A molten salt reactor (MSR) is a type of nuclear reactor that uses salt heated to a liquid state as its coolant, its fuel, or both. Unlike conventional nuclear plants, which pump water under extreme pressure to carry heat away from solid fuel rods, an MSR circulates molten salt at near-atmospheric pressure and at temperatures that can reach 700°C. This fundamental difference in chemistry and physics gives MSRs a distinct set of advantages and engineering challenges compared to the water-cooled reactors that generate most of the world’s nuclear electricity today.
How MSRs Differ From Conventional Reactors
The vast majority of operating nuclear plants are pressurized water reactors (PWRs). In a PWR, water serves as both the coolant and the moderator (the substance that slows neutrons to sustain the chain reaction). That water must be kept under enormous pressure, roughly 150 times atmospheric pressure, to prevent it from boiling at reactor temperatures. This high-pressure environment demands thick steel vessels and creates the risk of a pressure rupture, which is a central concern in reactor safety design.
Molten salt absorbs huge amounts of heat without needing to be pressurized. An MSR can operate at high temperatures and low pressures, which eliminates the possibility of a pressure-driven steam explosion. The reactor vessel and piping can be lighter and simpler because they don’t need to contain a high-pressure system. And because salt has a boiling point far above normal operating temperatures, the coolant stays liquid throughout routine operation without the constant mechanical effort of pressurization pumps.
Two Main Design Approaches
MSRs in development today fall into two broad categories. The first uses traditional solid fuel, typically ceramic-coated fuel pebbles, with molten salt flowing past them as a coolant instead of water. Kairos Power’s Hermes reactor, now under construction in Oak Ridge, Tennessee, follows this model. It uses a fluoride salt coolant paired with solid fuel pebbles, and received its construction permit in late 2024.
The second design dissolves the nuclear fuel directly into the salt itself, creating a liquid fuel. Uranium or thorium is mixed into the molten salt so that the fuel and coolant are the same fluid. This eliminates the need to manufacture and eventually dispose of solid fuel rods. It also means fuel can be added or removed from the circulating salt during operation, without shutting the reactor down. This liquid-fuel approach was first demonstrated in the 1950s and 1960s at Oak Ridge National Laboratory and is now the basis for several next-generation designs.
The Freeze Plug Safety System
One of the most distinctive safety features in many MSR designs is the freeze plug. Below the reactor sits a drain tank connected by a pipe. During normal operation, a small plug of solidified salt blocks that pipe, kept frozen by active cooling. If the reactor overheats or loses power, the plug melts on its own, and gravity drains the entire fuel salt into the tank below. Once spread out in the drain tank, the fuel can no longer sustain a chain reaction. The reactor shuts itself down without any human intervention, backup generators, or emergency pumps.
This is what engineers call passive safety: the laws of physics handle the emergency rather than relying on equipment or operators to respond correctly. In a conventional reactor, emergency shutdowns depend on control rods being mechanically inserted and backup cooling systems activating. The freeze plug works even during a total power failure, which is precisely the scenario that caused the 2011 Fukushima disaster in Japan’s water-cooled reactors.
Operating Temperatures and Efficiency
MSRs run significantly hotter than conventional reactors. While a typical PWR operates around 300°C, molten salt systems reach 600 to 700°C. That higher temperature isn’t just a technical detail. It directly translates into better thermal efficiency, meaning more of the heat gets converted into electricity rather than being wasted.
A standard PWR converts roughly 33% of its thermal energy into electricity. MSR designs project thermal efficiencies around 44%. That’s a substantial jump: for the same amount of nuclear fuel consumed, an MSR could generate about a third more electricity. The higher temperatures also open the door to industrial applications beyond electricity, such as hydrogen production and chemical manufacturing, which require intense heat that water-cooled reactors can’t easily provide.
The Thorium Fuel Option
MSRs are uniquely suited to using thorium as a fuel, which is roughly three to four times more abundant in Earth’s crust than uranium. Thorium itself isn’t directly fissionable, but when it absorbs a neutron inside a reactor, it transforms through a series of steps into uranium-233, which is. The thorium atom captures a neutron, becomes an unstable intermediate that decays over about 27 days, and ultimately produces a form of uranium that sustains the chain reaction.
This conversion process works especially well in a liquid-fuel MSR because the chemistry of the circulating salt can be managed continuously. Fresh thorium can be fed in and byproducts removed during operation. China’s experimental TMSR-LF1 reactor in Wuwei, Gansu Province, is currently the only operating molten salt reactor in the world loaded with thorium fuel. It achieved first criticality in October 2023 and reached full operation by June 2024, successfully demonstrating thorium-to-uranium conversion and collecting experimental data on the process.
The Corrosion Problem
The central engineering challenge for MSRs is that molten salt, particularly at 700°C, is extremely corrosive. It attacks most conventional metals and can degrade reactor components over time. This has been the primary reason MSRs haven’t moved from laboratory experiments to commercial power plants in the decades since the concept was first proven.
Specialized alloys have been developed to withstand this environment. Oak Ridge National Laboratory created Hastelloy N, a nickel-based alloy containing molybdenum and chromium, specifically for use with fluoride salts. China’s Institute of Metal Research developed a similar alloy called GH3535. Both have shown good corrosion resistance in fluoride salt systems at 700°C. Chloride salts, used in some newer designs, present an even tougher corrosion challenge, and researchers at institutions like Delft University of Technology are actively working on understanding how fission products behave in these salts at high temperatures.
Beyond corrosion, the radioactive salt itself poses maintenance questions. Every pump seal, valve, and heat exchanger in the primary loop contacts intensely radioactive fluid, making inspection and repair more complex than in reactors where the fuel stays contained in solid rods.
Waste: Not a Simple Story
MSR advocates often claim the technology will dramatically reduce nuclear waste. The reality is more nuanced. A 2022 study published in the Proceedings of the National Academy of Sciences found that several small modular reactor designs, including molten salt types, could actually increase the volume of nuclear waste needing management by factors of 2 to 30 compared to conventional large reactors.
One reason is burnup, the measure of how thoroughly a reactor consumes its fuel. A conventional PWR achieves a burnup around 50 to 55 megawatt-days per kilogram of fuel. Terrestrial Energy’s 400-megawatt molten salt design, by comparison, projects a burnup of only 14 megawatt-days per kilogram. Lower burnup means more fuel passes through the reactor per unit of energy produced, generating more spent material. That spent fuel also contains higher concentrations of plutonium isotopes, which increases its long-term radioactivity.
Liquid-fuel MSRs do offer an important theoretical advantage: the ability to continuously reprocess the salt, removing waste products and recycling usable fuel while the reactor runs. If this online reprocessing works at commercial scale, it could significantly improve fuel utilization and reduce the most problematic long-lived waste. But continuous salt reprocessing has never been demonstrated outside of small experiments, and scaling it up remains one of the technology’s biggest open questions.
Where MSR Development Stands
MSRs are not yet generating commercial electricity anywhere in the world. China’s 2-megawatt TMSR-LF1 is a research reactor, not a power plant. For context, a typical large nuclear plant produces 1,000 megawatts or more of electrical power.
In the United States, Kairos Power is furthest along in the licensing process, with construction of its Hermes test reactor underway in Tennessee. Hermes is a fluoride-salt-cooled design using solid fuel and will not generate electricity. It’s a demonstration project to validate the technology before building a commercial version. Kairos has also applied for permits to build Hermes 2, a follow-on pair of test reactors at the same site.
Several other companies are pursuing MSR designs at various stages. Terrestrial Energy in Canada is developing a liquid-fuel molten salt design. Copenhagen Atomics in Denmark is working on a thorium-fueled, heavy-water-moderated molten salt reactor. Most of these projects are still years away from producing power commercially, with optimistic timelines placing the first grid-connected MSRs in the early to mid-2030s. The technology is real, well-grounded in physics, and advancing. But the gap between a working experiment and a cost-competitive power plant remains significant.