Vienna and Beijing Build the First Nuclear Clocks

Time is the most precisely measured quantity in physics, and the race to build better clocks has just taken an unusual turn. Two independent teams, one at TU Wien in Vienna and one in Beijing, report the first working clocks that keep time with the nucleus of an atom instead of its electrons. The Austrian newspaper DER STANDARD framed it as a race between the two cities.

This post explains how atomic clocks work, what the Vienna team built, and where the newspaper’s account deserves a closer look. It draws on Klaus Taschwer’s article of 7 October 2026 and on the research papers, which I read as arXiv preprints. The published Nature versions may differ in detail.

How atomic clocks tick

Every clock needs something that swings at a steady rhythm and a counter that tallies the swings. A pendulum does this, and so does the quartz crystal in a wristwatch. Atoms offer a far better rhythm, because all atoms of one element are identical. When an electron jumps between two energy levels, the atom absorbs or emits light at one exact frequency. The official definition of the second still rests on such a jump in caesium atoms. Modern optical clocks use visible light instead of microwaves, which makes the rhythm faster and the measurement finer. The key is a feedback loop: a laser is tuned close to the atomic frequency, the atoms react most strongly when it is exactly right, and any drift is corrected automatically.

Swapping electrons for nuclei

In 2003, Peik and Tamm proposed using the atomic nucleus as the pendulum instead. A nucleus is more than ten thousand times smaller than the electron shell around it, so stray fields disturb it far less. There was a catch: almost every nucleus needs gamma rays to change its energy state, and no laser can deliver those. Thorium-229 is the exception. Its excited state sits only 8.4 electron volts above the ground state, reachable with a laser at 148 nanometres. Laser excitation of this transition was first achieved in 2024. The Vienna team embeds thorium in a millimetre-sized calcium fluoride crystal that works at room temperature.

What Vienna achieved

In the new paper, Luca Toscani De Col, Thorsten Schumm and colleagues from TU Wien, PTB Braunschweig and other institutes describe what they call the first nuclear clock that runs as a stand-alone device. A laser stabilised by an optical cavity provides short-term steadiness. The thorium nuclei then correct its slow drift: they absorb light best at exactly the right frequency, and the system adjusts the laser whenever the absorption changes. The signal is tiny, with only 65 picowatts of light reaching the detector and under one percent of it absorbed. Even so, the clock ran for about 23 hours in April 2026 while being compared with a ytterbium ion clock at the Austrian federal office BEV. The data also allowed a search for dark matter. The limits compete with the best atomic clocks for one type of coupling and reach further for couplings to the strong force and quarks. The Beijing team, using a much stronger laser, reports a working clock too. Two of its crystals agree on the clock frequency to about three parts in ten trillion.

Promise versus reality

Neither clock beats today’s best atomic clocks. DER STANDARD notes that a lutetium clock in Singapore holds the record at about one part in ten quintillion, and the Vienna authors project that nuclear clocks could eventually reach the level of the best optical clocks. One figure in the article needs care. It gives Vienna’s accuracy as roughly 1×10⁻¹⁵, or one second in 30 million years. The preprint describes that number as stability, meaning how smoothly the clock ticks when averaged over time. It is not accuracy, meaning how closely the clock matches the true frequency. The paper itself reports that results on different days differ by about 5×10⁻¹³, roughly one second in 60,000 years, which the authors trace to strain inside the crystal. Even the stability value is stated differently in different parts of the preprint, so the Nature version is worth checking.

Whether the nuclear clock becomes the most precise clock in the world is still open. What is already clear is the change in purpose. The better clocks get, the more they act as sensors for gravity, dark matter and possible changes in the constants of nature.

Sources: DER STANDARD, K. Taschwer, 7 Oct 2026. Toscani De Col et al., A thorium-229 optical nuclear clock with feedback loop (arXiv:2606.04997). Huang et al., A nuclear clock based on 229Th (arXiv:2606.08870).

Photo: Merlin Lightpainting via Pexels