Big tech firms in the US are pouring billions into advanced nuclear power to feed energy-hungry artificial intelligence. Across the Atlantic, European scientists are tackling the technology’s biggest engineering headache: corrosive liquid salt.Dutch startup Thorizon has teamed up with research institute TNO. Together, they have launched an experimental materials program. The goal is straightforward. They need to find metals that can survive inside molten salt reactors without failing. Early testing has already begun in Petten and Amsterdam.Increasing demand for molten salt reactorsMolten salt reactors run at near-atmospheric pressure and rely on passive safety mechanisms if power fails. They can also be configured to produce far less long-lived waste than conventional US commercial reactors. These traits make them attractive for data centres and heavy industry needing round-the-clock power.Yet their internal environment is hostile. Liquid fuel salts operate at intense heat, often topping 1,000 degrees Fahrenheit. Over time, this mixture corrodes structural metals while radiation weakens them from within.Thorizon plans to work around this problem rather than fight it indefinitely. Standard reactors rely on permanent vessels built to last half a century. Thorizon takes a modular route. It seals the liquid salt inside self-contained cartridges. Operators will simply pull out the old units and slot in fresh ones every five to ten years.“Our cartridge design makes material degradation a manageable engineering challenge,” said Kiki Lauwers, chief executive of Thorizon. She noted that the new trials will give regulators and operators hard data on cartridge lifespans.“This programme will provide the independent evidence needed to validate that design and give regulators and operators confidence in material performance and cartridge lifetime,” added Lauwers.Running stress trialsSafety agencies require strict experimental proof before licensing advanced reactor designs. TNO will run stress trials, simulate corrosion rates, and track failure points to show how these alloys hold up over time.“These innovations can only move from concept to deployment on the basis of strong scientific evidence,” remarked Tjark Tjin-A-Tsoi, chief executive of TNO.The findings will support safety licensing for Thorizon’s PIONEER test reactor in Petten. Later, they will underpin Thorizon One, the company’s planned commercial model.The two Dutch groups are expanding the work across borders. They are bringing in partners such as Delft University of Technology, NRG PALLAS, the French atomic agency CEA, and the European Commission’s Joint Research Centre.Need for steady baseload energyThe initiative arrives as power grids face unprecedented demand. In both North America and Europe, clean electrification and digital infrastructure are outstripping renewable additions like wind and solar. Modern grids still require steady baseload energy that runs day and night.To help close that gap, the Netherlands aims to build 7 gigawatts of nuclear capacity by 2050. Small modular reactors are central to that plan. Officials want compact plants that hook directly into factories and server farms.If the Petten trials succeed, Thorizon’s swappable core could offer a clear path forward. Proving the materials work may finally turn molten salt from a lab promise into a commercial reality.