{"id":"3ca7ef73-c3d9-4d3d-a3b3-496a113bd5b7","arxiv_id":"2502.05386","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"The 3P2 to 3P0 electric quadrupole transition in Ni12+ was computed at 20081(10) cm^-1 and measured at 20078.984(10) cm^-1, confirming the ab initio prediction and enabling an optical clock.","lead":"Physicists predicted the energy of an ultra-narrow, strongly forbidden transition in nickel ions to within 2 cm^-1 and then found the transition in the lab in hours. The result enables a new optical clock with a natural linewidth of 8 mHz and validates a 16-electron atomic-structure calculation method for future highly charged ion clocks.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'Estimate full CI' extrapolation has no quantified uncertainty; if the empirical weight-energy relation bends in the omitted tail, the 10 cm^-1 uncertainty budget is understated despite the small -3 cm^-1 nominal correction.","rationale":"The reader's weakest_assumption correctly identifies the empirical full-CI extrapolation in the 'Theory: pure CI method' section as a soft spot. I partially agree: the direct correction for 3P0 is only -3 cm^-1, so as a numerical contribution it is minor, but the lack of a quantified uncertainty for that estimate is more load-bearing than the reader's treatment suggests. The 10 cm^-1 total uncertainty is justified by the difference between two methods and by the 3P1 comparison, but not explicitly by the uncertainty of the full-CI tail estimate. The observed 1S0 error of -66 cm^-1 shows that the method's accuracy is not uniformly at the 0.05% level, and the 3P0 agreement to 2 cm^-1 could be an artifact of error cancellation rather than a demonstration of the uncertainty budget. The proposed test, sampling configurations from the omitted tail or validating the linear relation on a small solvable system, would directly assess whether the extrapolation is trustworthy. Since the reader's verdict is already CONDITIONAL and asks for re-evaluation of uncertainties, my concern does not change the verdict, hence UNCHANGED.","tokens_in":8914,"tokens_out":11973,"duration_ms":112508,"concrete_test":"Perform a pure CI run for Ni12+ that includes a statistically significant random sample of configurations from the omitted tail (selected via the neural-network tool), and measure the 3P0 energy shift. Compare this shift to the extrapolated -3 cm^-1 prediction. If the observed shift deviates by more than ~5 cm^-1, the full-CI estimate's uncertainty is larger than assumed, and the 10 cm^-1 uncertainty budget must be revised. Alternatively, validate the linear weight-energy relation on a smaller model system where full CI is exactly solvable and the extrapolation can be checked against the exact correction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the 16-electron CI prediction has 0.05% (10 cm^-1) uncertainty rests on the reliability of the 'Estimate full CI' procedure (Table II). The paper estimates the contribution of all omitted configurations by a linear relation between configuration weight and energy contribution, tested only up to 9500 reference configurations (Theory: pure CI method). For 3P0 this correction is only -3 cm^-1, so a direct error in the slope would have small impact. However, the extrapolation is over billions of configurations in the tail, and the uncertainty of the estimate is not quantified. If the linear relation becomes nonlinear for very small weights, the cumulative correction could be many times larger than -3 cm^-1, potentially exceeding the 10 cm^-1 budget. The agreement with experiment (2 cm^-1) suggests the error is small for this specific level, but the 1S0 level shows an error of -66 cm^-1 (Table II), demonstrating that the method's accuracy is strongly level-dependent and the 3P0 success may rely on error cancellation. The broader claim that the method 'qualifies for future calculations' depends on the uncertainty being reliable, not just on one fortuitous agreement.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper predicts the energy of the strongly forbidden 3s2 3p4 3P2 → 3s2 3p4 3P0 electric quadrupole transition in Ni12+ as 20081(10) cm−1, using two independent many-body approaches: a hybrid CI+all-order method and a pure 16-electron CI calculation. The prediction was then used to search for and find the transition experimentally at 20078.984(10) cm−1, only 2 cm−1 from the theoretical value. The authors claim that this level of agreement, with a stated 0.05% uncertainty for a 16-electron system, is unprecedented and qualifies their method for future atomic-structure calculations.","tokens_in":9131,"tokens_out":5702,"duration_ms":54250,"significance":"If the uncertainty estimate holds, this is a landmark result: it demonstrates that a full 16-electron configuration-interaction calculation can predict a state energy to about 0.05% accuracy, and the experimental confirmation in a few hours is a powerful validation. The paper also contributes an open-source parallel CI package, a neural-network-based configuration selection tool, and a two-method cross-check that agree to 5 cm−1 for the target level. The main weakness is the unquantified extrapolation used to estimate the residual full-CI contribution, which is central to the claimed 10 cm−1 uncertainty budget.","major_comments":[{"comment":"The estimate of the contribution of all omitted configurations, labeled 'Estimate full CI', is obtained from a linear relation between configuration weight and energy contribution, but no uncertainty is assigned to this extrapolation. The 10 cm−1 uncertainty budget appears to rely on the difference between the two methods and the agreement for the 3P1 level, and does not explicitly include the error in this extrapolation. The 1S0 level in Table II shows a discrepancy of −66 cm−1 from experiment, with an 'Estimate full CI' correction of only −24 cm−1, demonstrating that the method's accuracy is strongly level-dependent. Therefore, the small −3 cm−1 correction for 3P0 does not by itself guarantee that the extrapolation error is small. Please provide a quantitative bound for the extrapolation error, for example by testing the linear relation on additional levels or by varying the slope within a plausible range, or explicitly state that the 10 cm−1 budget excludes this source and justify why it is negligible.","section":"Theory: pure CI method (Table II, column 'Estimate full CI')"},{"comment":"The description 'a relative weight of 0.01% leading to a correction of about −11 cm−1' is ambiguous: it is unclear whether this refers to the contribution of a single configuration or to the integrated contribution of all configurations with weights below a threshold. The manuscript does not provide the actual fitting parameters, the range of weights over which the linear relation was validated, or a plot of the data supporting the 'linear dependence (similar for all four levels)' claim. Without this information, the 'Estimate full CI' column in Table II is not reproducible and its reliability cannot be independently assessed.","section":"Theory: pure CI method, paragraph on 'Estimate full CI'"},{"comment":"The conclusion states that the work demonstrates 'full convergence of a 16-electron CI computation', but this overstates the status of the calculation: the largest part of the configuration space is not computed directly but estimated via the extrapolated linear relation. A more precise wording would be 'convergence with respect to explicitly included configurations and an estimated residual contribution', which would be consistent with the actual procedure described in the text.","section":"Conclusion"}],"minor_comments":[{"comment":"Several formulas are typeset incorrectly in the text as provided, e.g., '3s23p4' instead of '3s^2 3p^4' and 'Ni12+' instead of 'Ni$^{12+}$'. These should be corrected in the final version.","section":"Throughout"},{"comment":"Reference [12] is not a standard citation but a footnote explaining the treatment of k and l partial waves. It should be renumbered as a footnote or integrated into the main text.","section":"References"},{"comment":"The column groups such as '17 spd fg', 'l >6', 'Extra conf.', and 'Estimate full CI' are not fully defined in the captions. Please add a sentence explaining the meaning of the grouped columns, e.g., that '17 spd fg' refers to orbitals with n up to 17 and partial waves s, p, d, f, g, and that the contributions are incremental from the previous column.","section":"Table I and Table II"},{"comment":"The caption 'Illustration of the basis set upscale' is vague. The figure should describe what the blocks represent (e.g., principal quantum number ranges and partial waves) and how it relates to the convergence procedure.","section":"Figure 1"}],"recommendation":"major_revision","confidential_remarks":"The experimental confirmation is impressive and strongly supports the central prediction. However, the uncertainty budget, which is a key part of the paper's claim of 0.05% accuracy, is not fully justified because the full-CI residual estimate lacks any quantified uncertainty. This is a fixable issue: the authors could provide additional validation or a conservative bound on the extrapolation. The paper is otherwise well within the scope of the journal and likely to be an important contribution once the uncertainty analysis is strengthened."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a read if you care about HCI clocks or CI calculations for mid-Z ions. The new result is both theoretical and experimental: a 16-electron CI+all-order prediction of the 3P2-3P0 E2 transition at 20081(10) cm^-1, and the first direct laser excitation of that transition, measured at 20078.984(10) cm^-1. Two independent theoretical treatments agree to 5 cm^-1, and nothing is fitted to the target energy. That is a real achievement and the 0.05% uncertainty claim is believable for this level.\n\nWhat is genuinely new is the convergence strategy: separating the CI into additive classes, pushing to 9.6 million configurations, and using a neural-network configurator to manage the memory. They also give a quantitative estimate of the residual full-CI contribution, which they say has never been done before. The experimental search logic (off-resonant ODF detection, scanning 100 GHz in six hours) is elegant and the detection scheme is convincing.\n\nThe soft spots are modest. The 'Estimate full CI' column relies on an empirical linear relation between configuration weight and energy shift, extrapolated over billions of omitted configurations, and no uncertainty is attached to that extrapolation. For the 3P0 level the correction is only -3 cm^-1, so even a sizeable error in the slope would not break the 10 cm^-1 budget. But the 1S0 level shows a -66 cm^-1 deviation against experiment, which tells you the method's accuracy is level-dependent and the 3P0 agreement may involve some error cancellation. The paper is honest about this: they present the 3P0 uncertainty as conservative based on the 3P1 comparison. I think that is fair, but a referee should ask whether the 'Estimate full CI' uncertainty can be bounded rather than assumed.\n\nThe citation pattern is fine; the QED corrections come from previous work with 10% accuracy, and the NIST values are used as benchmarks, not fitted. The central measurement is independently reproducible in principle and the code is released. For someone in the field, this is a solid contribution: it identifies a workable clock transition and demonstrates that the theory pipeline can guide a fast experimental search. I would send it to peer review; the only substantive request would be a clearer quantification of the full-CI residual uncertainty, maybe by testing the linear relation on smaller subsets.","headline":"First optical excitation of the Ni12+ clock transition, with a 10 cm^-1 ab initio prediction that landed 2 cm^-1 from the measured value; a strong letter with one modest caveat about the full-CI residual estimate.","tokens_in":9739,"tokens_out":1527,"would_cite":true,"duration_ms":14707,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The energy of the ultra-narrow 3P2 → 3P0 electric quadrupole transition in Ni12+ is predicted at 20081(10) cm^-1 and measured at 20078.984(10) cm^-1, an agreement of 2 cm^-1.","keywords":["Ni12+","highly charged ion","optical clock","electric quadrupole transition","configuration interaction","coupled-cluster","atomic physics","frequency metrology"],"falsifier":"If another independent computation of the 3P0 energy, or a future more precise measurement, deviates from 20081 $cm^{-1}$ by more than 10 $cm^{-1}$, the claimed uncertainty and the extrapolation of the weight-energy relation would be falsified. A direct check would be an independent CI calculation with a different code or basis that does not use the same linear extrapolation.","tokens_in":8699,"feed_emoji":"⚛️","tokens_out":4350,"duration_ms":37377,"temperature":0.7,"pith_summary":"This paper predicts the energy of the strongly forbidden 3s2 3p4 3P2 → 3s2 3p4 3P0 electric quadrupole transition in the Ni12+ ion, a candidate for a highly stable optical clock with a natural linewidth of only 8 mHz. Using two independent theoretical methods — a hybrid approach combining configuration interaction with coupled-cluster and a pure 16-electron CI calculation — the authors obtain 20081(10) $cm^{-1}$, an uncertainty of 0.05%. Working from this prediction, they located the transition experimentally in six hours, measuring 20078.984(10) $cm^{-1}$, only 2 $cm^{-1}$ away. The close agreement for a 16-electron system is presented as evidence that full CI calculations can guide searches for clock transitions in other highly charged ions.","feed_headline":"Ni12+ clock transition found 2 cm-1 from prediction","feed_subtitle":"A 0.05% ab initio prediction let researchers locate the 8-mHz-wide line in six hours of scanning.","key_machinery":"The load-bearing tool is a pure CI calculation for all 16 electrons of the Ni12+ ground configuration, broken into additive contributions from excitations of the outer six electrons, inner ten electrons, extra reference configurations, and an estimated residual full-CI term. The residual is estimated through a linear relation between a configuration's relative weight and its energy contribution, extrapolated from computed configurations. A hybrid CI+all-order method with a coupled-cluster effective Hamiltonian provides an independent cross-check, and a neural-network-based configuration selection reduced memory and time needs.","core_discovery":"The central claim is that a fully converged 16-electron configuration-interaction calculation can predict the energy of the forbidden 3P2 → 3P0 clock transition in Ni12+ with an uncertainty of only 10 $cm^{-1}$, sufficient to locate the transition with rapid laser scans. The prediction, 20081(10) $cm^{-1}$, differs from the measured value 20078.984(10) $cm^{-1}$ by 2 $cm^{-1}$. The authors argue this validates their method for other complex atomic systems and enables a Ni12+ optical clock.","pith_inferences":["The neural-network configuration-selection algorithm, developed for this calculation, could be applied to other large CI problems where the full expansion is intractable.","The measured 2 cm^-1 discrepancy suggests the remaining uncertainty may be dominated by the estimated full-CI residual or QED corrections; a systematic study of similar transitions could refine these estimates.","The same fast-scanning search strategy, combined with precise predictions, could be used to find other forbidden transitions in HCIs that were previously considered too weak to search for."],"forward_implications":["The Ni12+ clock transition is now accessible for quantum-logic spectroscopy and the development of a high-precision optical clock.","The demonstrated accuracy of the 16-electron CI method supports its application to other highly charged ions where clock transitions have not yet been located.","The separation of CI into additive contributions and the weight-energy estimate extends the practical size of CI calculations beyond the current limits of exact diagonalization."],"supporting_citations":[{"why":"First proposed the 3P2 → 3P0 transition in Ni12+ as a clock candidate, establishing the transition's narrow width and suitability.","marker":"[6]"},{"why":"Supplies the CI+all-order hybrid method with the coupled-cluster effective Hamiltonian used for core-valence correlations.","marker":"[10]"},{"why":"Provides the recurrent basis-set construction and CI approach used in the pure CI calculation.","marker":"[11]"},{"why":"The optimized parallel CI code made the 9.6-million-configuration computation feasible.","marker":"[14]"},{"why":"Method for QED corrections used in both calculations.","marker":"[15]"},{"why":"Method for three-electron interaction corrections used in the CI+all-order approach.","marker":"[16]"},{"why":"The fast laser frequency scanning technique that allowed the search to cover 100 GHz in six hours.","marker":"[8]"},{"why":"Provides the experimental 3P1 energy and other benchmarks used to validate the calculations.","marker":"[7]"}],"fun_headline_variants":["Ni12+ clock line found just 2 cm−1 from prediction","8-mHz transition in Ni12+ pinpointed by 0.05% theory","Clock search in Ni12+ ends in hours, not years","16-electron theory nails Ni12+ clock transition"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The estimate of the residual full-CI contribution assumes that the linear weight–energy relation found for computed configurations holds for the far more numerous configurations that cannot be computed directly, with the 3P0 correction being small (-3 $cm^{-1}$) but not independently verified.","fun_headline_variants_meta":{"raw":{"variants":["Ni12+ clock line found just 2 cm−1 from prediction","8-mHz transition in Ni12+ pinpointed by 0.05% theory","Clock search in Ni12+ ends in hours, not years","16-electron theory nails Ni12+ clock transition"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000516,"raw_usage":{"total_tokens":2482,"prompt_tokens":899,"completion_tokens":1583,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":515,"completion_tokens_details":{"reasoning_tokens":1508}},"tokens_in":515,"tokens_out":1583,"duration_ms":14250,"temperature":1.0,"reasoning_tokens":1508,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T19:32:23.307631+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"If another independent computation of the 3P0 energy, or a future more precise measurement, deviates from 20081 $cm^{-1}$ by more than 10 $cm^{-1}$, the claimed uncertainty and the extrapolation of the weight-energy relation would be falsified. A direct check would be an independent CI calculation with a different code or basis that does not use the same linear extrapolation.","supporting_citations":[{"cited_title":"Yu and B","cited_arxiv_id":null,"evidence_quote":"First proposed the 3P2 → 3P0 transition in Ni12+ as a clock candidate, establishing the transition's narrow width and suitability."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the recurrent basis-set construction and CI approach used in the pure CI calculation."},{"cited_title":"Cheung, M","cited_arxiv_id":null,"evidence_quote":"The optimized parallel CI code made the 9.6-million-configuration computation feasible."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Method for QED corrections used in both calculations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Method for three-electron interaction corrections used in the CI+all-order approach."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The fast laser frequency scanning technique that allowed the search to cover 100 GHz in six hours."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the experimental 3P1 energy and other benchmarks used to validate the calculations."}],"review_version":1}