Thorium‑229 Nuclear Clocks Demonstrated in Vienna and Beijing; Vienna Shows First Self‑Stabilizing Prototype

Scientists in Vienna have demonstrated the first self-stabilizing nuclear clock, a milestone that moves the long-discussed idea of nuclear timekeeping into the realm of working prototypes.

The result, published Oct. 7 in Nature, came from researchers at TU Wien working with Germany’s Physikalisch-Technische Bundesanstalt, or PTB, the national metrology institute in Braunschweig. In the same week, a separate team in Beijing published its own operational thorium-229 nuclear clock in Nature, reinforcing that the field has advanced beyond theory.

The Vienna paper, “A thorium-229 optical nuclear clock with feedback loop,” reported something researchers have been chasing for years: a clock in which the thorium-229 nuclear resonance itself provided the feedback signal that kept the laser on frequency, without continuous referencing to a conventional atomic clock. “What you really want is a self-stabilizing nuclear clock,” Thorsten Schumm, the TU Wien group leader, said in a university press release.

A nuclear clock differs from a conventional atomic clock in where it gets its timing signal. Today’s atomic clocks use transitions in an atom’s electrons. These new systems use a transition in the nucleus of thorium-229, an isotope with an unusually low-energy nuclear state that can be addressed with vacuum-ultraviolet light at about 148 nanometers. Physicists see promise in that approach because the nucleus is much smaller and more shielded from environmental disturbances than the electron cloud, which could eventually make nuclear clocks more robust.

Both the Vienna and Beijing teams used thorium-doped calcium fluoride crystals as a solid host for the thorium nuclei. That makes these devices solid-state nuclear clocks rather than the atom-trapping systems often associated with the most advanced laboratory clocks.

In Vienna, first author Luca Toscani De Col and colleagues reported continuous autonomous operation for more than 24 hours. The team measured a fractional frequency instability of 3 × 10⁻¹² divided by the square root of averaging time in seconds, improving to about 10⁻¹⁵ over a day of operation. TU Wien and ScienceDaily described that level as roughly equivalent to losing or gaining one second over 30 million years.

The Beijing group, led by first author Beichen Huang and senior author Shiqian Ding of Tsinghua University and related institutions, reported a fractional frequency instability of 5 × 10⁻¹³ divided by the square root of averaging time. The team also said the clock-transition frequencies measured in two independently fabricated crystals agreed at the 10⁻¹³ level, an important sign that the approach can be reproduced across separate samples.

Still, the headline result is not that nuclear clocks have overtaken the best existing clocks. They have not. The world’s top optical atomic clocks operate at about the 10⁻¹⁸ to 10⁻¹⁹ level, several orders of magnitude better than these first nuclear-clock prototypes. TU Wien said so directly in its press material: “This is not yet at the level of the world's best optical atomic clocks, but for a first prototype it is a fantastic result.”

That caveat matters because nuclear clocks have often been discussed in sweeping terms. What the two Nature papers show is something more concrete and more useful: thorium-229 can now support operational clock systems in the lab, including one that can keep itself on frequency for more than a day.

The new demonstrations build on a crucial earlier step. On April 29, 2024, researchers reported direct laser excitation of the thorium-229 nucleus, helping make the 2026 clock experiments possible.

More accurate and stable clocks underpin technologies such as GPS and other satellite navigation systems, telecom synchronization and geodesy, the science of measuring Earth’s shape and gravitational field. Vienna’s team also used its clock data to look for periodic fluctuations and slow drifts in the nuclear transition energy, setting limits on some models of ultralight dark matter. For now, though, the main news is simpler: nuclear clocks are no longer just a theoretical ambition. They are working devices, even if they are still early ones.