quant-ph·Quantum Physicsphysics·Physics

Thorium-229 Nuclei Stabilize an Operational Solid-State Clock

A 148.4-nm laser locked through frequency-modulated absorption in thorium-doped calcium fluoride reaches 5×10⁻¹³/√(τ/s) fractional instability.

Editorial Desk·October 21, 2026·5 min readstrong

Underlying Paper

A nuclear clock synchronized to $^{229}$Th

Beichen HuangGaowei YanQi XiaoWenhao BuChengchun ZhaoZhen ZhangChao YanZhi-Ang ChenPeixiong ZhangGleb PenyazkovZhenhai ZhanLingfeng YanYuefei WangLin LiShanming LiDapeng JiangXiaobo QianXuegang LiuQiange HeTaoxiang SunHaochen TianBingkun LuNingyuan MaJuxian LiYanzhang WuQiaorui GongYuxiang LiHaoyu ShiXiangliang LiLongsheng MaShining ZhuYuxiang MoJun LinLi YouYige LinXibo ZhangYin HangLiangbi SuShiqian Ding

Atomic clocks have made time and frequency the most precisely measured quantities in physics, progressing from microwave standards that realize the SI second to optical clocks with unprecedented precision. A nuclear clock transfers the frequency reference from an electronic to a nuclear transition, and the uniquely low-lying, laser-accessible isomeric transition in $^{229}$Th currently offers the most practical route to compact, robust timekeeping and sensitive tests of fundamental physics. Realizing such a clock requires turning spectroscopy of the $^{229}$Th nuclear resonance into a stable discriminator for steering a traceable oscillator. Here we demonstrate a $^{229}$Th nuclear clock by stabilizing a continuous-wave, narrow-linewidth 148.4-nm vacuum-ultraviolet (VUV) laser to a resolved, weakly temperature-sensitive nuclear transition in $^{229}$Th$:$CaF$_2$ crystals. A $10$-$\mu\mathrm{W}$ VUV source generated by four-wave mixing in cadmium vapour and phototube-based frequency-modulation absorption readout provide a fast, high-signal-to-noise nuclear discriminator. The clock reaches a fractional frequency instability of $5\times10^{-13}/\sqrt{\tau/\mathrm{s}}$ for averaging time $\tau$. Clock-transition frequencies measured in two independently fabricated crystals agree at the $10^{-13}$ level and are consistent with previous VUV-comb measurements on other $^{229}$Th:CaF$_2$ crystals. These results establish laser-addressed nuclei as operational clock references and provide a reproducible solid-state platform for compact nuclear clocks, nuclear quantum sensors and precision tests of fundamental physics.

arXiv:2606.08870Submitted: Oct 8, 2026v2
arXiv Abstract PDF

Optical clocks derive their precision from narrow electronic transitions, but their performance often depends on elaborate atomic or ionic systems. A nuclear reference could be less sensitive to environmental perturbations, provided the nuclear transition can be addressed by a laser and converted into a stable error signal. This work reports that step for the low-energy isomeric transition of 229^{229}Th embedded in CaF2_2: a continuous-wave vacuum-ultraviolet laser is locked to a resolved nuclear line rather than used only for spectroscopy.

The central result is operational rather than merely spectroscopic. The authors use the nuclear absorption signal to steer the VUV source and report a fractional frequency instability of 5×10−13/τ/s5\times10^{-13}/\sqrt{\tau/\mathrm{s}}. They also measure the clock-transition frequency in two independently fabricated crystals and find agreement at the 10−1310^{-13} level. That combination makes the result a credible demonstration of a nuclear transition functioning as a repeatable frequency reference in a solid-state host.

Core Contribution

The paper turns an ensemble nuclear resonance into a clock discriminator. Prior measurements of 229^{229}Th:CaF2_2 established laser excitation and mapped spectral structure, but a clock needs a dispersive signal around a selected transition and a feedback path that continuously corrects the oscillator. Here, the authors lock to line b, a quadrupole-resolved component associated with mg=5/2m_g=5/2 and mis=3/2m_{is}=3/2, using frequency-modulated absorption rather than fluorescence as the primary clock readout.

That choice matters because it makes the spectroscopy signal directly useful for feedback. The zero crossing of the lock-in-demodulated absorption trace defines the lock point; its local slope maps fluctuations in normalized signal to residual frequency detuning. The paper therefore demonstrates a full reference-and-servo chain, not simply a measured resonance frequency.

Technical Approach

The apparatus generates 148.4-nm VUV light by resonantly enhanced four-wave mixing in cadmium vapour. Two 375-nm photons drive a two-photon cadmium resonance, while a 710-nm photon completes the sum-frequency process. The reported VUV power is 10 μW\mu\mathrm{W}. A 750-nm laser and the 710-nm laser are phase-locked to a self-referenced erbium-fibre frequency comb; the comb repetition rate is tied through a 4-km phase-stabilized fibre link to a cryogenic-silicon-cavity-stabilized 1,397-nm laser, while the carrier-envelope offset is referenced to a hydrogen maser.

The VUV beam interrogates home-grown 229^{229}Th:CaF2_2 crystals. The authors compare an S1 crystal and a TS1 crystal, as well as photomultiplier and phototube readout configurations. In the optimized TS1-plus-phototube configuration, the modulation frequency is 0.888 kHz, the modulation amplitude is 12 kHz, the lock-in time constant is 0.1 s, the measurement time is 10 s, and the feedback-cycle duration is 11.8 s. The servo corrects the optical chain through an acousto-optic modulator.

Figure 1 lays out this chain from VUV generation through absorption detection and feedback, making clear that the stabilized quantity is the VUV frequency at the nuclear resonance.

Figure 1. Experimental platform for a solid-state ^229Th nuclear clock. a, Experimental setup. Continuous-wave 148.4-nm VUV light is generated by four-wave mixing in cadmium vapour and sent to a home-grown ^229Th:CaF_2 crystal. Nuclear fluorescence is collected with a photomultiplier tube (PMT), while the transmitted VUV power is recorded with a phototube or a PMT. During absorption spectroscopy and clock operation, the 710-nm laser is modulated, and the transmitted-VUV photocurrent is detected via lock-in demodulation.

Results and Analysis

The spectroscopy data resolve multiple quadrupole and quadrupole-unresolved components, with absorption scans acquired in as little as 10 s in TS1 using the phototube. Fine scans of the unresolved component did not reveal reproducible narrow features on scales of roughly 500 Hz or larger. The observed line profile also contains material-specific complexity: a full-range fit to line b in TS1 finds a narrow component of 26(1) kHz and a broad pedestal of 334(44) kHz, while fitting the central region alone gives 27(1) kHz. The clock can thus operate on a resolved central feature, but the pedestal is a reminder that the crystal host is part of the metrology problem.

The Allan-deviation comparison across four configurations is the key performance test. The optimized TS1 phototube system is the reported best configuration, reaching 5×10−13/τ/s5\times10^{-13}/\sqrt{\tau/\mathrm{s}}. Figure 3 compares it against S1 with a photomultiplier, S1 with a phototube, and a non-optimized TS1 phototube implementation. The evidence supports the narrower claim that absorption readout, crystal selection, and feedback optimization materially improve short-term stability. It does not yet establish that the solid-state nuclear clock outperforms mature optical clocks; its significance is the demonstrated nuclear reference and reproducibility across crystals.

Figure 3. Fractional frequency instability of the nuclear clock. Overlapping Allan deviations measured for four configurations: S1 + PMT, S1 + phototube, TS1 + phototube and optimized TS1 + phototube. Dashed lines, weighted ^-1/2 fits.

Caveats in Practice

The evaluation is confined to two CaF2_2 crystals and to the reported VUV, modulation, and servo configurations. The broad spectral pedestal and tens-of-kilohertz ensemble linewidth show that inhomogeneity and host-dependent line-shape effects remain relevant. In addition, the paper reports instability rather than a complete long-term systematic-uncertainty budget. Compact nuclear clocks and fundamental-physics sensors are plausible beneficiaries, but those applications require stability to be paired with equally well-characterized accuracy.

Evidence Box

strong

Key Claims

  • •A laser-addressed ²²⁹Th nuclear transition can serve as a clock reference
  • •Frequency-modulated absorption provides a usable nuclear clock discriminator
  • •Independently fabricated ²²⁹Th:CaF₂ crystals provide reproducible transition frequencies

Key Results

  • •Fractional frequency instability of 5×10⁻¹³/√(τ/s) in the optimized TS1 phototube configuration
  • •Clock-transition frequencies agree between two crystals at the 10⁻¹³ level
  • •Optimized feedback uses 0.888 kHz modulation, 12 kHz amplitude, and an 11.8 s feedback cycle
  • •TS1 line b has a 26(1) kHz narrow component and a 334(44) kHz broad pedestal in a double-Lorentzian fit

Limitations & Caveats

  • •Evaluation covers two CaF₂ crystals rather than broad host-to-host reproducibility
  • •Broad 237–334 kHz pedestals accompany the resolved 26–75 kHz spectral components
  • •No complete long-term systematic frequency-uncertainty budget is reported
  • •Fine scans found no reproducible features narrower than approximately 500 Hz in the QU structure

Related Articles

Readers are encouraged to consult the original arXiv paper for complete details. SOTA Papers does not make claims beyond what is supported by the authors' reported evidence.