Today’s best atomic clocks, which track time by measuring the way electrons consistently jump between energy states when exposed to particular laser light, would lose less than one second of accuracy over the entire age of the universe so far. They are incredibly accurate, but also incredibly delicate: an entire lab’s worth of vacuum chambers and shielding is required to prevent interference from stray electric or magnetic fields. To fix these issues scientists aim to go one layer deeper into the atom, from the electrons on the outside to the nucleus nestled within, and track how energy flips a single neutron between quantum states. According to Thorsten Schumm, a physicist at the Vienna University of Technology, such a switch could make the clocks up to 10 times more precise, and the atom itself would shield the clock from some outside interference. Plus, such atoms could be embedded protectively in a crystal.

In a new study in Science, Schumm and his colleagues have mapped the four places a thorium atom can sit inside a crystal and found one close to ideal for building the most precise clock ever.

In most atoms, the energy needed to flip a neutron in this way would require a gamma-ray laser, which scientists can’t make yet. There’s just one known exception, an isotope called thorium-229, which researchers showed in 2024 can be triggered with an ultraviolet laser instead.

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

“Thorium-229 is a quirk of nature. Its nuclear transition energy is remarkably low, making it accessible to high-purity laser light,” explains Andrei Derevianko, a physicist at the University of Nevada, Reno, who was not involved in the study. Researchers sought a crystal structure that could house and protect thorium-229 atoms, removing the need for intensive vacuum chambers for the clock to function.

“Crystal growing is like alchemy, or like cooking,” Schumm says. “You try, and then you try something else, and then you keep trying.” After 15 years of experiments, Schumm’s group ended up with a recipe for solid calcium fluoride crystals, doped with little impurities of thorium-229, that looked like transparent millimeter-sized cubes. “In the end it looks like a tiny little piece of glass, but there’s tens of thousands of hours of learning how to do that,” he adds. The researchers realized, however, that if they targeted thorium-229 in a bad internal position, the crystal would have an uneven electric field that would harm the clock’s timing.

To investigate the possible ways thorium atoms can be positioned in crystals, for this latest study, the scientists shined a custom-built ultraviolet laser at one wavelength for 60 seconds, switched it off and watched for five minutes as excited nuclei decayed back down to lower energy levels and emitted the light back out. Then the researchers altered the frequency and did it again. It turned out thorium atoms can sit at four sites in the crystal lattice. Three return light at multiple wavelengths, which indicate an uneven electric field around them. The fourth responds to just one wavelength, a sign of an even field that made it perfect for the clock.

“The data in this paper is something we’ve all wanted to see for some time,” says Eric Hudson, a physicist at the University of California Los Angeles who was not involved in the study. “This is a very important result for the research on the solid-state nuclear clock,” adds Ekkehard Peik, head of the Time and Frequency Department at the National Metrology Institute in Germany, who was also not involved in the study. Based on their discovery, Schumm and his colleagues have already built first early prototypes of new clocks, and have patented a route to building one that fits on a small chip. At the same time another team in China led by Shiqian Ding, a physicist at Tsinghua University, has also demonstrated a prototype nuclear clock that achieved comparable results with a more powerful laser aimed at a crystal containing a slightly lower concentration of thorium-229. Schumm thinks the progress in the field is gaining momentum.

“I’m not going to give you a time scale on when we will beat the best atomic clocks,” Schumm says, “but we will be better [than our current prototypes] by at least three orders of magnitude by the end of the year.” For now, though, his team isn’t chasing superior precision. The priority, they say, is miniaturizing the clocks—first going from lab-sized to shoebox-sized, Schumm says, a form that could be used in a data center server rack to keep things like banking transactions going when communications might otherwise be lost with today’s satellite-based clocks.