Preprint reports operation of thorium-229 nuclear clock locked to nuclear transition in crystal

Researchers reported in an arXiv preprint that they have operated a thorium-229 nuclear clock by locking a laser to a nuclear transition in a crystal, rather than to the electronic transitions used in today’s atomic clocks. In the paper, posted June 7 as arXiv:2606.08870v1, first author Beichen Huang and co-authors write: “Here we demonstrate the operation of a 229Th nuclear clock by stabilizing a continuous-wave narrow-linewidth 148.4 nm vacuum-ultraviolet (VUV) laser to a resolved nuclear transition in a solid-state host.”

The work has not yet been peer-reviewed, and its significance lies in what it would represent if confirmed: a move from spectroscopy of thorium-229’s much-studied nuclear resonance to a functioning clock reference. Atomic clocks keep time using transitions in electrons around an atom. A nuclear clock would instead use a transition inside the nucleus itself.

According to the authors, the clock is built around a continuous-wave, narrow-linewidth vacuum-ultraviolet laser at 148.4 nanometers, with a reported power of 10 microwatts. They say that light was generated by four-wave mixing in cadmium vapor, a nonlinear optical technique used to create difficult wavelengths. The laser was then stabilized to a nuclear resonance in a home-grown thorium-doped calcium fluoride crystal, written as 229Th:CaF2. Clock operation, the preprint says, was enabled by fast frequency discrimination using phototube photocurrent readout of transmitted VUV power — in plain terms, detecting tiny changes in the light passing through the crystal and using that signal to keep the laser locked to the resonance.

On performance, the authors report a fractional frequency instability of 2 × 10^-12 divided by the square root of the averaging time in seconds, a standard way of describing how a clock’s noise falls with longer measurement time. The abstract also says measurements made with two separate crystals matched closely. “Remarkably, nuclear-clock frequencies measured with two distinct crystals agree at the 10^-13 level, demonstrating the reproducibility of solid-state nuclear frequency references,” the authors write. As with the rest of the paper, that result is so far a claim in a preprint, not an independently validated benchmark.

Thorium-229 has long stood out because it has a uniquely low-energy nuclear isomer — an excited nuclear state that, unusually, can be reached with lasers. That has made it the leading candidate for a nuclear clock. The new report follows two pieces of enabling work published in Nature in 2026: one on frequency reproducibility in 229Th:CaF2 crystals, including operation near 196 kelvin, where first-order thermal sensitivity vanishes, and another on a continuous-wave, narrow-linewidth VUV source near 148.4 nanometers generated by four-wave mixing in cadmium vapor. Together, those advances addressed two central obstacles: creating the needed light and showing the solid-state crystal reference could behave reproducibly.

If the new claim holds up, it would mark a notable extension of precision timekeeping from electronic transitions to a nuclear one. In restrained terms, that would mean researchers are no longer only probing the thorium resonance, but beginning to use it as a working reference. That could support more compact precision clocks and new precision measurements, including tests of fundamental physics, though the immediate takeaway is narrower: a preprint now says the field has crossed an important operational threshold.