The First Working Nuclear Clock Marks a New Era in Timekeeping
June 13, 2026
For decades, the most precise clocks on Earth have been atomic clocks, devices that lock their rhythm to the quantum dance of electrons orbiting an atom. Cesium fountain clocks define the second itself. Optical lattice clocks using strontium or ytterbium now reach uncertainties below one part in 1018, losing less than a second over the age of the universe.
But a different kind of clock has long beckoned. What if, instead of tuning to an electron transition, you locked to the nucleus itself? The nucleus is thousands of times smaller and far more isolated from external fields than its orbiting electrons. A nuclear clock would be inherently more stable, more compact, and sensitive to forces that electrons simply do not feel.
It has been a tantalizing vision. And as of June 3, 2026, it is no longer a vision at all. A team led by Luca Toscani De Col and Thorsten Schumm at TU Wien, together with Ekkehard Peik at the Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig, has built the first working nuclear clock. Their paper, posted on arXiv as 2606.04997, reports the first time a laser has been actively stabilized to and steered by a nuclear transition in real time. This is not a proof of principle or a single-shot excitation. It is a fully operational feedback loop, a genuine clock.
Why Thorium-229?
The reason this breakthrough centers on thorium-229 is deeply rooted in nuclear physics. Most nuclear transitions require energies in the gamma-ray range (tens of thousands to millions of electronvolts), far beyond what any laser can reach. But thorium-229 is exceptional. Its nucleus possesses an isomeric state only about 8.3 electronvolts above the ground state. That energy corresponds to a vacuum ultraviolet (VUV) wavelength of 148.4 nanometers, a regime accessible to modern laser technology.
This “low-energy isomer” was first hinted at in the 1970s and confirmed over decades of painstaking measurements. It is effectively the only nuclear transition in existence that can be driven by a laser. That single fact makes thorium-229 the foundation upon which all nuclear clock research is built.
Previous milestones include the first direct excitation of the thorium nuclear transition by a JILA/NIST team, published in Nature in September 2024, and improved reproducibility demonstrated in a January 2026 Nature paper. Those experiments showed that the transition could be driven. What they did not yet achieve was a closed feedback loop: a clock that runs continuously, correcting its laser in real time.
That is precisely what the Vienna/Braunschweig team has now delivered.
How the Clock Works
The design is elegant. Thorium-229 nuclei are doped into a calcium fluoride (CaF2) crystal, which operates at room temperature. A continuous-wave VUV laser tuned to 148.4 nm illuminates the crystal. When the laser frequency matches the nuclear transition, thorium nuclei absorb photons and jump to the isomeric state. A photomultiplier tube (PMT) monitors the crystal’s fluorescence, providing a real-time readout of how many nuclei are being excited.
The key innovation is the feedback loop. The team modulates the laser frequency slightly above and below the estimated resonance. By comparing the fluorescence signal on either side, they generate an error signal that tells the control system exactly how far the laser has drifted. The system then corrects the laser frequency, keeping it locked to the nuclear resonance at all times.
This is the same principle used in any atomic clock: lock an oscillator to a quantum reference, measure the error, correct it, repeat. The difference is that the quantum reference here is a nuclear transition, not an electronic one. The laser’s frequency is then compared to a secondary standard: an ytterbium ion (Yb+) optical clock at PTB, one of the best timekeepers in the world.
The results speak clearly. The team measured the transition frequency at 2.0204 x 1015 Hz with a linewidth of approximately 100 kHz. The clock instability is 3 x 10-12 divided by the square root of the averaging time in seconds, approaching 10-15 after one full day of operation.
The Power of a Nuclear Reference
Why go through the enormous difficulty of building a nuclear clock when atomic clocks already perform so brilliantly? The answer lies in what the nucleus offers that electrons cannot.
The key property is isolation. Electron orbitals are sensitive to electric and magnetic fields from nearby atoms, from stray charges, from the environment. Clock designers spend immense effort shielding atomic clocks from such perturbations. The nucleus, by contrast, is shielded by its surrounding electron cloud. It barely notices external electric fields. Magnetic sensitivity exists but is many orders of magnitude smaller than for electrons. A nuclear clock should, in principle, be far more robust against environmental noise.
This robustness has a direct practical consequence: compactness. Optical lattice clocks require ultracold atoms, vacuum chambers, and elaborate laser cooling setups. They fill rooms. A nuclear clock, based on a solid-state crystal at room temperature, could eventually be shrunk to a portable device. Imagine a frequency standard that fits in a shipping container, or even a backpack, with stability rivaling laboratory-scale atomic clocks. That prospect has enormous implications for GPS-independent navigation, geodesy, and field-deployed scientific instruments.
A New Window on Fundamental Physics
Beyond timekeeping, the nuclear clock opens a new experimental window on the fundamental laws of physics. Because the nuclear transition is governed by the strong force rather than the electromagnetic force (which governs electron transitions), it is sensitive to physics that atomic clocks cannot easily probe.
One immediate triumph has already been reported. The same team used the nuclear clock to set the best laboratory limits to date on variations of the strong force coupling constant. Certain dark matter candidates, such as topological defects or ultralight bosons, would cause subtle oscillations in fundamental constants as they pass through Earth. A nuclear clock, with its unique sensitivity to the strong force, is an ideal detector for such effects.
The clock also enables searches for temporal variation of fundamental constants over days, months, and years. The ratio of nuclear transition frequency to electronic transition frequency is a sensitive probe of whether the fine-structure constant or the quark mass has changed over time. Even a null result is valuable, tightening constraints on theories of physics beyond the Standard Model.
Where It Stands Today
It is important to be precise about what this clock is and is not, yet. The current instability, around 10-15 at one day, is roughly a factor of 1,000 worse than the best optical atomic clocks, which reach the low 10-18 range. The limitations come primarily from crystal quality and laser power. The CaF2 crystal host introduces some inhomogeneous broadening, and the VUV laser power is limited, constraining the signal-to-noise ratio of the feedback loop.
These are engineering challenges, not fundamental ones. The thorium transition itself is capable of far narrower linewidths. As crystal growth improves and laser technology advances, the nuclear clock should improve rapidly. Many researchers expect it to surpass atomic clock performance within a decade.
A Historic Moment
The significance of this achievement should not be understated. Since the first atomic clock was built in 1949, every major advance in timekeeping has relied on electronic transitions. The idea of using the nucleus was always there, but it seemed perpetually out of reach. The thorium isomer was known, but the laser technology to drive it did not exist. When the laser became available, the challenge shifted to stabilizing it to the nuclear resonance with sufficient precision.
That challenge has now been met. For the first time, a human-made device counts time by the heartbeat of an atomic nucleus. The implications for metrology, fundamental physics, and technology are only beginning to unfold. The nuclear clock era has begun, and it promises to be as transformative as the atomic clock era that preceded it.
Reference: L. Toscani De Col et al., “A thorium-229 optical nuclear clock with feedback loop,” arXiv:2606.04997 (2026). See also: JILA/NIST, Nature (September 2024); JILA/NIST, Nature (January 2026).