Nuclear Clocks: Scientists Have Built a Clock That Measures Time Inside an Atomic Nucleus

The best optical atomic clocks can measure frequency differences so accurately that they can detect effects associated with differences in gravitational potential across surprisingly small height changes.

Instead of using the energy transitions of electrons surrounding an atom, a nuclear clock uses a transition between quantum states inside an atomic nucleus.

On October 7, 2026, two independent research teams published papers in Nature demonstrating operational nuclear clocks based on thorium-229. One team, led by researchers associated with TU Wien and Germany's national metrology institute PTB, stabilized a laser to a nuclear transition in a thorium-doped crystal. A second team, led by researchers in China, independently demonstrated a nuclear clock using a related thorium-229 system.

The studies are:

• "A thorium-229 optical nuclear clock with feedback loop", published in Nature on October 7, 2026.
• "A nuclear clock synchronized to ²²⁹Th", also published in Nature on October 7, 2026.

These were not simply proposals for a future instrument. Both teams demonstrated systems in which the nuclear transition served as a real-time frequency reference.

But there is an important qualification: these first operational nuclear clocks do not yet outperform the world's best optical atomic clocks in overall performance.

Their significance lies in establishing a new working platform that may eventually combine precision measurement with a more compact and potentially more robust physical design.

The breakthrough raises several questions. Why would a transition inside a nucleus make a better clock? What exactly did the experiments demonstrate? And how could a clock help scientists investigate dark matter or test the laws of physics?

To answer those questions, we need to look at what a clock actually measures.

A clock is fundamentally a system that produces a regular physical signal.

A mechanical clock uses oscillations of a balance wheel or pendulum. A quartz clock relies on the vibrations of a crystal. Atomic clocks use carefully selected transitions between atomic energy levels to stabilize an electromagnetic oscillator.

The more stable and reproducible the reference frequency, the better scientists can measure elapsed time and compare frequencies.

A simplified relationship between frequency and period is:

Here, f is the frequency, measured in cycles per second, and T is the period of one cycle.

For an optical clock, the reference oscillation is associated with light at an extremely high frequency. The clock compares a laser's frequency with a well-defined transition in an atom or ion and uses feedback to correct deviations.

The transition acts like a reference against which the laser can be disciplined.

If the laser begins to drift, the measured response indicates the direction and size of the error. A control system adjusts the laser to keep it aligned with the reference.

The resulting stabilized frequency can be used for precision timing, comparisons between clocks, and tests of physical theories.

The challenge is finding a reference transition that is sufficiently narrow, stable, reproducible, and insensitive to unwanted environmental influences.

That is where thorium-229 becomes interesting.

Most familiar atomic clocks use transitions involving electrons. Nuclear clocks instead exploit a transition between two energy states of the nucleus itself.

The isotope thorium-229 has a remarkably low-energy excited nuclear state. Its energy is approximately 8.4 electronvolts above the nuclear ground state, corresponding to vacuum-ultraviolet light with a wavelength near 148 nanometers.

The relationship between photon energy and wavelength is:

where E is photon energy, h is Planck's constant, c is the speed of light, and λ is wavelength.

For most nuclei, transitions occur at much higher energies, commonly requiring X-rays or gamma rays. Those transitions are difficult to address using the narrow-band, tunable lasers needed for an optical clock.

Thorium-229 is unusual because its low-energy nuclear transition can be addressed with specially engineered vacuum-ultraviolet laser technology.

This makes it a practical candidate for a nuclear clock.

Why use a nucleus instead of an electron?

An atomic nucleus is much smaller than the electron cloud surrounding it. Some nuclear transitions can therefore be less sensitive to certain external electric and magnetic disturbances that affect electronic transitions.

That does not mean a nuclear clock is immune to its environment. The surrounding crystal, magnetic fields, temperature, and other interactions can still affect the observed transition frequency.

Nevertheless, the different physical structure of the reference transition creates an opportunity to develop clocks with distinct systematic errors and potentially improved robustness.

A nuclear clock could also provide a new way to test fundamental physics because its transition depends on nuclear structure and the forces governing nucleons, rather than only on electronic energy levels.

The important word is potentially. The advantages must be demonstrated experimentally, and each clock design has its own limitations.

The idea of a thorium-229 nuclear clock is not new.

Researchers proposed the concept in 2003, motivated by the unusually low energy of the isotope's excited nuclear state. For years, the challenge was identifying the transition precisely enough to excite it with the right laser frequency.

A clock requires more than evidence that a transition exists. Researchers must be able to locate the resonance, measure it, and use it to control a frequency reference.

In 2024, experiments established important milestones by directly exciting the thorium-229 nuclear transition with laser light and measuring its frequency relative to an established atomic-clock reference.

Those results helped establish thorium-229 as a viable candidate for nuclear timekeeping.

The next challenge was operational control.

A clock must continuously monitor its reference and correct its oscillator when it drifts. This requires a measurable signal that indicates whether the laser is above or below the desired transition frequency.

The October 2026 studies achieved that next step by using absorption spectroscopy and feedback control.

Instead of merely observing a nuclear resonance, the researchers used it to stabilize a laser in operation.

That is the essential distinction between measuring a transition and demonstrating a working clock.

The European team, associated with TU Wien and PTB, used thorium-229 nuclei embedded in a calcium fluoride crystal.

The crystal was approximately millimeter-sized and operated at room temperature. A narrow-linewidth continuous-wave laser at 148 nanometers was stabilized to the nuclear transition.

The experiment used absorption spectroscopy to determine how the laser interacted with the thorium nuclei. By observing the absorption signal, the researchers obtained information about the laser's detuning from the transition.

A feedback loop then adjusted the laser frequency to keep it aligned with the nuclear reference.

This closed-loop control is what made the system an operational clock rather than a one-time spectroscopic measurement.

What did the team demonstrate?

The researchers demonstrated laser stabilization to the thorium-229 nuclear transition and characterized the clock's frequency instability.

Their work showed that a nuclear reference embedded in a solid-state crystal can be used to control an optical oscillator in real time.

The crystal's solid-state structure offers a potential advantage over systems that require individual atoms to be cooled and trapped in elaborate vacuum arrangements.

However, the clock still required specialized laser technology, optical components, and experimental controls. It is not yet a small consumer device or a drop-in replacement for a commercial timing system.

The experiment also demonstrated that a nuclear clock can be used for fundamental-physics measurements. The team compared the nuclear clock against a remote ytterbium-ion clock through a stabilized optical-fiber link to search for possible time-dependent changes associated with ultralight dark matter.

That search did not reveal the predicted variation over the timescales studied. A null result, however, can still constrain theories by ruling out some combinations of possible effects.

A second team independently demonstrated a nuclear clock using thorium-229 embedded in calcium fluoride crystals.

The researchers stabilized a continuous-wave laser at approximately 148.4 nanometers to the nuclear transition. They generated vacuum-ultraviolet light using four-wave mixing in cadmium vapor and used frequency-modulation absorption spectroscopy to monitor the resonance.

The team reported a fractional frequency instability scaling approximately as:

Here, σy(τ) represents fractional frequency instability as a function of averaging time τ, expressed in seconds.

This equation describes how the clock's short-term instability decreases as more measurements are averaged. It is not, by itself, a statement of absolute accuracy.

The researchers also compared transition frequencies measured in two independently fabricated crystals. Their results agreed at approximately the 10⁻¹³ fractional level and were consistent with previous measurements.

That reproducibility matters because a clock is useful only if its reference can be characterized reliably across different samples and experimental conditions.

The study established a solid-state platform for nuclear clock operation and future tests of fundamental physics.

This is where headlines can be misleading.

The October 2026 demonstrations are major milestones, but the nuclear clocks do not yet surpass the best optical atomic clocks in overall performance.

According to the American Physical Society's analysis of the two experiments, the reported instabilities correspond to illustrative accumulated timing errors of roughly one second over three million years for the European system and one second over nineteen million years for the Chinese system.

These are simplified extrapolations of measured instability — not demonstrations that either clock was operated for millions of years, and not direct measurements of total long-term timekeeping accuracy.

The best optical atomic clocks can reach performance corresponding to gaining or losing approximately one second over tens of billions of years under suitable comparisons.

That means the first nuclear clocks are not yet the winners in the precision race.

Why, then, are they important?

Because precision is only one property of a useful clock.

Scientists also care about stability, systematic uncertainty, environmental sensitivity, physical size, power requirements, reproducibility, and the ability to operate outside highly specialized laboratory conditions.

A new clock architecture can be scientifically valuable even before it outperforms existing systems.

It may offer different sources of error, new experimental capabilities, or a path toward a future instrument with advantages that current clocks cannot easily provide.

The 2026 results establish that nuclear clock operation is possible. Improving the performance enough to outperform the best optical clocks remains a research challenge.

One of the most intriguing potential applications is the search for dark matter.

Dark matter is inferred from its gravitational effects on galaxies, galaxy clusters, and the large-scale structure of the universe. Yet scientists still do not know what it is made of.

Some theoretical models propose that dark matter could include ultralight fields that vary slowly over time. If such a field interacts with ordinary matter, it could produce tiny variations in fundamental constants or in the energies of atomic and nuclear transitions.

A clock is an attractive instrument for searching for these effects because it measures frequency with extraordinary precision.

Suppose two different clocks respond differently to a hypothetical change in a fundamental constant. Researchers can compare their frequencies over time.

If one clock changes relative to the other in a pattern predicted by a particular model, that could provide evidence for new physics.

A simplified comparison is:

Here, R(t) is the ratio between the nuclear-clock frequency and an atomic-clock frequency.

The ratio is useful because some disturbances may affect both clocks similarly, while a new interaction could affect the nuclear and electronic transitions differently.

A changing ratio would not automatically prove dark matter. Researchers would need to rule out environmental effects, systematic errors, and other explanations. They would also need to check whether the observed signal matches the predicted behavior of a particular model.

The Vienna team used its nuclear clock in an initial search for ultralight-dark-matter effects. It found no variation over the timescales studied, but the measurement placed constraints on some proposed interactions.

That is a meaningful result: a clock can test a theory even when it finds no evidence for the predicted signal.

Nuclear clocks may eventually become particularly useful in this area because nuclear transitions respond to aspects of physics that are not probed in exactly the same way by conventional atomic clocks.

Some of the most fundamental quantities in physics are dimensionless constants, such as the fine-structure constant, which characterizes the strength of electromagnetic interactions.

Modern physics generally treats these constants as fixed in space and time. But some theories beyond the Standard Model allow small variations under particular conditions.

Precision clocks offer a way to test these possibilities.

Different atomic transitions respond differently to changes in fundamental constants. A nuclear transition provides a distinct sensitivity pattern because its energy depends on nuclear forces and structure.

By comparing clocks based on different transitions, researchers can test whether their frequency ratios remain constant.

This approach is especially valuable because it allows scientists to investigate new physics without necessarily needing a particle accelerator capable of directly producing a hypothetical particle.

Instead, they search for tiny deviations from expected behavior in extremely precise measurements.

However, the interpretation must remain careful. A measured frequency shift can have ordinary causes, including temperature changes, electromagnetic fields, or other environmental influences.

The scientific challenge is to separate those effects from any genuine change in fundamental physics.

The first nuclear clocks provide a new reference for conducting such tests. They do not yet demonstrate that fundamental constants vary or that dark matter has been detected.

A long-term goal is to create nuclear clocks that are not only precise but also compact and robust.

The 2026 experiments embedded thorium-229 nuclei in calcium fluoride crystals. This offers a different architecture from conventional optical clocks based on individual atoms or ions held in elaborate trapping systems.

A solid-state system can host many nuclei in one sample, potentially increasing the number of usable reference systems. It may also simplify some aspects of mechanical construction and environmental control.

But a crystal introduces its own challenges.

The local environment around each thorium nucleus can vary slightly, causing differences in the observed transition frequencies. Temperature and crystal structure can affect the resonance. Laser generation at approximately 148 nanometers also remains technically demanding.

Researchers will need to improve the stability of the reference, characterize systematic shifts, and develop more practical laser and optical systems.

If those challenges can be addressed, nuclear clocks could eventually find uses in:

• Precision timing and frequency comparisons.
• Tests of fundamental constants.
• Searches for ultralight dark matter.
• Portable scientific instruments.
• Specialized navigation or synchronization systems.

These are potential applications, not proven commercial capabilities of the first-generation devices.

The best path forward may involve combining the strengths of different designs. The European team benefited from a relatively high concentration of thorium nuclei, while the Chinese team used a more powerful vacuum-ultraviolet laser source. The researchers have identified the possibility that combining these strengths could improve future performance.

That is one reason the independent demonstrations are important: they establish multiple experimental approaches that can now be compared and refined.

A major scientific development should be judged by the evidence rather than the most dramatic possible interpretation.

Demonstrated in 2026:

• Two independent research teams operated nuclear clocks based on thorium-229.
• Both stabilized lasers to the nuclear transition using absorption measurements and feedback.
• The Chinese team reported a quantified fractional frequency instability and agreement between measurements in two crystals.
• The European team demonstrated clock operation and used its system in a search for time-dependent effects associated with ultralight dark matter.

Not yet demonstrated:

• Nuclear clocks outperforming the best optical atomic clocks in overall precision.
• A compact, mass-produced nuclear clock ready for widespread commercial use.
• A confirmed detection of dark matter through nuclear-clock measurements.
• A proven variation in the fundamental constants of nature.
• A nuclear clock that is immune to environmental disturbances.

This distinction is central to understanding the breakthrough. The experiments crossed a major threshold from nuclear spectroscopy to operational timekeeping, but the technology is still developing.

The significance of nuclear clocks extends beyond measuring time.

Precision measurement is one way scientists test whether their theories describe the physical world accurately. If a clock's frequency responds to a subtle effect predicted by a new theory, it can provide evidence that is difficult to obtain through other methods.

The nuclear transition of thorium-229 gives researchers a different physical reference from conventional atomic clocks. That difference can make comparisons especially sensitive to certain hypothetical interactions.

A nuclear clock therefore serves two roles: it is a timekeeping instrument and a scientific probe.

This is similar to how atomic clocks have become tools for measuring gravitational time dilation, testing relativity, and studying variations in fundamental constants. A better clock does not merely tell time more reliably; it can make new experiments possible.

The first operational nuclear clocks mark the beginning of that research program, rather than its completion.

In October 2026, two independent teams demonstrated operational nuclear clocks based on thorium-229. By stabilizing lasers to a nuclear transition, they established that a reference based on the nucleus can function as a working clock.

The devices do not yet outperform the world's best optical atomic clocks, and they are not ready to replace everyday timing systems. But they offer a new way to perform precision measurements, with potential advantages in robustness and new sensitivities to fundamental physics.

The most intriguing applications may be scientific rather than commercial: comparing nuclear and atomic clocks to search for ultralight dark matter, testing whether fundamental constants remain unchanged, and probing the nuclear forces that govern matter.

The next milestones will involve improving stability, controlling systematic effects, making vacuum-ultraviolet lasers more practical, and testing the clocks over longer periods and under a wider range of conditions.

The bigger lesson is that advances in timekeeping can become advances in our understanding of reality itself.

A clock measures an oscillation. A precision clock lets scientists test the rules governing that oscillation. And a nuclear clock may offer a new way to ask whether the deepest constants of nature are truly as constant as we believe.