A Laser Built for Nuclear Timekeeping

Chuankun Zhang; Tian Ooi

Viewpoint

A Laser Built for Nuclear Timekeeping

¹Department of Applied Physics and Materials Science, Institute for Quantum Information and Matter, Caltech, Pasadena, CA, US
²JILA, National Institute of Standards and Technology and University of Colorado Boulder, Boulder, CO, US

February 11, 2026• Physics 19, 19

Researchers have designed and demonstrated an ultraviolet laser that removes a major bottleneck in the development of a nuclear clock.

Figure caption — **Figure 1:** Xiao and colleagues combined three lasers (two with the same frequency) in a cell containing cadmium vapor [1, 2]. The result was a single-frequency continuous-wave ultraviolet laser. This laser can drive a specific transition—in thorium-229 nuclei—that could form the basis of a nuclear clock.

Whereas ordinary atomic clocks keep time using transitions of electrons in atoms, a prospective nuclear clock would harness a transition between states of the nucleus. Compared with electronic transitions, nuclear ones are much less sensitive to environmental disturbances, which would potentially give nuclear clocks unprecedented precision and stability. Such devices could improve GPS systems and enable more sensitive probes of fundamental physics. The main hurdle has been that nuclear transitions are extremely difficult to drive controllably using existing laser technology. Now Qi Xiao at Tsinghua University in China and colleagues have proposed and realized an intense single-frequency ultraviolet laser that can achieve such driving for thorium-229 nuclei [1, 2]. Beyond timekeeping, the team’s laser platform could find uses across quantum information science, condensed-matter physics, and high-resolution spectroscopy.

For most nuclear transitions, the energy difference between the two states lies in the kilo-electron-volt to mega-electron-volt range. Consequently, such transitions are inaccessible to today’s high-precision lasers, which can deliver photons of typically a few electron volts in energy. A long-known exception is the transition between the ground state and first excited state of thorium-229 nuclei. Indirect measurements over the past 50 years have gradually pinned down that transition’s energy difference to only about 8.4 eV. As a result, this transition is being actively investigated as a candidate for developing a nuclear clock.

A thorium-229 nuclear clock would open a plethora of new opportunities [3]. Whereas today’s most accurate atomic clocks require their atoms or ions to be trapped in a technically challenging ultrahigh vacuum, a thorium-229 clock could operate even if its atoms were embedded in a simple solid crystal. Such a solid-state clock would be highly portable and easily deployable outside the lab. More interestingly, by packing almost an Avogadro’s number of thorium-229 atoms into the crystal, the clock’s stability could be drastically improved simply by averaging over many identical nuclei. Lastly, comparing the nuclear clock with conventional atomic clocks would provide a stringent test of the standard model of particle physics by probing a possible time variation of fundamental constants [3].

Operating such a clock, however, would demand specially designed lasers because addressing the thorium-229 transition requires light with a precise wavelength of about 148 nm, corresponding to the vacuum-ultraviolet region of the electromagnetic spectrum. The first successful laser excitations of this transition used nanosecond-long pulses repeating at rates of tens of hertz, produced by four-wave mixing [4, 5]. In this process, three existing laser frequencies can be combined to generate a new, otherwise inaccessible one. The short pulse duration naturally provides high peak intensities, a typical requirement for generating vacuum-ultraviolet light through nonlinear optics. However, the use of short pulses also results in a broad optical spectrum, with a linewidth usually on the order of 10 GHz [4, 5]. A recent implementation of this scheme narrowed the laser linewidth to roughly 30 MHz [6], setting an upper limit on the resolution of nuclear spectroscopy based on this approach.

Later experiments relied on trains of femtosecond pulses with a high repetition rate of about 100 MHz. Possessing a precisely controlled optical phase, these trains corresponded to a set of regularly spaced spectral lines in the frequency domain. Known as frequency combs, such laser sources combined the high peak intensity needed for nonlinear frequency up-conversion with the narrow spectral linewidth needed for high-precision spectroscopy. The combs enabled scientists to resolve key detailed features of the thorium-229 transition and to characterize the long-term reproducibility of a nuclear clock based on that transition [7, 8]. The narrowest observed transition feature, with a width of about 20 kHz [8], was limited by crystal inhomogeneity instead of laser linewidth. However, the frequency comb distributed its power among roughly 100,000 spectral lines, among which only one was resonant with the nuclear transition. This property severely limited the power available for clock operation, compared with an ideal single-frequency continuous-wave laser with a similar average power.

Xiao and colleagues aimed to efficiently generate such a single-frequency vacuum-ultraviolet laser by engineering a giant resonant nonlinear effect in the medium used for four-wave mixing [1, 2]. Through electron-correlation calculations, the researchers identified cadmium vapor as the ideal medium owing to its strong light–matter coupling and suitable energy-level structure. The team considered various experimental schemes by closely examining available laser technologies and their suitability for providing the three lasers driving the four-wave-mixing process. The researchers then predicted the optimal operational conditions by realistically modeling how the lasers interact with the cadmium vapor [1]. This detailed theoretical investigation served as the roadmap for the team’s experimental realization of a high-power continuous-wave vacuum-ultraviolet laser operating at the required 148-nm wavelength (Fig. 1) [2].

A similar concept for generating a continuous-wave vacuum-ultraviolet laser had previously been investigated at other wavelengths [9]. But whether such a laser would retain the crucial property of optical coherence had not yet been experimentally verified. Using the same set of driving lasers, Xiao and colleagues produced two vacuum-ultraviolet lasers from separate cadmium-vapor cells. The interference of those two lasers showed that the team’s resonantly enhanced four-wave-mixing process faithfully preserves the optical coherence of the driving lasers. The researchers placed a conservative upper bound of 100 Hz on the linewidth of their vacuum-ultraviolet laser [2]. If the laser beam could be tightly focused to a spot about 2 µm in diameter, the resulting intensity would be high enough to drive the nuclear transition at a rate comparable to the laser’s linewidth. This capability shows a clear path toward coherent laser manipulation of nuclear states.

Although the demonstrated laser power of 100 nW is already impressive, the team’s simulations and experiments suggest it could be pushed much higher. By simply boosting the power of the driving lasers using commercially available laser sources, more than 30 µW of vacuum-ultraviolet power should be achievable [1]. Beyond its application as a nuclear-clock laser, the demonstrated source also features a wide spectral tunability, making this scheme attractive to general vacuum-ultraviolet spectroscopy applications in atomic and condensed-matter physics. In parallel work, scientists generated a continuous-wave laser at the same wavelength using direct frequency doubling in a solid crystal [10]. That laser had a comparatively low output power of about 1 nW, but such a laser using a solid-state medium may facilitate a future portable nuclear clock. The results of Xiao and colleagues demonstrate the exciting pace at which vacuum-ultraviolet lasers are evolving, opening the door for coherent laser manipulation of the thorium-229 nuclear state.

References

Q. Xiao et al., “Proposal for the generation of continuous-wave vacuum-ultraviolet laser light for Th-229 isomer precision spectroscopy,” Phys. Rev. Appl. 25, 024034 (2026).
Q. Xiao et al., “Continuous-wave narrow-linewidth vacuum ultraviolet laser source,” Nature (2026).
E. Peik et al., “Nuclear clocks for testing fundamental physics,” Quantum Sci. Technol. 6, 034002 (2021).
J. Tiedau et al., “Laser excitation of the Th-229 nucleus,” Phys. Rev. Lett. 132, 182501 (2024).
R. Elwell et al., “Laser excitation of the ²²⁹Th nuclear isomeric transition in a solid-state host,” Phys. Rev. Lett. 133, 013201 (2024).
T. Hiraki et al., “Laser Mössbauer spectroscopy of ²²⁹Th,” arXiv:2509.00041.
C. Zhang et al., “Frequency ratio of the ^229mTh nuclear isomeric transition and the ⁸⁷Sr atomic clock,” Nature 633, 63 (2024).
T. Ooi et al., “Frequency reproducibility of solid-state thorium-229 nuclear clocks,” Nature 650, 72 (2026).
K. S. E. Eikema et al., “Continuous wave coherent Lyman- $𝛼$ radiation,” Phys. Rev. Lett. 83, 3828 (1999).
V. Lal et al., “Continuous-wave laser source at the 148 nm nuclear transition of Th-229,” Optica 12, 1971 (2025).

About the Authors

Chuankun Zhang is an IQIM postdoctoral scholar in the Department of Applied Physics and Materials Science at Caltech, working on quantum-network nodes in Andrei Faraon’s group. He obtained his PhD in 2025 from Jun Ye’s group at JILA, a joint institute of the National Institute of Standards and Technology and the University of Colorado Boulder, developing a thorium-229 nuclear clock using a vacuum-ultraviolet frequency comb.

Tian Ooi is a graduate student in Jun Ye’s group at JILA, a joint institute of the National Institute of Standards and Technology and the University of Colorado Boulder. She is working on the creation and characterization of solid-state thorium nuclear clocks. Previously, she studied engineering physics as an undergraduate researcher in Lindsay LeBlanc’s group at the University of Alberta in Canada.