
That distinction is the entire story. X-Energy is now doing something many advanced-reactor companies have avoided: spending money on nuclear-grade hardware before it has secured the full commercial order book. That is evidence of seriousness. It is also where the difficult economics begin.
The Xe-100 is a 80-MWe high-temperature gas reactor, with four units typically configured as a 320-MWe plant. The Indiana concept would use 12 reactors for 960 MWe. The architecture is attractive for one reason utilities understand immediately: smaller units can be added in blocks rather than committing the balance sheet to one enormous reactor. That can reduce construction-period exposure and allow capacity to track load growth.
But smaller does not automatically mean cheaper. A 960-MWe site made up of 12 reactors carries 12 reactor pressure boundaries, 12 sets of instrumentation, 12 maintenance programs, and more duplicated equipment than a conventional two-unit plant. X-Energy’s economic case depends on standardization, factory production, shared site infrastructure, and a high enough production run to spread engineering and licensing costs across many units. Without that repetition, the SMR label is merely a description of scale.
The fuel system creates another economic variable. Xe-100 uses TRISO fuel, a robust, accident-tolerant fuel form manufactured as coated particles. Its safety characteristics are a major technical advantage, but the commercial fuel supply chain is not as deep as the one serving conventional light-water reactors. Fuel qualification, production capacity, quality control, and pricing all have to mature alongside the reactor. A reactor design can be licensed on schedule and still miss its operating date if its fuel plant is not ready.
The first tubing order is valuable precisely because it addresses this less glamorous problem. Long-lead components define schedules. They require qualified suppliers, documented processes, inspection records, and production slots. They cannot be conjured after a customer signs a power purchase agreement. X-Energy is showing that it understands nuclear deployment as an industrial program, not simply as a licensing exercise.
The financial benchmark remains brutal. A project must compete with gas turbines that are fast to build, with renewables whose equipment costs continue to fall, and with batteries that can arbitrage electricity over shorter durations. Nuclear wins when buyers value firm, low-carbon power, fuel-price stability, land efficiency, and decades of operation. Amazon’s interest reflects that exact problem: data centers need power at a scale and reliability profile that intermittent supply alone cannot guarantee.
The first plant will not prove the mature fleet’s economics. It will carry first-of-a-kind engineering, regulatory, financing, and manufacturing costs. The relevant question is not whether the first Xe-100 is cheap. It will not be. The relevant question is whether X-Energy can make the second, fourth, and twelfth units materially cheaper by preserving the same design and building an actual production system.
That is why the tubing order matters more than another announced customer. Customers create demand signals. Components create schedule credibility.
The immediate beneficiary is X-Energy’s Dow project in Texas. The order gives the project a procurement trail and forces design decisions that cannot be postponed indefinitely. It also strengthens X-Energy’s position with industrial customers that need process heat or dedicated power, not merely another certificate for clean electricity.
Amazon and AEP benefit in a different way. Their proposed 960-MWe Indiana project would repurpose a retired coal site, use existing grid infrastructure, and anchor the development around a large commercial buyer. That structure is more credible than a developer asking ratepayers to fund speculative capacity. Amazon’s proposed role as development funder and power purchaser addresses the central problem facing advanced nuclear: who absorbs the early risk before the technology has a bankable operating history?
It still does not eliminate that risk. A feasibility agreement is not a final investment decision, and a target of deciding by 2029 leaves years for cost escalation, licensing delays, supply-chain bottlenecks, and changing data-center demand. A project with 12 units also raises the standard for execution. One delayed module is manageable. A production line that cannot deliver all 12 on schedule destroys the central economic argument.
Doosan Enerbility is another important name to watch. The company is not merely providing a logo for a partnership. Its role in manufacturing major nuclear components gives X-Energy access to industrial capability that advanced-reactor startups generally lack. The reactor developer still owns the design and licensing risk, but deployment requires a builder with experience turning engineering drawings into inspected metal.
This also puts pressure on TerraPower, NuScale, GE Vernova, and other advanced-nuclear contenders. The market is moving from design claims to procurement evidence. NuScale has the stronger regulatory history for its approved light-water design, while X-Energy is pursuing a more differentiated reactor and fuel platform. Neither advantage is sufficient by itself. The winner will be the company that can deliver repeatable units at a price customers can finance.
The next meaningful signals are not additional memoranda. They are a construction permit application, a firm engineering and procurement contract, fuel-production readiness, a binding customer commitment, and a financing structure that protects utility customers without hiding the cost from investors. If X-Energy reaches those milestones, this tubing order will be recognized as the beginning of a real deployment chain. If it does not, the order will remain an expensive component attached to an unbuilt reactor.
The nuclear revival is entering its most useful phase, where optimism has to survive contact with procurement, metallurgy, financing, and construction management. X-Energy has cleared one of those hurdles. It has not won the race, but it has stopped pretending that a press release is the same thing as a plant.