Stripped Atoms and the Size of the Nucleus

How highly charged ions settled one rare-earth puzzle and uncovered another

Oct 20, 2026 · @Thomas Lee Abshier, ND

The news and what it actually says

A recent news summary announced that “highly charged ion spectroscopy resolves the Lu–Yb nuclear size puzzle.” The headline is half right. The work behind it, two preprints posted in November 2025 by a collaboration centered at Clemson University, Japan’s National Institute for Fusion Science, and the University of Electro-Communications in Tokyo, with nuclear theorists from Michigan State and Erlangen, did settle a long-standing oddity in the size of the lutetium nucleus. But in doing so, the same measurement sharpened a second, deeper problem: the size pattern it confirmed is far larger than the best current nuclear theory can explain. One puzzle closed; a harder one opened.

This essay explains what was measured, why stripping electrons from an atom makes the measurement possible, what the old anomaly was, and why the new one matters.

How big is a nucleus, and how do we know?

An atomic nucleus is tiny. The lutetium-175 nucleus has a charge radius of about 5.3 femtometers, roughly one twenty-thousandth the size of the atom around it. Physicists describe nuclear size by the root-mean-square charge radius, a measure of how far, on average, the protons’ charge extends from the center.

There are three classic ways to measure it:

1. Electron scattering. Fire high-energy electrons at a target and record how they deflect. The diffraction pattern encodes the charge distribution’s shape.
2. Muonic atoms. Replace an orbiting electron with a muon, which is about 207 times heavier and therefore orbits about 207 times closer. The muon spends real time inside the nucleus, and its energy levels shift measurably with nuclear size.
3. Optical isotope shifts. Compare a spectral line in two isotopes of the same element. The electrons are identical; only the nucleus differs. The tiny shift reveals how the radius changes when you add a neutron.

The third method is extremely precise, but it has a built-in limitation: it compares isotopes of one element. It cannot tell you how the nucleus of element 71 compares with the nucleus of element 70 because the electron clouds of different elements differ completely, and the size effect is buried under much larger differences in electronic structure. Cross-element comparisons have therefore rested on scattering and muonic data, which become difficult in heavy, strongly deformed nuclei such as those of the rare earths.

Why strip the electrons?

The nucleus’s size affects atomic energy levels through what is called the volume shift (or field shift). If the nucleus were a mathematical point, an electron near the center would feel the full Coulomb pull all the way in. Because the charge is actually spread over a small sphere, an electron that penetrates that sphere feels slightly less attraction than it would from a point. Its binding energy is slightly weaker, and this weakening grows with the size of the nucleus.

Only electrons that have appreciable probability of being at the nucleus feel this: chiefly s electrons and, relativistically, p₁/₂ electrons. In a neutral heavy atom, dozens of inner electrons screen the outer electrons that produce convenient optical lines, which spend little time near the center.

The experimenters reversed the situation. In the Tokyo electron-beam ion trap, an intense electron beam strips lutetium (71 electrons) and ytterbium (70 electrons) down to only 11 remaining electrons (the sodium-like charge state) or 12 (magnesium-like). The outermost electron now sits in an n = 3 shell around a nucleus carrying a charge of +71 or +70, with only ten core electrons to screen it. It is pulled in tightly; it overlaps the nucleus far more, and the volume shift becomes large enough to read. The transition they watched, the so-called D₁ line from the 3p₁/₂ to the 3s₁/₂ level, falls in the extreme ultraviolet, and they resolved it to about 0.01 nanometers.

The second and less obvious reason stripping helps is that it makes the theory tractable. An ion with one or two electrons outside a closed neon-like core can be calculated with relativistic many-body perturbation theory, including quantum-electrodynamic corrections, to a precision that is impossible for a neutral rare-earth atom with an open 4f shell. This enables a cross-element comparison. The measured Lu–Yb energy difference in the sodium-like ions was 3.235 ± 0.003 eV; theory, evaluated at the previously recommended radii, predicted 3.2144 eV. The residual of roughly 20 milli-electron-volts is the signal: it tells how the true radii must differ from the old values. The magnesium-like ions, which require a harder calculation, gave an independent result that agreed. Two charge states, two calculations, one answer is a meaningful check.

The old anomaly: lutetium was too large

Nuclei show a pattern called odd–even staggering. Compared to its neighbors, a nucleus with an odd number of protons (or neutrons) tends to be slightly smaller than the average of its even neighbors. The usual explanation involves pairing: like nucleons in matched pairs behave somewhat like a superfluid, and an unpaired nucleon disrupts that correlation and alters the nucleus’s size and shape.

In the deformed rare earths, the data had long suggested that odd-proton nuclei are more compact than their even-proton neighbors, with one conspicuous exception. Lutetium (Z = 71) appeared larger than expected relative to ytterbium (Z = 70) and hafnium (Z = 72). The staggering pattern seemed to invert at lutetium. The recommended lutetium radius rested on scarce data, including a 1979 MIT electron-scattering thesis, and that data had a sizable uncertainty.

The new measurement gives a lutetium-175 radius of 5.291 ± 0.011 fm, three times more precise than the electron-scattering value. Compared with ytterbium-174, which has the same number of neutrons (104), lutetium is slightly smaller by 0.017 ± 0.010 fm. The inversion disappears. Lutetium behaves like the other odd-proton nuclei after all.

Two honest qualifications belong here. First, the difference is small compared with its uncertainty: it is less than two standard deviations from zero, so the sign is favored rather than nailed down. Second, the authors note that their result agrees with the direction of the old electron-scattering difference but disagrees with its absolute values, suggesting a systematic error in that older experiment that affected both radii. In other words, the anomaly appears to have been a measurement artifact rather than a feature of nature.

The new puzzle: the staggering is too large

With the expected pattern restored, the team examined its magnitude along a chain of nuclei sharing 94 neutrons (europium through lutetium), combining their result with existing muonic-atom and isotope-shift data. Here the trouble begins.

The odd–even staggering along that chain is enormous: about thirty times larger than the staggering seen in other chains, whether in spherical nuclei near the magic numbers or in the deformed ytterbium isotopes.

The theorists compared this with nuclear density functional theory, the most successful general framework for heavy nuclei. They used three versions: one Skyrme functional and two Fayans functionals, each calibrated to the same nuclear data and each with quantified uncertainties. The models do well on what one would hope: they reproduce the absolute radii, the staggering in spherical nuclei, and the staggering in the deformed ytterbium isotope chain. But along the 94-neutron chain, they predict a staggering that is many times too small.

Why would theory fail here? Two mechanisms normally produce odd–even staggering. Pairing dominates in spherical nuclei. In soft, easily reshaped nuclei, an odd nucleon can polarize the whole nucleus’s shape. But the nuclei in question are well deformed, rigid footballs rather than soft ones, and for them both effects are expected to be weak. The calculations confirm that: the computed pairing and deformation show little staggering along this chain. The authors therefore suggest that current models may be missing a structural effect.

They are also careful about the alternative. The large staggering depends on recommended radii for terbium and holmium, which come mainly from muonic-atom measurements that are notoriously difficult in this deformed region. The authors call for better terbium and holmium radii, or their differences relative to dysprosium, to decide whether the enhancement is a real feature of nuclear structure or a statistical or systematic artifact in the older data. Highly charged ion spectroscopy is well suited to that task.

Why it matters

Nuclear charge radii are not merely catalog entries. They feed into precision atomic calculations, including those behind optical clocks. They are needed to interpret isotope-shift searches for new forces: ytterbium in particular has been used to bound a hypothetical boson coupling electrons to neutrons, and such searches depend on understanding the nucleus well enough to separate new physics from ordinary nuclear structure. Radius trends across elements also constrain nuclear matter properties, including how the strong force treats protons and neutrons differently.

The method itself is the more lasting contribution. Until now, precise cross-element comparisons of nuclear size were largely out of reach. With few-electron ions and accurate atomic theory, these comparisons are now routine. The group’s next steps include neighboring rare earths, radioactive nuclides, and heavier elements; the lead author also listed a 2026 conference presentation on the thorium-229 to thorium-232 radius difference among targets, which bears on the proposed thorium nuclear clock.

What the episode teaches

The headline said a puzzle was resolved. More precisely, an apparent exception was shown to be a measurement problem, and once it was removed the regular pattern came into view at a magnitude theory does not predict. That is how measurement usually advances understanding: not by confirming what is believed, but by showing more exactly where belief and nature part company. The lutetium anomaly is probably gone. The question of why odd-proton deformed nuclei shrink so much more than theory allows, if indeed they do, is now open and better defined than before.

References

• E. Takacs, H. Staiger, et al., “Puzzling Isotonic Odd-Even Staggering of Charge Radii in Deformed Rare Earth Nuclei,” arXiv:2511.19395 (2025). https://arxiv.org/abs/2511.19395
• H. Staiger, E. Takacs, et al., “Extreme Ultraviolet Spectroscopy of Highly Charged Lu and Yb Ions for Nuclear Charge Radius Determination,” arXiv:2511.20537 (2025). https://arxiv.org/abs/2511.20537
• “Studying isotonic odd-even staggering of nuclear charge radii via spectroscopy of highly charged Yb and Lu,” APS DAMOP 2026 abstract. https://meetings-archive.aps.org/damop/2026/n12/3/
• H. Staiger, “Extracting nuclear charge radii systematics from spectroscopy of highly charged ions,” NREC 2026 presentation.
• I. Angeli and K. P. Marinova, “Table of experimental nuclear ground state charge radii: An update,” Atomic Data and Nuclear Data Tables 99, 69 (2013).
• M. Door et al., “Probing New Bosons and Nuclear Structure with Ytterbium Isotope Shifts,” Physical Review Letters 134, 063002 (2025).