{"id":"2c1cbd07-dafa-4827-884b-dd6ba5d0c030","arxiv_id":"2502.01112","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"New relativistic calculations predict two forbidden optical clock transitions in Ir17+ and a full set of atomic properties, but with large energy uncertainties and no experimental confirmation yet.","lead":"Two theoretical methods predict that the highly charged ion Ir17+ has two very slow optical transitions, around 346 nm and 833 nm, that could serve as future atomic clock ticks. The same predictions give the ion an unusually strong sensitivity to changes in the fine-structure constant, making it a candidate for tests of fundamental physics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract assigns transition wavelengths 346 nm and 833 nm to the 4f14 1S0 and 4f12 5s2 3H6 states respectively, but Table I energies imply the opposite: 833 nm for 1S0 and 347 nm for 3H6, so the central clock-transition identification is internally inconsistent.","rationale":"The paper's core contribution is the identification of two specific optical clock transitions in Ir17+. The wavelengths are the most direct observable identifiers. An internal swap of the assigned wavelengths means that, as printed, the abstract tells experimentalists to look for a 346 nm line from 1S0 (actually at 833 nm) and an 833 nm line from 3H6 (actually at 347 nm). The associated 'excellent clock features' — especially the negative Delta-alpha and magic RF frequency — are then attached to the wrong transition, since Table III shows Delta-alpha < 0 only for 1S0 (833 nm) and Delta-alpha > 0 for 3H6 (347 nm). This is a definite factual error in the central claim, independent of any quantum-electrodynamics uncertainties. While the underlying calculations may be correct, the presented wavelengths cannot be used to guide experiment without correction. The QED omission noted by the reader is a legitimate secondary concern, but it is an uncertainty-estimate issue; the wavelength swap is a clear inconsistency that must be resolved before the central claim can be accepted. Therefore the paper should remain conditionally accepted, pending correction of the wavelength assignments and a check of the SM table.","tokens_in":33158,"tokens_out":9421,"duration_ms":87321,"concrete_test":"Compute transition wavelengths from the KRCI energies in Table I: lambda = 10^7 / E. For 4f14 1S0: 10^7/12006 = 833 nm; for 4f12 5s2 3H6: 10^7/28848 = 347 nm. Compare with the abstract's 346(30) nm and 833(100) nm; if they are swapped, correct the abstract and introduction and re-examine which transition carries the negative Delta-alpha. Additionally, reconcile Table XV's 664 nm and 368 nm with Table I; any residual discrepancy indicates a typo or error that must be fixed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim in the abstract and introduction is that the 4f13 5s 3Fo4 -> 4f14 1S0 transition has wavelength ~346(30) nm and the 4f13 5s 3Fo4 -> 4f12 5s2 3H6 transition has wavelength ~833(100) nm. However, the KRCI energies in Table I place 4f14 1S0 at 12006(1392) cm^-1 and 4f12 5s2 3H6 at 28848(2117) cm^-1. Using lambda = 10^7/E (cm^-1) yields 833 nm for 1S0 and 347 nm for 3H6 — the reverse of the abstract. The SM transition table (Table XV) lists yet different values (664 nm and 368 nm for these transitions), compounding the inconsistency. Consequently, the abstract assigns the negative differential polarizability (which Table III gives only for 1S0) to the 346 nm transition, while actually the 346 nm transition goes to 3H6 and has positive Delta-alpha. The magic RF trap frequency calculation in Sec. III C therefore uses a frequency for a transition whose wavelength may be misidentified. This internal contradiction directly undermines the paper's headline identification of the clock transitions and their properties.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports KRCI and FSCC calculations of Ir17+ excitation energies and properties under the Dirac-Coulomb-Gaunt Hamiltonian. It proposes two forbidden optical clock transitions from the 4f13 5s 3Fo4 ground state to the 4f14 1S0 and 4f12 5s2 3H6 states, with predicted wavelengths of about 346 nm and 833 nm, and claims that the negative differential polarizability of one transition enables a magic radio-frequency trap frequency while the two transitions have opposite Kα sensitivities. The manuscript also tabulates g-factors, lifetimes, polarizabilities, quadrupole moments, hyperfine constants, field-shift factors, and related properties for the low-lying states.","tokens_in":33535,"tokens_out":11940,"duration_ms":124293,"significance":"If the transition identification and properties are correct, Ir17+ would offer two long-lived forbidden optical transitions with useful features for HCI clocks and for searches of α-variation, including a possible magic RF trap frequency and opposite sensitivity coefficients. The paper's systematic use of multiple basis sets, explicit reference-state testing, and KRCI/FSCC cross-checks, together with the detailed supplementary tables, is a genuine strength. However, the headline identification is internally inconsistent, the two methods agree only at the one-sigma level for the clock states, and the error budget does not address omitted QED effects; these issues must be resolved before the claimed reliability can be accepted.","major_comments":[{"comment":"The central clock-transition identification is internally inconsistent. The abstract assigns 346(30) nm to the 3Fo4 -> 4f14 1S0 transition and 833(100) nm to the 3Fo4 -> 4f12 5s2 3H6 transition, but Table I places 1S0 at 12006(1392) cm^-1 and 3H6 at 28848(2117) cm^-1, which by lambda = 10^7/E (cm^-1) corresponds to 833 nm and 347 nm, respectively. SM Table XV gives yet another set of values, 664 nm and 368 nm, for the same two transitions. Section III C then attributes the negative differential polarizability (which Table III gives only for 1S0) to the 346-nm transition and computes the magic RF frequency from it, even though Table III assigns the 346-nm transition to 3H6 with positive Delta-alpha. Until this assignment is corrected and the numerical sets are reconciled, the headline clock claims and the magic-frequency estimate cannot be evaluated.","section":"Abstract; Sec. III A; Sec. III C; Table I; SM Table XV"},{"comment":"The claimed excellent agreement between KRCI and FSCC is overstated. For the two clock states, the FSCC final values are 10203(3563) cm^-1 for 4f14 1S0 and 27445(3007) cm^-1 for 4f12 5s2 3H6, while KRCI gives 12006(1392) cm^-1 and 28848(2117) cm^-1, respectively; the central values differ by about 1800 and 1400 cm^-1. The FSCC final values are obtained by adding Delta_T and Delta_basis corrections estimated at different truncation levels (SDT with the 2-zeta basis, SD with the 3-zeta/4-zeta bases), and the additivity of these separately estimated corrections is an assumption that is not tested. Given that the method differences are comparable to the quoted uncertainties, the robustness claim should either be demonstrated quantitatively or replaced by a more conservative statement.","section":"Sec. III A; SM Table V"},{"comment":"The Dirac-Coulomb-Gaunt Hamiltonian omits QED corrections (vacuum polarization and self-energy) and the frequency-dependent Breit interaction. For Z=77, these omitted contributions can be comparable to the 1000-3600 cm^-1 uncertainties quoted in Table I, and the manuscript provides no estimate of their size. A defensible error budget for candidate clock transitions needs at least a model estimate of these shifts or an explicit argument for their smallness, especially since the excited configurations differ in 4f/5s occupations and would not be expected to cancel.","section":"Sec. II A, Eq. (1)"}],"minor_comments":[{"comment":"The formula for nu_magic is dimensionally unclear as written; please state whether nu_magic is a cyclic frequency or an angular frequency, give the SI conversions used for Delta-alpha, and show how the quoted 500 x 2pi MHz follows from the corrected transition frequency.","section":"Sec. III C"},{"comment":"The columns for K_alpha and the LLI reduced matrix elements appear misaligned in the current rendering; please reformat the table and verify that the text's K_alpha values (0.7 and 1.9) are assigned to the intended states.","section":"Table II"},{"comment":"The section heading 'The field-field calculation' should read 'The finite-field calculation'.","section":"Sec. II D"},{"comment":"Reference [53] lists the year 2004, but the volume and page numbers (328, 109151) indicate the year should almost certainly be 2024.","section":"References"},{"comment":"The statement that the dyall.aae3z basis set 'tends to underestimate energy values' is based on a limited 3-zeta/4-zeta comparison; please clarify whether this is intended as a general observation or is specific to the present calculations.","section":"Sec. III A"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely and useful system, and the underlying calculations are standard and carefully documented. However, the internal contradiction in the abstract's transition assignment is load-bearing and cannot be treated as a mere typographical issue. I would encourage the editor to seek a revision that corrects the assignment, reconciles the energy tables, and separates claimed accuracy from the systematic omissions of QED and extrapolation additivity."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful core here is the property package: KRCI and FSCC values for polarizabilities, quadrupole moments, hyperfine constants, isotope-shift factors, and the magic RF frequency analysis for Ir17+. The low-lying 3Fo3 level agrees with EBIT data to about 100 cm^-1, and the supplementary material is transparent about reference-state choices and basis-set convergence. That is real, citable work if the inconsistencies are fixed.\n\nThe soft spots are serious. The abstract and introduction assign 346 nm to the 4f14 1S0 transition and 833 nm to the 4f12 5s2 3H6 transition, but Table I gives KRCI energies of 12006 cm^-1 for 1S0 and 28848 cm^-1 for 3H6, which correspond to 833 nm and 347 nm respectively. The Supplementary Table XV lists different wavelengths again (664 nm and 368 nm). So the headline identification is internally inconsistent, and the negative differential polarizability that supports the magic RF trap is attributed to the wrong transition. That directly affects the clock proposal.\n\nSecond, the two methods do not agree as well as claimed. KRCI and FSCC differ by roughly 1400–1800 cm^-1 on the two clock states, and the FSCC final values carry 3000–3600 cm^-1 uncertainties after large extrapolations from triple-excitation and basis-set corrections. Calling this \"excellent agreement\" overstates it. The energies are not fitted to the predicted transitions, so there is no circularity, but the precision is not yet at the level the abstract implies.\n\nThird, the Dirac-Coulomb-Gaunt Hamiltonian omits QED and the full Breit interaction. For Z=77 these effects can plausibly shift transition energies by amounts comparable to the quoted 1400–3600 cm^-1 uncertainties, and the paper gives no estimate. That is a legitimate gap for a heavy-ion clock candidate.\n\nBottom line: the paper is a competent computational study that deserves a serious referee, but only after the authors correct the wavelength assignment, reconcile the transition table with the energy table, and either add QED estimates or temper the reliability claims. I would not cite it in its current form, but I would send it to peer review with a request for major revision.","headline":"A broadly useful property set for Ir17+ clock candidates, but the abstract swaps the two clock-transition wavelengths, which flips the sign of the magic-trap claim and needs correction before the paper is reliable.","tokens_in":34049,"tokens_out":2449,"would_cite":false,"duration_ms":26628,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Two forbidden transitions in the Ir17+ ion are predicted to serve as optical clock references, at 346 nm and 833 nm.","keywords":["Ir17+","highly charged ion optical clock","forbidden optical transitions","relativistic configuration interaction","Fock-space coupled cluster","fine-structure constant variation","magic radiofrequency trap","atomic polarizability"],"falsifier":"A spectroscopic search in an electron-beam ion trap for the two forbidden lines would settle the claim: if no transitions appear near the predicted 346 nm and 833 nm positions, or if the $4f^{14} {}^1S_0$ level is found far outside the quoted $12006 \\pm 1392$ cm$^{-1}$ window, then the two clock-candidate transitions are refuted.","tokens_in":32943,"feed_emoji":"⏱️","tokens_out":10136,"duration_ms":94934,"temperature":0.7,"pith_summary":"This paper predicts that the highly charged ion Ir$^{17+}$ offers two forbidden optical transitions suitable as clock references: from the $4f^{13}5s$ ground state to the $4f^{14} {}^1S_0$ state at about 346 nm, and to the $4f^{12}5s^2 {}^3H_6$ state at about 833 nm. The predictions come from two independent relativistic many-body methods, configuration interaction and Fock-space coupled cluster, whose results agree once basis-set and triple-excitation corrections are applied. If correct, Ir$^{17+}$ would provide narrow, long-lived clock lines whose negative differential polarizability allows a magic radiofrequency trap and whose opposite sensitivity coefficients make their frequency ratio a sensitive probe of variation in the fine-structure constant. The paper also supplies the supporting atomic properties—lifetimes, polarizabilities, quadrupole moments, hyperfine constants, isotope-shift factors, and Lorentz-invariance matrix elements—needed to plan experiments.","feed_headline":"Ir17+ predicted to clock at 346 nm and 833 nm","feed_subtitle":"Two independent relativistic methods converge on narrow, long-lived transitions that could test the fine-structure constant.","key_machinery":"The machinery is the relativistic many-body calculation under the Dirac-Coulomb-Gaunt Hamiltonian. Two independent methods carry the argument: a Kramers-restricted configuration-interaction (KRCI) calculation with single and double excitations of 32 electrons, and a Fock-space coupled-cluster (FSCC) calculation with single, double, and estimated triple excitations, both taken to increasingly large correlation-consistent basis sets. The argument is carried by the convergence pattern: basis-set enlargement and triple-excitation corrections move the two methods into agreement, and the same two methods are used for finite-field calculations of polarizabilities, quadrupole moments, hyperfine constants, and isotope-shift factors. The finite-field technique, which fits energies against applied electric-field or field-gradient perturbations, is what produces the differential polarizability and the 'magic' trap-frequency estimate.","core_discovery":"The central claim is that Ir$^{17+}$, with a ground configuration $4f^{13}5s$ and near-degenerate $4f^{14}$ and $4f^{12}5s^2$ excited configurations, has two strongly forbidden optical transitions with clock-grade properties. The KRCI calculation places the $4f^{14} {}^1S_0$ level at 12006 cm$^{-1}$ and the $4f^{12}5s^2 {}^3H_6$ level at 28848 cm$^{-1}$, corresponding to transitions at roughly 346 nm and 833 nm from the ground state; the paper quotes uncertainties of about 30 nm and 100 nm on these wavelengths. The first transition has a negative differential scalar polarizability of $-0.203$ a.u., which supports cancellation of the trap-induced Stark and micromotion shifts at a 'magic' radiofrequency near $500 \\times 2\\pi$ MHz, and the two transitions have opposite relativistic sensitivity coefficients ($K_\\alpha \\approx 68.7$ and $-35.5$), so their frequency ratio would be unusually sensitive to variation of the fine-structure constant. Long lifetimes, including 610 ms for the lowest excited state and states with lifetimes exceeding 100 ms, and small electric quadrupole moments for the ground and ${}^3H_6$ states are presented as additional clock advantages.","pith_inferences":["A direct experimental target is electron-beam-ion-trap spectroscopy of Ir$^{17+}$ searching for the predicted 346 nm and 833 nm lines; confirmation would also settle whether the $4f^{14} {}^1S_0$ level sits near $12000$ cm$^{-1}$ rather than the $5000$-$7000$ cm$^{-1}$ range some earlier calculations suggested.","The same two-method convergence strategy could be transferred to neighboring charge states of iridium or to other $4f$-hole ions, where similar level crossings may hide clock-grade forbidden transitions.","A measurement of the differential polarizability of the 346 nm transition would test the magic-trap prediction more sharply than a frequency measurement alone, since a sign error in the computed $\\Delta\\alpha_d^S$ would eliminate the cancellation.","If omitted QED and full-Breit corrections shift the $4f^{14} {}^1S_0$ energy by thousands of cm$^{-1}$, the 833 nm transition may be the more robust candidate because its upper state's energy shows less scatter across the two methods and earlier calculations."],"forward_implications":["If the central claim is correct, Ir$^{17+}$ becomes one of only a few highly charged ions with two viable optical clock transitions, one near-ultraviolet and one infrared.","A clock on the 346 nm transition could operate at a trap drive frequency near $500 \\times 2\\pi$ MHz where the DC Stark shift and the micromotion time-dilation shift cancel.","The frequency ratio of the two transitions would shift if the fine-structure constant changes, with the two lines moving in opposite directions, giving a self-calibrating $\\alpha$-variation test.","The small quadrupole moments and low polarizabilities of the proposed clock states would suppress the dominant systematic shifts, while the large $T^{(2)}$ matrix elements would make Ir$^{17+}$ competitive in Lorentz-invariance searches.","The opposite-sign field-shift factors of the two transitions provide a path to isotope-shift and nuclear-structure studies of iridium."],"supporting_citations":[{"why":"Identifies Ir17+ as highly sensitive to alpha variation from the 4f-5s level crossing, the motivation for this study.","marker":"[37]"},{"why":"Establishes the mechanism by which 4f-5s level crossings in iridium give enhanced sensitivity to the fine-structure constant.","marker":"[55]"},{"why":"Reports EBIT measurements of M1 transitions in Ir17+ that validate the level-crossing spectrum the present calculations refine.","marker":"[56]"},{"why":"Provides earlier CI lifetimes and transition rates, including the 3F4 lifetime that this work revisits and finds much longer.","marker":"[57]"},{"why":"Gives the earlier CI prediction of the 4f14 1S0 energy near 12000 cm^-1 with which the present KRCI result agrees.","marker":"[58]"},{"why":"Supplies recent CI energy levels and Lorentz-invariance matrix elements used for direct comparison.","marker":"[59]"},{"why":"Supplies the KRCI method used for all configuration-interaction calculations.","marker":"[65]"},{"why":"Provides the 'magic' radiofrequency trap formula used to derive the 500 MHz cancellation frequency.","marker":"[86]"},{"why":"Establishes that large T(2) matrix elements in long-lived states enable Lorentz-invariance tests, motivating the reported LLI values.","marker":"[87]"}],"fun_headline_variants":["Ir17+ clock transitions at 346 nm and 833 nm","Two methods converge on Ir17+ clock lines","Ir17+ clock candidate: alpha-sensitive pair","Ir17+ predicted clock lines: 346 and 833 nm","Ir17+ clock ticks may probe fine-structure alpha"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Dirac-Coulomb-Gaunt Hamiltonian without QED corrections (vacuum polarization and self-energy) and without the full Breit interaction describes the Ir$^{17+}$ spectrum accurately enough that the predicted clock-transition wavelengths and properties are trustworthy.","fun_headline_variants_meta":{"raw":{"variants":["Ir17+ clock transitions at 346 nm and 833 nm","Two methods converge on Ir17+ clock lines","Ir17+ clock candidate: alpha-sensitive pair","Ir17+ predicted clock lines: 346 and 833 nm","Ir17+ clock ticks may probe fine-structure alpha"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000254,"raw_usage":{"total_tokens":1601,"prompt_tokens":1012,"completion_tokens":589,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":628,"completion_tokens_details":{"reasoning_tokens":508}},"tokens_in":628,"tokens_out":589,"duration_ms":6223,"temperature":1.0,"reasoning_tokens":508,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T16:32:29.287774+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A spectroscopic search in an electron-beam ion trap for the two forbidden lines would settle the claim: if no transitions appear near the predicted 346 nm and 833 nm positions, or if the $4f^{14} {}^1S_0$ level is found far outside the quoted $12006 \\pm 1392$ cm$^{-1}$ window, then the two clock-candidate transitions are refuted.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies Ir17+ as highly sensitive to alpha variation from the 4f-5s level crossing, the motivation for this study."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the mechanism by which 4f-5s level crossings in iridium give enhanced sensitivity to the fine-structure constant."},{"cited_title":"Windberger, J","cited_arxiv_id":null,"evidence_quote":"Reports EBIT measurements of M1 transitions in Ir17+ that validate the level-crossing spectrum the present calculations refine."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides earlier CI lifetimes and transition rates, including the 3F4 lifetime that this work revisits and finds much longer."},{"cited_title":"Cheung, M","cited_arxiv_id":null,"evidence_quote":"Gives the earlier CI prediction of the 4f14 1S0 energy near 12000 cm^-1 with which the present KRCI result agrees."},{"cited_title":"Knecht, H","cited_arxiv_id":null,"evidence_quote":"Supplies the KRCI method used for all configuration-interaction calculations."},{"cited_title":"Relativistic configuration-interaction and coupled-cluster calculations of Ir$^{17+}$ transition energies and properties for optical clock applications","cited_arxiv_id":"2502.01112","evidence_quote":"Establishes that large T(2) matrix elements in long-lived states enable Lorentz-invariance tests, motivating the reported LLI values."}],"review_version":1}