Top Molten Salt Reactor Stocks 2026: Meltdown-Proof Nuclear

Every current nuclear reactor has the same weak spot: water.

Water boils at 100°C. So a normal reactor runs its coolant at 300°C+ under enough pressure to keep it liquid, wrapped in a steel vessel thick enough to hold that pressure, inside a concrete dome in case the vessel fails. Most of the cost of a nuclear plant is the cost of babysitting hot water.

A molten salt reactor (MSR) throws the water out. The coolant is a salt (fluoride or chloride salts, not table salt) that melts around 450°C and doesn't boil until roughly 1,400°C. It runs at atmospheric pressure. In the most radical versions, the uranium is dissolved in the salt, so the fuel and the coolant are the same liquid.^[1]

Water boils at 100°C. Reactor salt boils at 1,400°C. That gap is the investment thesis.

If something goes wrong, a frozen salt plug at the bottom of the vessel melts, the fuel drains by gravity into a tank shaped so it can't sustain a chain reaction, and it freezes solid. No pumps, operators, or backup diesel required. The fuel is already melted, so there's nothing left to melt down.

Oak Ridge proved all of this works between 1965 and 1969. Then the U.S. picked water-cooled reactors, shut the experiment down, and forgot about it for 50 years.

Here's why it matters now. The thing was reinvented at exactly the moment hyperscalers started begging for 24/7 power and refineries started asking for 600°C+ process heat that water reactors can't deliver. So the last 24 months looked like this:

• China's TMSR-LF1 went critical in October 2023 and, in November 2025, became the first reactor ever to breed uranium-233 from thorium. A 10 MWe follow-on is under construction for 2030.
• Terrestrial Energy listed on Nasdaq in October 2025 with roughly $292 million in the bank, the first liquid-fuel pure-play any retail investor could buy.
• Kairos Power broke ground on Hermes 2 in April 2026, the first Gen IV power reactor the NRC has ever let anyone build, with Google as the anchor customer.
• The DOE approved its first-ever safety design agreement for a molten salt reactor (Natura's MSR-1) in July 2026.

Investors get paid, or don't, based on whether those dates hold.

We expect the first Western liquid-fuel reactor to go critical in the next 18 months (Natura, at Abilene Christian University), the first commercial-scale construction permit application to be filed (Terrestrial, at Texas A&M), and at least one private MSR developer to announce an IPO. The technology risk is mostly retired. What's left is schedule risk.

We're slicing the theme by Pure-plays (the reactor is the business), Integrated (big utilities using MSRs as an R&D option), Picks-and-shovels (whoever sells the salt, the fuel, and the isotopes), and Private bellwethers, because four of the five most important MSR companies on Earth don't have a ticker yet.

There is one. Every other liquid-fuel developer is private, state-owned, or buried inside a conglomerate, so for now this segment is a single stock and it carries the whole theme's volatility.

Terrestrial Energy NASDAQ: IMSR

Terrestrial Energy came public on October 29, 2025 through a merger with HCM II Acquisition Corp, raising about $292 million gross with almost no redemptions. The product is the Integral Molten Salt Reactor: a 390 MWe plant where the fuel salt, graphite moderator, pumps, and heat exchangers all sit sealed inside one swappable "core-unit" that gets replaced every seven years. The plant is designed to be sited next to the customer, whether that's a data center or a chemical plant, and to sell high-temperature heat as well as electricity.

Two things separate IMSR from the rest of the SMR pile. First, it runs on standard-assay low-enriched uranium, the same stuff every existing reactor uses. No HALEU. That matters because HALEU is the supply-chain bottleneck strangling Oklo, X-energy, and half the field. Westinghouse is already lined up to supply the uranium tetrafluoride. Second, management isn't planning to own plants. The model is to sell core-units and fuel salt under long-term contracts, and in the Q2 2026 report they raised their estimate of lifetime revenue per plant to $2.7 billion at a 33% blended gross margin, with 79% of that revenue landing after construction.

What's actually happened in 2026: the NRC approved two foundational topical reports (Principal Design Criteria and Postulated Initiating Events), Texas A&M leased the company 77 acres at its RELLIS campus for the first commercial plant, Zachry signed on for site engineering, and Riot Platforms signed an MOU to co-locate up to 4 GW of IMSR capacity with its data centers. Liquidity was $283.4 million at June 30, and burn is only about $2.2 million a month, so the runway is long. The stock has round-tripped from the mid-teens down into the single digits this year, which tells you the market wants a construction permit application, not press releases.

Big utilities that treat molten salt as an R&D option on their balance sheet. You're buying a regulated dividend payer with a free call option attached.

Southern Company NYSE: SO

Southern Company is the prime contractor on the Molten Chloride Reactor Experiment (MCRE), a partnership with TerraPower, Core Power, and the DOE to build the first fast-spectrum, liquid-fuel salt reactor ever to go critical. It's tiny, a few hundred kilowatts inside a test cell at Idaho National Laboratory, and the first batch of fuel salt was produced in December 2025. The experiment is slated to run in 2028.

Why a Georgia utility? Southern already built the last two big reactors in America (Vogtle 3 and 4) and wants first dibs on whatever comes next. The company also runs, with TerraPower, the largest chloride salt test loop on the planet in Everett, Washington. To be clear: MCRE won't move Southern's earnings. It's a $170 million cost-shared program at a company with a market cap north of $100 billion. But if you want MSR exposure that pays you a dividend while you wait, this is the only one.

An MSR needs three things nobody else needs in quantity: purified fluoride or chloride salt, uranium in a fluoride form instead of oxide pellets, and lithium-7.^[2] The reactor companies are exciting. The companies selling them the ingredients are where the fewer-headlines, steadier-money bet usually lives.

Cameco NYSE: CCJ

Cameco is the world's second-largest uranium miner and owns 49% of Westinghouse, which is the outfit converting uranium into the tetrafluoride Terrestrial needs for its fuel salt. Every IMSR plant is a multi-decade fuel-salt contract, and Westinghouse is the front of that supply line. Scale check: MSR fuel is a rounding error to Cameco today. You buy CCJ because you believe nuclear demand goes up across the board, and molten salt is one more reactor type pulling on the same uranium supply. The kicker is that IMSR uses ordinary enrichment levels, so Cameco's existing conversion and enrichment partners can serve it without building anything new.

ASP Isotopes NASDAQ: ASPI

ASP Isotopes is the speculative one. The company enriches isotopes in South Africa using a laser process and a proprietary aerodynamic method, and its nuclear-fuel subsidiary, Quantum Leap Energy, has lithium-7 on its product roadmap alongside HALEU. Here's why that matters: lithium-7 is the "Li" in FLiBe, the salt most thermal-spectrum MSRs use, and each reactor needs tens of tonnes of it enriched to 99.99%+. The World Nuclear Association pegs global demand at about one tonne a year today, rising to roughly 250 tonnes a year once MSRs commercialize. Today, essentially all of it comes from China and Russia. Natura had to get its FLiBe from the DOE's leftover 1960s stockpile. A Western producer of lithium-7 at scale doesn't exist yet, and whoever gets there first has every MSR developer on Earth as a captive customer. ASPI is pre-revenue on that specific line and trades like a lottery ticket. Size accordingly.

Here's the awkward truth about this watchlist: the four most important MSR programs in the West have no ticker. Track them anyway, because their milestones move IMSR and the suppliers.

Kairos Power (Private)

Kairos is the best-funded name in the space and a cousin rather than a sibling: its KP-FHR uses solid TRISO fuel pebbles cooled by molten FLiBe salt, so the salt is the coolant but not the fuel. The regulatory path is easier, and it shows. Hermes 1, a 35 MWt test reactor in Oak Ridge, has been under nuclear construction since May 2025. Hermes 2 broke ground on April 17, 2026, will deliver 50 MWe to the Tennessee Valley Authority starting in 2030, and is the first deployment under Kairos's deal to supply Google with 500 MW by 2035. TVA's power purchase agreement is the first any U.S. utility has ever signed with a Gen IV reactor. No IPO has been announced, but with Google as an anchor customer and steel already in the ground, Kairos is the obvious candidate to go next.

TerraPower (Private)

Everyone knows TerraPower for Natrium, the Bill Gates-backed sodium-cooled reactor going up in Wyoming. Fewer people know that Natrium's molten salt is just heat storage, and that TerraPower's other reactor, the Molten Chloride Fast Reactor, is a true liquid-fuel MSR. MCRE at Idaho (see Southern Company above) is the physics test for it, with fuel already made and operation targeted for 2028. The MCFR is designed to eat spent fuel and run hot enough for industrial heat, and Core Power wants a marine version for ships. TerraPower has raised well over a billion dollars privately and has never needed public markets. If it ever does, it would be the largest name in the segment.

Natura Resources (Private)

Natura is a Texas company founded by an oil-and-gas veteran, and it's about to do the thing no one else in America has done since 1969: run a liquid-fuel reactor. Its 1 MWt MSR-1 at Abilene Christian University got the first NRC construction permit ever issued for a liquid-fuel design in September 2024, the building is finished, the DOE handed it the original Oak Ridge FLiBe salt in January 2026, and on July 21, 2026 the DOE approved its Nuclear Safety Design Agreement, the first for any molten salt reactor. Management says criticality in 2026; we'd pencil in the first half of 2027. A 100 MWe commercial follow-on, the MSR-100, is planned with Texas A&M, and a June 2026 deal with Quadrant lines up HALEU for it. No IPO chatter yet, but a Texas MSR going critical on a college campus is the sort of headline that changes that.

Copenhagen Atomics (Private)

The Danish thorium company is the one saying the quiet part out loud. CEO Thomas Jam Pedersen has told the press he's planning an IPO, prefers the U.S. market because nuclear names fetch higher valuations there, and thinks the company could be worth up to $7 billion (versus roughly $700 million today) once its test reactor is running. Copenhagen Atomics raised close to DKK 400 million in 2026 toward a €200 million target, has run a molten salt pump continuously for two years, and plans a physical test in Switzerland in 2027 or 2028 and a 1 MWt test reactor in 2028. Unlike most of the field it already has revenue, selling salt test loops, purified salts, and enriched lithium to labs and fusion programs. The pitch is mass-manufactured 100 MWt thorium breeders at $50 million a pop. Ambitious. Also the most likely new MSR ticker of 2027.^[3]

The salt eats the reactor. Hot fluoride salt full of fission products is nasty stuff, and the nickel alloy Oak Ridge invented for it (Hastelloy-N) was only ever tested for four years. Terrestrial sidesteps this by throwing the whole core-unit away every seven years. TerraPower is using chloride salts with different corrosion chemistry. Everyone is running irradiation campaigns (Terrestrial's graphite is being cooked at NRG Petten in the Netherlands right now) precisely because nobody has 30-year data. The tell: any developer quietly extending its test campaigns or adding "materials qualification" line items to its budget.

The calendar. This industry's business model is a promised date. Natura said 2026. Kairos said 2030. Copenhagen Atomics said 2028. Every slipped quarter is a capital raise at a worse price, and IMSR's burn, while low today, is designed to climb as construction starts. Three dates matter: Natura's criticality, IMSR's construction permit filing at RELLIS, and MCRE loading fuel at Idaho in 2028. Miss two of those three and the theme goes back in the drawer for another five years.

The quieter risk is supply chain. Lithium-7 comes from China and Russia. China has the only operating MSR on the planet and a 100 MW demo scheduled for 2035. If the West stalls, molten salt becomes something we import.

The next 12 to 18 months are the most eventful stretch molten salt has had since Lyndon Johnson was president. Here's what to circle on the calendar:

• Natura MSR-1 criticality (late 2026 to H1 2027). The first liquid-fuel reactor to run in the West in 57 years. This is the headline that gets retail money to learn the acronym.
• Terrestrial's construction permit application for Texas A&M-RELLIS. Site characterization is underway; the filing is the single biggest de-risking event for IMSR. Also watch whether the Riot Platforms MOU turns into a binding agreement.
• MCRE construction at Idaho's LOTUS test bed ahead of the 2028 run, plus DOE progress on Terrestrial's TETRA reactor pilot and TEFLA fuel pilot.
• Kairos Hermes 1 completion and the operating license application for Hermes 2. Any sign of Kairos raising publicly would reprice the whole category.
• A Copenhagen Atomics IPO filing. Management has telegraphed it. If it lands on a U.S. exchange, IMSR finally gets a listed comp.
• China's 10 MWe TMSR construction milestones in Gansu, and Stellaria's prototype authorization in France, as a read on whether Europe and Asia move faster than the NRC.

Bottom line: molten salt went from museum exhibit to construction site in about three years. By this time next year, the list of tickers should be longer. Subscribe and we'll tell you when.