Lande g-factor measurements for the 5d6s 3D2 hyperfine levels of 176Lu+

K. J. Arnold; M. D. Barrett; M. D. K. Lee; N. Jayjong; Qin Qichen; Qi Zhao; Zhao Zhang

REVIEW 3 major objections 4 minor 16 references

Lande g-factor measurements for the 5d6s 3D2 hyperfine levels of 176Lu+

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read g-factor measurements bound lutetium clock shift to 10^-19 level

desk verdict Careful g-factor measurements for 3D2 in 176Lu+; the RQM estimate is honest but model-dependent, and the abstract/Eq. (15) uncertainty mismatch needs fixing. read the letter →

arxiv 2507.16292 v1 pith:LSNLWYXH submitted 2025-07-22 physics.atom-ph quant-ph

Qi Zhao , M. D. K. Lee , Qin Qichen , Zhao Zhang , N. Jayjong , K. J. Arnold , M. D. Barrett This is my paper · ORCID

classification physics.atom-phquant-ph

keywords Landeg-factor176Lu+residualquadrupolemomenthyperfine-mediatedeffectsopticalclocktrappedionZeemansplittingshift

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper pins down the last major unquantified systematic of the $^{176}\mathrm{Lu}^+$ $^1S_0 \leftrightarrow {}^3D_2$ optical clock transition: the residual quadrupole moment (RQM) left over when hyperfine averaging cancels the dominant $J$-dependent quadrupole shift. The authors measure the Landé $g$-factors of all five $^3D_2$ hyperfine levels to a fractional inaccuracy of $5\times 10^{-7}$, and combine those numbers with a theoretical decomposition of hyperfine-mediated corrections. From this they infer $\delta\Theta = 1.59(34)\times 10^{-4}\,ea_0^2$, which bounds the clock's quadrupole shift to the low $10^{-19}$ level under typical operating fields. If correct, the result removes a key obstacle for using the $^3D_2$ transition as a second clock reference in the same ion, enabling a system-level validation of the lutetium frequency reference.

What carries the argument

The key machinery is the orthogonal projection of the five measured $g$-factors onto a basis that separates the electronic and nuclear $g$-factor contributions from the hyperfine-mediated corrections, leaving four independent parameters plus a residual $\beta$ that measures the size of all higher-order terms. The crucial physical relation is Eq. 14, which converts each magnetic-dipole hyperfine parameter $\beta^{(k)}$ into the corresponding electric-quadrupole correction $\beta_Q$ via the ratio of reduced matrix elements $\langle J \| \Theta^{(2)} \| J' \rangle / \langle J \| m \| J' \rangle$. To close the system, the analysis takes the coupling to the $^1D_2$ singlet state to be negligible and uses the nuclear $g$-factor $g_I$ from prior $^3D_1$ work.

What would settle it

A direct measurement of the electric-quadrupole shift of the $^1S_0 \leftrightarrow {}^3D_2$ clock transition, made by varying the applied dc electric field gradient and looking for a frequency shift proportional to the gradient at the level of $\delta\Theta \approx 1.6\times 10^{-4}\,ea_0^2$, would settle the claim: a matching shift confirms the inference, while a null result at that sensitivity would falsify it.

Watch

Extended reading notes

Core claim

The paper reports measurements of the Landé $g$-factors for the $5d6s\,{}^3D_2$ hyperfine levels of $^{176}\mathrm{Lu}^+$ with fractional inaccuracy $5\times 10^{-7}$. The $g$-factors are obtained from interleaved measurements of the Zeeman splittings between $m=\pm1$ substates in the $^3D_1(F=6)$ and $^3D_2(F=5,\ldots,9)$ manifolds, giving the ratios $r_F = g_F/\bar{g}_6$; the only significant systematic correction is the optical ac Stark shift from off-resonant 804 nm couplings. Combining the five $g_F$ values with the theoretical expressions for hyperfine-mediated $g$-factor corrections and the previously measured nuclear $g$-factor allows the authors to solve for the hyperfine parameters and, using the relation between magnetic-dipole and electric-quadrupole hyperfine couplings, to infer the residual quadrupole moment $\delta\Theta = 1.59(34)\times 10^{-4}\,ea_0^2$ for the $^1S_0 \leftrightarrow {}^3D_2$ hyperfine-averaged clock transition, a shift at the low $10^{-19}$ level.

Load-bearing premise

The inferred residual quadrupole moment assumes the hyperfine coupling to the nearby $^1D_2$ singlet level is negligibly small and that every omitted higher-order correction is no larger than the measured residual parameter $\beta = 1.431(19)\times 10^{-6}$; if either assumption fails, the extracted $\delta\Theta$ could shift by more than the quoted uncertainty.

Editorial extensions

If this is right

The $^3D_2$ clock transition of $^{176}\mathrm{Lu}^+$ can now be assessed with a bounded quadrupole systematic at the low $10^{-19}$ level, putting it on par with the $^3D_1$ transition.
A ratio measurement of the two lutetium clock transitions becomes a practical system-level validation tool, since both transitions' leading systematics are now quantified.
The extracted $\beta^{(1)}_{2,1}$ agrees with the value inferred from the $^3D_1$ measurements, supporting the theoretical link between $g$-factor and quadrupole corrections used here.
Future clock operation should treat the RQM as a known shift, not an unknown systematic, when assessing the $^3D_2$ transition's accuracy budget.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct quadrupole-shift measurement by varying the electric field gradient could confirm $\delta\Theta$ with existing correlation spectroscopy, providing an independent check of the inference.
The residual $\beta$ here is larger than expected from $^3D_1$ contributions, hinting that a joint analysis of both transitions' $g$-factor data could identify a common higher-order source.
Extending the same measurement to other lutetium isotopes, which have different nuclear magnetic moments, would test the predicted scaling of the hyperfine-mediated residual quadrupole moment.
If $\delta\Theta$ is used to correct clock data, the ratio of the two $^{176}\mathrm{Lu}^+$ clock transitions could serve as a real-time diagnostic of electric field gradient or magnetic field drift.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The new thing here is the measurement of the 3D2 hyperfine g-factors in 176Lu+ to about 5e-7 fractional accuracy. That part is solid. The interleaved ratio scheme, QPN analysis, systematic corrections for off-resonant probe shifts, and repeat checks on F=5 and F=7 all look careful and credible. If you work on lutetium clocks, these numbers are directly useful. The paper also extends the authors' earlier 3D1 analysis to infer the residual quadrupole moment for the 1S0–3D2 hyperfine-averaged clock transition, which was previously unquantified. That inference is the softer part, and the authors say so themselves. They rely on beta^(2)_{2,S}=0 with an ad hoc 0.1 beta uncertainty, take gI from a prior paper, and borrow the sum rule from 3D1 to fix beta^(2)_{2,1}. The extracted delta-Theta is therefore not determined by the new g-factor data alone; it carries model assumptions. The stress-test note's worry is fair: if the singlet coupling is larger than assumed, the central value could shift beyond the quoted uncertainty. That said, the paper discloses these assumptions explicitly and treats the result as an estimate. I do not see a load-bearing flaw. There is one concrete editorial problem: the abstract quotes delta Theta = 1.59(34)e-4 while Eq. (15) gives 1.59(38)e-4. The discrepancy is small but it will confuse readers; it should be reconciled before publication. Overall, the measurement claim is robust and the RQM estimate is appropriately hedged. The paper deserves a serious referee, mostly to check the algebra and the way uncertainties are propagated through the model assumptions. I would not desk-reject it. If I were still working on trapped-ion clocks, I would cite the g-factor table. For a general physics audience, the interest is narrow, but for the clock community it is a useful incremental step.

Referee Report

3 major / 4 minor

Summary. The paper reports interleaved measurements of the Landé g-factor ratios r_F = g_F(3D2)/gbar_6(3D1) for F = 5,...,9 of 176Lu+ with fractional inaccuracies around 5e-7, corrects the dominant 804-nm probe-light shift, and converts the ratios to absolute g-factors using the previously measured gbar_6. It then analyzes hyperfine-mediated corrections using the formalism of Ref. [5], projects the measured g-factors onto orthogonal components, neglects singlet-state couplings, and combines the extracted beta parameters with theoretical matrix elements to infer a residual quadrupole moment for the 1S0-3D2 clock transition of deltaTheta = 1.59(38)e-4 ea0^2 (abstract: 1.59(34)). The central measurement claim is supported by interleaved probing, QPN-limited noise analysis, and repeat checks for F = 5 and F = 7; the RQM inference is conditional on several stated approximations.

Significance. If the RQM estimate holds, it bounds the quadrupole clock shift to the low 10^-19 level under typical operating conditions, a useful input for a relatively unexplored clock transition. The g-factor ratios themselves are a precise, independent set of measurements that usefully extend the Lu+ data set and expose hyperfine-mediated corrections. Strengths include the interleaved measurement design, explicit QPN limit with a sqrt(2) statistical penalty, correction of 804-nm ac Stark shifts, and repeatability checks. The RQM inference, however, is not determined by the new data alone: it depends on external beta^(2)_{2,1} from [5], on beta^(k)_{2,S}=0 with an ad hoc 10% uncertainty, and on treating residual beta as a higher-order error bound. These assumptions need stronger support before the headline deltaTheta can be taken at face value.

major comments (3)

[Section IV, after Eq. (12)] The extraction of beta^(2)_{2,1} and hence deltaTheta rests on the assumption beta^(k)_{2,S}=0. The manuscript supports this only by noting that beta^(k)_{1,S} from [5] were a few percent of beta^(k)_{1,2}; no analogous calculation of beta^(2)_{2,S} is shown. The 0.1 beta^(k)_{2,1} uncertainty added later is an assumption, not a demonstrated bound. A concrete sensitivity check makes the concern quantitative: substituting a singlet contribution of 30% of beta^(2)_{2,1} into Eq. (7b) changes the extracted beta^(2)_{2,1} by roughly 40%, moving deltaTheta well outside the quoted uncertainty. Because the new g-factor data cannot identify beta^(2)_{2,S}, the RQM estimate is conditional on an unverified model assumption.
[Abstract and Eq. (15)] The abstract quotes deltaTheta = 1.59(34)e-4 ea0^2, while Eq. (15) gives 1.59(38)e-4 ea0^2. The manuscript does not explain which number is correct. Since this is the headline result, the discrepancy must be resolved; if the abstract is intended, the uncertainty statement in the analysis or in Eq. (15) needs correction.
[Section IV, treatment of higher-order corrections] The treatment of higher-order corrections is ad hoc: after Eq. (11e), beta = 1.431(19)e-6 is larger than anticipated from 3D1 contributions, and the authors "take beta as a bound on each delta g^(2)_F," treating each as zero-mean with variance beta^2, which is equivalent to inflating measurement uncertainties until beta is statistically consistent with zero. This procedure may be a reasonable conservative estimate, but it is not derived from a calculation of omitted terms. The authors should either provide a physical estimate of the largest omitted correction or state explicitly that the quoted uncertainty is a model-dependent allowance rather than a measured bound.

minor comments (4)

[Section III, around Table I] The sentence "we take the statistical uncertainty in the result to be sqrt(2) above the projection noise limit" is missing a word and should read "to be a factor of sqrt(2) above".
[Section IV, after Eq. (12)] There is a typo in "effecively treating each as an error term"; it should be "effectively".
[Section II] The notation "Sect.II" and "Fig 1" should be standardized as "Sect. II" and "Fig. 1".
[Title and text] The rendering "Land´ e" contains an accent-encoding artifact and should be corrected to "Landé".

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the g-factor ratios are direct measurements, and the RQM estimate is a transparent model-dependent extraction using external prior results, not a fit renamed as a prediction.

full rationale

The central g-factor measurements are self-contained: interleaved Zeeman-splitting ratios rF are measured with QPN-limited Allan deviations, corrected for the dominant 804-nm optical ac-Stark shift, and repeated for F=5 and F=7. The gF values are obtained by multiplying these ratios by the previously measured ¯g6, which is a calibration from Ref. [5], not a parameter fitted to the same data. No equation in the paper defines a predicted quantity in terms of the quantity it purports to predict. The RQM inference in Sec. IV is model-dependent rather than circular: it uses the theoretical decomposition of Eq. (6), the prior values gI and Σ β^(2)_{1,J'} from Ref. [5], and the stated assumption β^(k)_{2,S}≈0. Those inputs are external, published results, and the new ratio data genuinely constrain the remaining parameter β^(2)_{2,3} and the consistency of β^(1)_{2,1}. The paper explicitly labels the result an estimate and quotes uncertainties that include an ad hoc 10% model allowance. The fragility of the singlet assumption and the discrepancy between the abstract's δΘ=1.59(34)×10^(-4) and Eq. (15)'s 1.59(38)×10^(-4) are correctness or model risks, not circular reductions. Accordingly no circular step is exhibited and the score is low.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The g-factor ratios are directly measured and are the strongest part of the paper. The RQM inference rests on the theory expansion from the authors' earlier work, on the nuclear g-factor and matrix elements from Refs. [5,10], and on two explicit model assumptions (singlet coupling zero, higher-order terms bounded by β). No new physical entities are postulated.

free parameters (8)

β^(1)_{2,1} = 8.4(1.6) × 10^-4
Hyperfine-mediated magnetic-dipole coupling to the 3D1 fine-structure level, obtained by solving Eq. 6 against the measured g-factors.
β^(1)_{2,S} = 0.0(9) × 10^-4
Coupling to the 1D2 singlet is set to zero, with an uncertainty assigned as 0.1 times β^(1)_{2,1}; this is a stated modeling assumption.
β^(1)_{2,3} = -0.9(2.5) × 10^-4
Magnetic-dipole coupling to the 3D3 level, extracted from the same fit.
β^(2)_{2,1} = -3.1(3) × 10^-5
Electric-quadrupole hyperfine coupling to 3D1; this term most directly sets the residual quadrupole moment.
β^(2)_{2,S} = 0.0(3) × 10^-5
Singlet quadrupole coupling set to zero with an uncertainty of 0.1 times β^(2)_{2,1}.
β^(2)_{2,3} = 9.1(3.5) × 10^-5
Electric-quadrupole coupling to 3D3, extracted from the fit.
gJ = 1.156180(20)
Electronic g-factor of the 3D2 state obtained from the fit; part of the set of inferred parameters.
β (residual higher-order term) = 1.431(19) × 10^-6
Residual from the four-parameter fit after orthogonal decomposition; used as a bound on each δg^(2)_F error term.

assumptions (6)

domain assumption The first-order perturbation expansion for hyperfine-mediated g-factor corrections, Eqs. 2-5 from Ref. [5], is valid for the 3D2 manifold.
All β parameters are extracted by solving these equations; if second-order contributions are not bounded by the residual β, the extracted values shift.
domain assumption The nuclear g-factor gI = -2.435047(16)×10^-4 from Ref. [5] is accurate.
Used as a fixed input to solve Eq. 6; its uncertainty is not propagated into the RQM uncertainty.
domain assumption The reduced matrix elements in Table II, taken from Refs. [5,10], are accurate to within a few percent.
Eq. 14 converts fitted β parameters into quadrupole moments; the paper adds a 10% allowance for this uncertainty.
ad hoc to paper Coupling to the 1D2 singlet state is negligible, β^(k)_{2,S} ≈ 0.
Section IV, after Eq. (12); supported by expectation that singlet coupling is small, not by a direct measurement.
ad hoc to paper Higher-order corrections δg^(2)_F are bounded by the residual β from Eq. 11e and are treated as zero-mean errors with variance β².
Section IV; this inflates the rF uncertainties and converts a model residual into a statistical error.
domain assumption Hyperfine-averaging over a fixed m_F leaves an effective J=0 level with only a small residual quadrupole moment.
Background framework from Ref. [1]; the RQM concept depends on this cancellation being imperfect but describable by a single δΘ.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Lande g-factor measurements for the 5d6s 3D2 hyperfine levels of 176Lu+." pith.science (2026). https://pith.science/paper/LSNLWYXH

@misc{pith2026250716292,
  author       = {Pith},
  title        = {Pith review of: Lande g-factor measurements for the 5d6s 3D2 hyperfine levels of 176Lu+},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LSNLWYXH}},
  note         = {Machine review of arXiv:2507.16292}
}

abstract

We report measurements of the Lande g-factors for the 5d6s $^3$D$_2$ hyperfine levels of $^{176}$Lu$^+$ to a fractional inaccuracy of $5\times 10^{-7}$. Combining these measurements with theoretical calculations allows us to estimate hyperfine-mediated modifications to the quadrupole moments for each state and infer a value of $\delta\Theta = 1.59(34)\times 10^{-4} \,ea_0^2$ for the residual quadrupole moment of the $^1S_0\leftrightarrow{^3}D_2$ hyperfine-averaged clock transition.

Figures

Figures reproduced from arXiv: 2507.16292 by the authors.

**Figure 2.** FIG. 2. Fractional Allan deviation of the ratios [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[5]

Beloy, D

K. Beloy, D. R. Leibrandt, and W. M. Itano, Hyperfine- mediated electric quadrupole shifts in Al + and In + ion clocks, Physical Review A 95, 043405 (2017)

work page 2017
[1]

Lasers at 350, 622 and 895 nm initialize population in the 3D1 state. A laser at 646 nm, which has three inde- pendent frequency components coupling 3D1(F = 6, 7, 8) to 3P0(F = 7), provides Doppler cooling and state detec- tion for the 3D1 state with fluorescence collected onto a single photon counting module (SPCM). Switching the 3D1(F = 7) to 3P0(F = 7)...

work page Pith review arXiv 2025
[2]

M. D. Barrett, Developing a field independent frequency reference, New Journal of Physics 17, 053024 (2015)

work page 2015
[3]

Zhiqiang, K

Z. Zhiqiang, K. J. Arnold, R. Kaewuam, and M. D. Barrett, 176Lu+ clock comparison at the 10 -18 level via correlation spectroscopy, Science Advances 9, eadg1971 (2023)

work page 2023
[4]

K. J. Arnold, S. Bustabad, Z. Qi, Q. Qichen, Z. Zhang, Z. Zhao, and M. D. Barrett, Validating a lutetium fre- quency reference., in Journal of Physics: Conference Se- ries, Vol. 2889 (IOP Publishing, 2024) p. 012040

work page 2024
[6]

Zhiqiang, K

Z. Zhiqiang, K. J. Arnold, R. Kaewuam, M. S. Safronova, and M. D. Barrett, Hyperfine-mediated effects in a Lu + optical clock, Physical Review A 102, 052834 (2020)

work page 2020
[7]

K. J. Arnold, N. Jayjong, M. L. D. Kang, Q. Qichen, Z. Zhang, Q. Zhao, and M. D. Barrett, Enhanced mi- cromotion compensation using a phase-modulated light field, Physical Review A 110, 033115 (2024)

work page 2024
[8]

Myerson, D

A. Myerson, D. Szwer, S. Webster, D. Allcock, M. Curtis, G. Imreh, J. Sherman, D. Stacey, A. Steane, and D. Lu- cas, High-fidelity readout of trapped-ion qubits, Physical Review Letters 100, 200502 (2008)

work page 2008

Show all 16 references

[9]

H. C. J. Gan, G. Maslennikov, K.-W. Tseng, T. R. Tan, R. Kaewuam, K. J. Arnold, D. Matsukevich, and M. D. Barrett, Oscillating-magnetic-field effects in high-precision metrology, Physical Review A 98, 032514 (2018)

2018
[10]

K. J. Arnold, R. Kaewuam, S. R. Chanu, T. R. Tan, Z. Zhang, and M. D. Barrett, Precision measurements of the 138Ba+ 6s 2S1/2 − 5d 2D5/2 clock transition, Physical Review Letters 124, 193001 (2020)

2020
[11]

K. J. Arnold, E. Hajiyev, E. Paez, C. H. Lee, M. D. Barrett, and J. Bollinger, Prospects for atomic clocks based on large ion crystals, Physical Review A92, 032108 (2015)

2015
[12]

E. Paez, K. J. Arnold, E. Hajiyev, S. G. Porsev, V. A. Dzuba, U. I. Safronova, M. S. Safronova, and M. D. Bar- rett, Atomic properties of Lu +, Physical Review A 93, 042112 (2016)

2016
[13]

V. I. Yudin, A. V. Taichenachev, C. W. Oates, Z. W. Barber, N. D. Lemke, A. D. Ludlow, U. Sterr, C. Lisdat, and F. Riehle, Phys. Rev. A. 82, 011804(R) (2010)

2010
[14]

Huntemann, B

N. Huntemann, B. Lipphardt, M. Okhapkin, C. Tamm, E. Peik, A. V. Taichenachev, and V. I. Yudin, Phys. Rev. Lett. 109, 213002 (2012)

2012
[15]

K. J. Arnold, R. Kaewuam, A. Roy, T. R. Tan, and M. D. Barrett, Blackbody radiation shift assessment for a lutetium ion clock, Nature communications 9, 1650 (2018)

2018
[16]

P. Dube, A. A. Madej, M. Tibbo, and J. E. Bernard, High-accuracy measurement of the differential scalar po- larizability of a 88Sr+ clock using the time-dilation effect, Phys. Rev. Lett. 112, 173002 (2014)

2014

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [5]

Beloy, D

K. Beloy, D. R. Leibrandt, and W. M. Itano, Hyperfine- mediated electric quadrupole shifts in Al + and In + ion clocks, Physical Review A 95, 043405 (2017)

work page 2017

[2] [1]

Lasers at 350, 622 and 895 nm initialize population in the 3D1 state. A laser at 646 nm, which has three inde- pendent frequency components coupling 3D1(F = 6, 7, 8) to 3P0(F = 7), provides Doppler cooling and state detec- tion for the 3D1 state with fluorescence collected onto a single photon counting module (SPCM). Switching the 3D1(F = 7) to 3P0(F = 7)...

work page Pith review arXiv 2025

[3] [2]

M. D. Barrett, Developing a field independent frequency reference, New Journal of Physics 17, 053024 (2015)

work page 2015

[4] [3]

Zhiqiang, K

Z. Zhiqiang, K. J. Arnold, R. Kaewuam, and M. D. Barrett, 176Lu+ clock comparison at the 10 -18 level via correlation spectroscopy, Science Advances 9, eadg1971 (2023)

work page 2023

[5] [4]

K. J. Arnold, S. Bustabad, Z. Qi, Q. Qichen, Z. Zhang, Z. Zhao, and M. D. Barrett, Validating a lutetium fre- quency reference., in Journal of Physics: Conference Se- ries, Vol. 2889 (IOP Publishing, 2024) p. 012040

work page 2024

[6] [6]

Zhiqiang, K

Z. Zhiqiang, K. J. Arnold, R. Kaewuam, M. S. Safronova, and M. D. Barrett, Hyperfine-mediated effects in a Lu + optical clock, Physical Review A 102, 052834 (2020)

work page 2020

[7] [7]

K. J. Arnold, N. Jayjong, M. L. D. Kang, Q. Qichen, Z. Zhang, Q. Zhao, and M. D. Barrett, Enhanced mi- cromotion compensation using a phase-modulated light field, Physical Review A 110, 033115 (2024)

work page 2024

[8] [8]

Myerson, D

A. Myerson, D. Szwer, S. Webster, D. Allcock, M. Curtis, G. Imreh, J. Sherman, D. Stacey, A. Steane, and D. Lu- cas, High-fidelity readout of trapped-ion qubits, Physical Review Letters 100, 200502 (2008)

work page 2008