🔬 atomic physics

Origin of the inhomogeneous linewidth in a solid state $^{229}$ Th:CaF $_2$ nuclear clock

This study identifies local electric-field gradient variations caused by thorium dopant concentration as the dominant source of inhomogeneous linewidth broadening in solid-state $^{229}$ Th:CaF $_2$ nuclear clocks, providing a microscopic explanation for the observed spectral characteristics and guiding future material engineering to achieve sub-Hertz precision.

Original authors: Kai Li, Tian Ooi, Jack F. Doyle, Konstantin Herb, Emil Pellett, Martin Pimon, Thorsten Schumm, Steven M. Girvin, Leo Radzihovsky, Jun Ye

Published 2026-10-06

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kai Li, Tian Ooi, Jack F. Doyle, Konstantin Herb, Emil Pellett, Martin Pimon, Thorsten Schumm, Steven M. Girvin, Leo Radzihovsky, Jun Ye

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Time, as we measure it, is defined by the most regular vibration nature can offer. For decades, the gold standard has been the oscillation of atoms, specifically the way electrons jump between energy levels. But scientists have long looked toward the atomic nucleus itself for a clock that is even more precise. The nucleus of a thorium-229 atom holds a secret: it possesses a special, low-energy state that can be triggered by light. Unlike most nuclear transitions, which require high-energy X-rays or gamma rays to activate, this specific jump in thorium-229 happens at an energy level reachable by ultraviolet lasers. Because the nucleus is so much heavier and more isolated than the surrounding electrons, it is less disturbed by its environment, promising a clock that is far more stable than current atomic timekeepers. This stability could allow us to test the fundamental laws of physics with unprecedented sensitivity, searching for changes in the constants of nature or the presence of dark matter.

To build such a clock, researchers embed these thorium atoms into a solid crystal, specifically calcium fluoride, creating a material where millions of nuclei can be studied at once. However, a significant obstacle has stood in the way: the signal from these nuclei is blurry. In a perfect vacuum, the transition should be incredibly sharp, defined by a natural width of less than one hertz. Yet, when the atoms are placed inside the crystal, the signal widens dramatically, stretching to tens or even hundreds of kilohertz. This blurring, known as inhomogeneous broadening, acts like static on a radio, making it difficult to lock onto the precise frequency needed for a clock. For years, the exact cause of this blur remained a mystery, with theories pointing to various interactions within the crystal lattice.

A team of researchers has now solved this puzzle by systematically testing how the blur changes under different conditions. They grew high-quality crystals doped with varying amounts of thorium and measured the width of the nuclear signal at different temperatures and concentrations. Their findings revealed a clear pattern that ruled out several common suspects. First, they found that the width of the signal did not change when they heated or cooled the crystal, which meant that thermal jitters or vibrating atoms were not the primary cause. Second, they discovered that the width grew in direct proportion to the number of thorium atoms added to the crystal. The more thorium present, the blurrier the signal became. Finally, they observed that the amount of blur depended heavily on the specific orientation of the nuclear spin, a property known as the quadrupole state.

These clues pointed the researchers toward a single, dominant cause: static variations in the local electric environment surrounding each nucleus. Inside the crystal, the thorium atoms do not sit perfectly still in a uniform grid; instead, they create tiny distortions in the lattice structure. These distortions alter the electric field gradient—the way electric forces change over a short distance—at the site of each nucleus. Because every thorium atom sits in a slightly different local environment, each one experiences a slightly different electric field, causing its transition frequency to shift by a tiny, unique amount. When you look at the entire collection of atoms, these millions of tiny shifts add up to create the broad, blurry line observed in the experiments. The researchers determined that this disorder scales linearly with the concentration of thorium, confirming that the dopant atoms themselves are the source of the disturbance.

To understand the microscopic details of this disturbance, the team combined their experimental data with advanced computer simulations. They tested three different possible arrangements for how the thorium atoms sit within the crystal and how the crystal compensates for the extra charge of the thorium. Two of the proposed arrangements failed to match the experimental data, predicting a pattern of blurring that did not align with what was measured. Only one configuration, where two thorium atoms sit next to each other as a pair, successfully reproduced the specific hierarchy of blurring seen across the different nuclear states. This dimer arrangement creates a specific type of electric field distortion that matches the experimental observations, suggesting that these paired atoms are the primary architects of the signal's width.

The study also looked at what happens when the concentration of thorium is reduced to zero. Even in the absence of thorium, a small amount of blurring remains, which the researchers attribute to inherent imperfections in the crystal itself, such as vacancies or stray impurities. By extrapolating their data, they estimated that if these sources of disorder could be eliminated, the remaining width of the signal would be around five kilohertz. This residual width is still much larger than the natural limit, but it is small enough to suggest that the path forward is clear. The work establishes that the primary barrier to a solid-state nuclear clock is not an unfixable flaw in the physics, but rather a manageable issue of material quality. By engineering crystals with fewer defects and perhaps using host materials that naturally accommodate thorium without distortion, scientists can now aim to narrow the signal significantly, bringing the dream of a nuclear clock within reach.

Technical Summary: Origin of the Inhomogeneous Linewidth in a Solid-State $^{229}$ Th:CaF $_2$ Nuclear Clock

Problem Statement
Solid-state $^{229}$ Th nuclear clocks offer a promising route toward next-generation optical frequency standards due to the exceptionally narrow natural linewidth (sub-Hertz) and high quality factor of the nuclear isomer transition. However, in practical solid-state implementations, such as thorium-doped calcium fluoride (CaF $_2$ ), the observed transition linewidths are broadened to the tens-to-hundreds-of-kilohertz range. This inhomogeneous broadening, which exceeds the radiative limit by several orders of magnitude, obscures the fundamental properties of the transition and limits clock performance. While previous theoretical work predicted homogeneous broadening on the order of 1 kHz (due to magnetic dipole interactions and second-order Doppler effects), the dominant source of the observed kilohertz-level broadening remained unidentified.

Methodology
The authors employed a combined experimental and theoretical approach to isolate the mechanisms driving this broadening:

Systematic Experimental Characterization: Using VUV frequency-comb spectroscopy, the team measured the linewidths of four allowed nuclear quadrupole transitions in $^{229}$ Th-doped CaF $_2$ single crystals. Measurements were conducted across a temperature range of 150–293 K and across three crystals with varying thorium concentrations ( $n \approx 0.3, 0.8, \text{and } 4.0 \times 10^{18} \text{ cm}^{-3}$ ).
Data Analysis via Hamiltonian Modeling: The researchers modeled the transition frequencies using a quadrupole-interaction Hamiltonian. They analyzed the linewidths as a function of the local electric-field gradient (EFG) variations ( $\delta V_{zz}$ and $\delta V_{yy}$ ), assuming an elliptical bivariate Cauchy distribution for the EFG disorder. This model allowed them to extract the magnitude and correlation of EFG fluctuations from the measured linewidths.
Theoretical Simulations: To identify the microscopic origin of the disorder, the team performed first-principles density-functional-theory (DFT) calculations and classical atomistic simulations. They evaluated three candidate defect configurations for the thorium dopant: the $C_{3v}$ configuration (Th with two interstitial fluorines), the $v_{Ca}$ configuration (Th with a calcium vacancy), and a dimer configuration (two adjacent Th ions). These simulations calculated the resulting EFG distributions and predicted the corresponding linewidths.

Key Results

Temperature Independence: The transition linewidths showed no measurable dependence on temperature across the 150–293 K range, despite significant temperature-induced frequency shifts. This indicates that the broadening is not driven by thermally excited phonons but by static disorder in the crystal lattice.
Linear Concentration Dependence: The linewidths increased linearly with thorium concentration for all measured transitions. The slope of this dependence varied by transition, following the ordering $a > c \approx d > b$ , which mirrors the relative magnitudes of the linewidths themselves.
Dominance of EFG Disorder: Data analysis identified static variations in the local electric-field gradient (EFG) as the primary source of inhomogeneous broadening. The extracted EFG variations ( $\Gamma_{zz}$ and $\Gamma_{yy}$ ) scaled linearly with thorium concentration.
Lorentzian Line Shapes: The observed spectra were well-described by Lorentzian profiles rather than Gaussian or Voigt profiles. Theoretical analysis attributed this to the $1/r^3$ scaling of long-range perturbations induced by randomly distributed dopants, which naturally yields a Lorentzian distribution of frequency shifts.
Microscopic Origin (Dimer Configuration): Among the simulated defect configurations, the dimer configuration (two neighboring Th ions) provided the most consistent description of the experimental data. It reproduced the observed hierarchy of transition linewidths and the specific ratio of EFG fluctuation components ( $\Gamma_{yy}/\Gamma_{zz} \approx 1.6$ ) and correlation ( $\rho \approx -0.6$ ). While the $C_{3v}$ and $v_{Ca}$ models predicted different linewidth hierarchies and fluctuation ratios, the dimer model uniquely matched the experimental signatures, suggesting that Th-Th dimers are the dominant defect structure responsible for the broadening.
Residual Broadening: Extrapolating the linewidth to zero thorium concentration revealed a common non-zero intercept for all transitions. This residual broadening is attributed to EFG disorder arising from host crystal imperfections (e.g., vacancies, interstitials) rather than thorium dopants, with an estimated magnitude of $\sim 5$ kHz.

Significance and Claims
The paper establishes that the dominant mechanism limiting the performance of solid-state $^{229}$ Th nuclear clocks is inhomogeneous broadening caused by static variations in the local electric-field gradient. These variations arise primarily from long-range lattice distortions induced by thorium dopants (specifically in dimer configurations) and residual crystal strain.

The authors claim that their work provides:

A definitive identification of the broadening mechanism, distinguishing it from previously hypothesized homogeneous effects.
A quantitative model linking dopant concentration and defect geometry to the observed linewidth, explaining both the linear concentration dependence and the Lorentzian line shape.
Experimental constraints on the microscopic structure of the thorium doping site, strongly supporting the dimer configuration over other charge-compensation models.
A pathway for engineering improved nuclear-clock materials. By identifying EFG disorder as the bottleneck, the authors suggest that reducing dopant-induced lattice distortions (e.g., by using thorium-based host materials like ThF $_4$ ) and minimizing residual crystal strain could substantially narrow the resonances, potentially allowing access to the homogeneous broadening regime ( $\sim 1$ kHz) and realizing the full potential of solid-state nuclear clocks.

More like this