Thorium is a naturally occurring, slightly radioactive metal that is about three times more abundant than uranium in the Earth’s crust. It is often pitched as the next great leap in nuclear power, but today it is used in tiny quantities for things like lamps and welding electrodes, and no commercial thorium-fuelled power plant exists. The technology remains experimental: the most advanced project is a 2 MWt research reactor in China that loaded thorium only in 2024.
This article explains what thorium is used for, how it becomes nuclear fuel, and why it has not replaced uranium despite its theoretical advantages. We look at the projects in China and India, past demonstration plants, company claims, and the reasons thorium is not mainstream yet.
What thorium is and what it is used for today
Thorium is element 90 on the periodic table. Its only natural isotope is thorium-232, which has a half-life of about three times the age of the Earth. It eventually decays into stable lead-208. The metal is found mainly in monazite, a rare-earth phosphate that averages 6–7% thorium phosphate. Because monazite is mined primarily for its rare earth elements, thorium is usually a by-product, and only a small fraction of it is recovered and used. Without demand for rare earths, thorium would largely remain in the ground.
Thorium has limited uses right now. According to the US Geological Survey (2026), its commercial applications are niche: catalysts, high-temperature ceramics, magnetrons in microwave ovens, metal-halide lamps, nuclear medicine, optical coatings, tungsten filaments and welding electrodes. Thorium dioxide (thoria) has a melting point of about 3,300 °C, which is why it has been used in lamp filaments, gas lantern mantles and heat-resistant ceramics. Because of its natural radioactivity, non-radioactive substitutes have largely replaced it: yttrium compounds in lamp mantles, and cerium, lanthanum, yttrium and zirconium oxides in welding electrodes. The scale is small: the value of thorium compounds imported into the US in 2025 was estimated at just $118,000.
Thorium is not used to generate commercial electricity anywhere in the world as of October 2026. The Chinese HTR-PM plant, which entered commercial operation in December 2023, is sometimes mentioned in this context but is not thorium-fuelled. Thorium has produced grid electricity in the past, but only in demonstration plants that closed decades ago (see below).
How thorium becomes nuclear fuel
To understand why thorium is not mainstream, you need to understand the fuel cycle. Thorium-232 is fertile, not fissile. This means it cannot sustain a chain reaction on its own. To generate power, it must first be converted into uranium-233, which is fissile.
The conversion begins when Th-232 absorbs a neutron. It becomes Th-233, which beta-decays (with a half-life of about 22 minutes) into protactinium-233. Pa-233 then decays, with a half-life of about 27 days, into U-233. U-233 can then undergo fission, releasing about 200 MeV of energy, similar to the energy released by uranium-235.
However, this process requires a “driver”. A fissile material such as U-233, U-235 or plutonium-239 is needed to start and sustain the reaction. Low-enriched uranium (up to 20% U-235) is generally uneconomic as a driver unless the fuel burn-up is very high. Thorium itself does not need enriching, but because it depends on a driver, usually enriched uranium or plutonium, it does not remove the need for enrichment or fissile material.
Not all reactors are suitable for this conversion. Heavy water reactors, such as the CANDU design, have excellent neutron economy and are well suited to thorium. In contrast, pressurised water reactors (PWRs) can use thorium but cannot convert significant amounts of it into U-233. Fast reactors gain little advantage from thorium compared with using depleted uranium.
Molten salt reactors and the search for efficiency
The design most commonly associated with thorium is the molten salt reactor (MSR). In a thorium MSR, the fuel is dissolved directly in molten fluoride salt; this version is also called the Liquid Fluoride Thorium Reactor (LFTR). The liquid fuel allows different safety features, such as a freeze plug that melts to drain the fuel into a subcritical tank in an emergency, and a negative temperature coefficient that slows the reaction as it heats up.
The history of MSRs is short. Oak Ridge National Laboratory in the USA ran the Molten Salt Reactor Experiment from 1965 to 1969. It used U-233 fuel and operated at up to 7.4 MWt, using a lithium-beryllium fluoride salt at 600–700 °C. It showed that the concept could work at small scale.
Supporters argue that MSRs are safer than solid-fuel reactors because they operate at near atmospheric pressure. That applies to molten salt designs, not to thorium fuel in general, which can also be used in conventional reactors at normal pressures. MSRs also face significant engineering hurdles, including corrosion of materials by the hot salts and online fuel processing.
Real-world projects: China, India and commercial claims
Why is there no thorium power station today? The answer lies in the current state of the major projects.
China: the most advanced thorium project
China currently leads thorium development. The Shanghai Institute of Applied Physics (SINAP) built the TMSR-LF1, a 2 MWt experimental liquid-fuel thorium molten salt reactor at Wuwei, Gansu. The reactor received its operating licence in June 2023 and reached first criticality on 11 October 2023. It reached full power in June 2024, and thorium was loaded into the core in October 2024.
The reactor holds about 50 kg of thorium and uses uranium enriched to below 20% U-235 as its driver. In October 2025 it achieved the first thorium-to-uranium conversion in a molten salt reactor, with protactinium-233 detected after ten days at full power (announced in November 2025).
Despite this progress, TMSR-LF1 is a small experimental reactor, not a power station. No electricity generation has been reported. The World Nuclear Association cites a start-up budget of $350 million for the programme. China plans a 10 MWe (about 30 MWt) reactor at Wuwei with first criticality targeted for 2030, and a 100 MWt-class demonstration by about 2035. These are targets, not guaranteed commitments.
India: a long-term three-stage strategy
India has some of the world’s largest thorium resources, held mainly in monazite sands. Its nuclear programme is built on a three-stage plan:
- Pressurised heavy water reactors on natural uranium produce plutonium.
- Fast breeder reactors burn that plutonium and breed more plutonium and U-233 using a uranium/thorium blanket.
- Advanced Heavy Water Reactors burn thorium-plutonium fuel and breed U-233 to drive a fleet of thorium reactors.
The Prototype Fast Breeder Reactor (PFBR) at Kalpakkam is the stage-two prototype. It is a 500 MWe reactor that began core loading in March 2024 and reached first criticality in April 2026. However, the PFBR runs on plutonium-uranium fuel, not thorium; its role in the thorium cycle comes later, through breeding. The third-stage Advanced Heavy Water Reactor (AHWR) was envisaged to operate around 2022, but in August 2025 the government told parliament that the project had not been formally launched.
Past demonstration plants
Thorium has made grid electricity before. Germany ran the AVR (15 MWe, 1967–1988) and the THTR-300 (300 MWe, 1983–1989, shut after technical problems). In the US, Peach Bottom 1 (40 MWe, 1967–1974), Fort St Vrain (330 MWe, 1976–1989) and the Shippingport Light Water Breeder Reactor (60 MWe, 1977–1982) also used thorium. None led to a commercial thorium fleet.
Other projects and company claims
Several companies are developing thorium reactors or fuels. Their timelines are company plans, not independently verified:
- Copenhagen Atomics (Denmark) says it has built two full-size non-fission prototypes of a containerised 100 MWt molten salt reactor. It plans its first nuclear chain reaction at the Paul Scherrer Institute in Switzerland in 2028, with commercial units planned for the early 2030s.
- Thorizon (Netherlands/France) is developing the 250 MWt Thorizon One, whose fuel combines thorium with long-lived elements from used fuel. The company says construction starts in the early 2030s.
- Clean Core Thorium Energy (USA) is developing a thorium-plus-HALEU fuel for CANDU-type reactors. Test rodlets have been irradiated at Idaho National Laboratory since May 2024, and the remaining rodlets passed 60 GWd/tU burn-up in May 2026. The fuel completed phase one of the Canadian Nuclear Safety Commission pre-licensing review in 2024.
The numbers: output, efficiency and costs
The economic case for switching is weak. As the World Nuclear Association notes, uranium is cheap and plentiful, so there is little economic incentive to move to thorium.
Thorium dioxide fuel is chemically inert, conducts heat better and melts at a higher temperature than uranium dioxide. The thorium cycle also produces fewer transuranic elements (such as plutonium and americium), which dominate the long-term radiotoxicity of nuclear waste. Waste is therefore reduced over very long timescales, but not eliminated: fission products still require long-term management.
Cost data is scarce because no commercial plants exist. Company claims vary widely; Copenhagen Atomics has cited electricity prices “below $20 per MWh” in one place and “$30–40/MWh” in another. Without built plants, these figures are speculative.
The UK angle
The UK has a historical connection to thorium. The Dragon reactor, a 20 MWth high-temperature gas-cooled reactor, used thorium-containing fuel and ran from 1964 to 1973.
As of October 2026, there is no thorium reactor operating, under construction or licensed in the UK, so none of the nuclear output in the live GB electricity mix comes from thorium. The Office for Nuclear Regulation’s Generic Design Assessment list includes the Rolls-Royce SMR, BWRX-300 and SMR-300, but no thorium designs. The USGS lists the UK among countries where thorium research has been under way, but UK new-build plans centre on large EPRs and small modular reactors that run on uranium.
Limitations: why thorium is not mainstream yet
Thorium is not a simple substitute for uranium. It requires a fissile driver, usually enriched uranium, to start. Fuel fabrication is expensive because of radioactive contaminants, and reprocessing is complex. The intermediate isotope, protactinium-233, absorbs neutrons, which reduces efficiency unless it is removed or allowed to decay.
The proliferation resistance of thorium is also only partial. U-233 bred from thorium is contaminated with U-232, whose decay products emit strong gamma radiation, making it hard to handle and easy to detect. Even so, the IAEA treats U-233 as direct-use material. The US tested a U-233/plutonium device in 1955, and India tested a small U-233 device in 1998.
Finally, the engineering challenges of molten salt reactors (corrosion, materials and online processing) remain unproven at commercial scale. Our guide to the pros and cons of nuclear energy covers the wider trade-offs; thorium adds a layer of complexity that has yet to be justified by cost or reliability.
Frequently asked questions
What is thorium?
Thorium is a naturally occurring, slightly radioactive metal found in the Earth’s crust. It is about three times more abundant than uranium, with major resources in India, Brazil, Australia and the US. Its main ore is monazite.
What is thorium used for?
Today, thorium is used mainly in non-energy applications: welding electrodes, high-temperature ceramics, lamp filaments, optical coatings, catalysts, magnetrons and nuclear medicine. It is not used to generate commercial electricity.
Is thorium better than uranium?
Thorium is more abundant, produces fewer long-lived transuranic elements and has better thermal properties as a fuel. However, it needs a fissile driver such as enriched uranium or plutonium, its waste is reduced rather than eliminated, and the fuel cycle is currently more expensive and complex than the established uranium cycle.
Is there a working thorium reactor?
Yes, but not a commercial one. China’s TMSR-LF1 in Wuwei reached criticality in October 2023 and loaded thorium in October 2024. It is an experimental 2 MWt reactor, not a power station.
Does the UK have a thorium reactor?
No. The UK’s Dragon reactor used thorium-containing fuel from 1964 to 1973, but no thorium reactor is currently operating, under construction or licensed in the UK.
Why isn’t thorium used in nuclear power stations yet?
Thorium reactors are still experimental. Thorium needs a fissile driver fuel, its fuel fabrication and reprocessing are complex, and new reactor types such as MSRs have not been proven commercially. With uranium cheap and plentiful, there is little economic incentive to switch.