Circulating salt delivers efficient energy production

Molten salt at 700 degrees Celsius bears a striking resemblance to ordinary tap water. ‘It’s quite fluid, not very viscous’, says Sander de Groot, founder and technical director of Thorizon, a Dutch start-up working on a molten salt reactor. This is a type of nuclear reactor in which the nuclear fuel is dissolved in molten salt at hundreds of degrees, which acts as a coolant. In other common nuclear reactors, the coolant is usually water.

‘We have tested one circuit with circulating salt, and are now improving the design’, De Groot explains at Thorizon’s new premises in the Energy Transition Campus Amsterdam (ETCA) building, on the north bank of the IJ. In the barely furnished laboratory, De Groot and chemist Jérémie Sobel show the stainless steel crucibles used for the molten salt experiments. ‘There’s plenty to work on: testing materials, implementing sensors, and all the while we’re learning how the salt behaves’, says Sobel. In a video, he demonstrates how the molten salt is pumped around using a special pump, like water in a fountain. The precise composition is a trade secret, ‘but it contains chloride.’

An established principle

The principle behind the molten salt reactor is not new: in the United States, the Molten Salt Reactor Experiment (MSRE) ran in Oak Ridge, Tennessee during the 1960s, but was never followed up. ‘Yet the principle simply works’, says De Groot, who previously worked at nuclear company NRG in Petten. He was one of the founders of Thorizon in 2018; the company now employs 45 people and has recently opened an office in France as well. Thorizon has so far received around €42 million in start-up grants and investment. It is, incidentally, not the only party working on this type of reactor. In China, the experimental TMSR reactor became operational in 2023, and research institutes in France, Germany, Japan and the United States are also working on molten salt reactors. In addition, start-ups such as the American companies TerraPower and Elysium, the Indonesian ThorCon, and the Danish firms Copenhagen Atomics and Saltfoss Technologies are also engaged in this work.

‘Molten salt is just like water, and achieving proper seals at the high temperatures we work at is a real challenge’

Jérémie Sobel, chemist at Thorizon

As in all nuclear reactors, this one too is driven by a nuclear chain reaction. An atomic nucleus, such as uranium, splits apart, releasing heat and a few neutrons. These can collide with other nuclei, causing them to split in turn, and in this way the chain reaction is sustained. The heat released in the process enters the circulating salt, and is converted into usable energy with the help of heat exchangers.

Various fuels

One advantage of molten salt reactors is that they can run on a wide variety of nuclear fuels: not only uranium, but also the long-lived nuclear waste from the current generation of nuclear reactors. The nuclear fission of uranium leaves behind fission products — fragments of split nuclei — which typically remain radioactive for a few hundred years. However, actinides are also produced: nuclei that have captured a neutron but have not split apart, and which are often radioactive for tens of thousands of years. Such long-lived isotopes, which must be stored safely for extended periods, can largely be ‘burned up’ in a molten salt reactor. Another option is the burning of the element thorium. The only naturally occurring isotope, thorium-232, is not fissile. But when a thorium nucleus captures a neutron, it can be converted into uranium-233, which is fissile. ‘The thorium cycle is attractive, but technically complicated and lies further in the future’, says De Groot. ‘Hence the company name.’

Leaks and corrosion

With a high operating temperature of 600 to 700 degrees, energy production is more efficient than that of light-water reactors, which operate with water at around 300 degrees and therefore under high pressure. Furthermore, because a molten salt reactor operates at low pressure, pressure-resistant circuits are not required. Even so, there are still some challenges to overcome. ‘Leaks and corrosion are our biggest challenges’, says Sobel. ‘Molten salt is just like water, and achieving proper seals at the high temperatures we work at is a real challenge.’ Seals between pipes that remain tight even at 600 degrees Celsius are particularly difficult to make. Sobel: ‘Salt itself is not corrosive, but because the temperature is so high, it acts as a catalyst for almost all reactions. So any impurity reacts easily with your structural material.’ Because salt attracts moisture, that impurity is often moisture itself, which makes keeping the reactor dry absolutely essential.

‘The thorium cycle is attractive, but technically complicated and lies further in the future’

Sander de Groot, co-founder of Thorizon

Cartridges

The composition of the salt changes during the reaction. As the reaction continues, fission products, corrosion products and actinides accumulate in the salt. Because these eventually poison the reaction, they must be removed from the salt. However, the primary limiting factor is the service life of the materials used. For this problem, Thorizon has developed its own solution: the salt circulates within multiple sealed cartridges — narrow modules approximately five metres tall that are interchangeable. Multiple cartridges are placed together, allowing the fuel to reach the critical density needed to initiate a nuclear reaction. ‘The neutrons fly straight through the cartridge walls’, explains De Groot. A meltdown, such as the one at Chernobyl, is inconceivable: if the temperature rises, the reaction shuts itself down.

Individual cartridges last five to ten years, depending on the service life of the materials used, and can be replaced with ease. ‘You remove them from the reactor, allow them to cool, and then they are transported to a reprocessing facility.’ There, the salt, fuel and waste are separated and purified, and new cartridges are filled and sent back.’

In this way, the reactor itself remains relatively simple, and it is possible, for instance, to swap in cartridges with different fuels, or to introduce new versions with a longer service life. ‘We will have the basic components of a complete cartridge by 2027’, says De Groot. A demonstration reactor could then be built in 2030. A few years after that, construction of Thorizon One would begin — a larger reactor with a capacity of 250 megawatts. ‘Yes, that is quite soon’, admits De Groot, ‘but we believe the timeline is realistic, if everything goes to plan.’