Teams in Vienna and Beijing build first working nuclear clocks using thorium-229

A new mechanism for timekeeping

Atomic clocks have served as global time standards for more than 70 years, tracking the movement of electrons between energy states to coordinate satellite navigation and telecommunication networks. Two independent research teams, one led by the Vienna University of Technology (TU Wien) and Germany's Physikalisch-Technische Bundesanstalt (PTB), and another at Tsinghua University in Beijing, have built the first operating nuclear clocks. Published in Nature on 7 October 2026, the two systems measure time by tracking transitions within the atomic nucleus of thorium-229 rather than outer electron shells. The concept originated in a 2003 proposal by Ekkehard Peik at the PTB, who identified thorium-229 as the only known isotope with a nuclear transition accessible by laser light. Because atomic nuclei are roughly 10,000 to 100,000 times smaller than whole atoms and bound by strong nuclear forces, they remain shielded from external electric fields, magnetic interference, and temperature variations.

Self-stabilizing laser prototypes

Converting the nuclear transition into a standalone timepiece required resolving laser stability and automated frequency correction. In April 2024, researchers at TU Wien and PTB observed the laser-driven thorium transition, and later that autumn they coupled the excitation apparatus to a conventional optical atomic clock as a stopwatch. The latest breakthrough establishes an autonomous system that uses the thorium-229 crystal to self-stabilize the laser without external atomic clocks. An internal feedback loop continuously monitors the laser and corrects frequency deviations on a timescale of seconds. Senior author Thorsten Schumm described the engineering requirement for autonomous operation:

The secret was getting the feedback to work on a scale of just a few seconds.

The Austrian prototype maintained stability for more than 24 hours without human intervention. Simultaneously, the Chinese team led by Shiqian Ding demonstrated that two separate nuclear clocks built with different crystal lattices produced matching resonant frequencies.

Precision targets and laser challenges

The initial prototypes achieve a precision level equivalent to losing one second every 30 million years, matching standard atomic clocks used in satellite navigation systems. They remain behind top laboratory optical atomic clocks, including a lutetium clock at the University of Singapore that loses one second in 55 billion years. The main technical hurdle lies in the generation of vacuum ultraviolet light, which remains inefficient and delivers relatively weak laser power to the thorium crystal. Schumm emphasized that early prototypes focus on establishing feasibility:

It is completely normal that a new type of clock does not immediately surpass all existing technologies; this is a first proof of concept, and what is relevant is the potential for future improvements.

Researchers project that incorporating more powerful lasers will eventually allow nuclear clocks to reach an accuracy of one second in 550 billion years, ten times more precise than existing atomic instruments.

Probing fundamental physical constants

Beyond advancing metrology, nuclear clocks provide experimental platforms to investigate fundamental physics and search for dark matter, which accounts for roughly 83% of the universe's matter. If passing dark matter particles interact with the thorium nucleus, they could induce measurable frequency perturbations in the clock's regular cycle. Daniel de Mercado, a metrology physicist at the Spanish Metrology Centre developing calcium-40 optical clocks, noted the broader experimental scope of precise frequency measurements:

The frequencies of atomic and nuclear transitions depend on the constants and fundamental forces of nature, so an extremely precise measurement makes it possible to look for small discrepancies between the observed value and what our theories predict.

Physicists plan to use these measurements to test whether fundamental physical constants remain invariant over time or drift under unknown fundamental interactions.