{"id":"bee405ab-5776-4991-a1d5-f74cc05e85a4","arxiv_id":"2608.00607","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Microwave spectroscopy of trapped HD+ determines the v=0, L=3 hyperfine interval as 1,050,804.511(11) kHz, 1.9σ above theory and consistent with scaled ground-state hyperfine data.","lead":"Physicists measured the total hyperfine splitting of the HD+ molecule in a rotationally excited state using microwave spectroscopy, reaching a relative uncertainty of 10 parts per billion. The measured value sits 1.9 standard deviations above the leading theoretical prediction, and the work should help resolve a long-standing hyperfine discrepancy in deuterated hydrogen molecular ions.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 10-ppb central claim rests on a theoretical Zeeman extrapolation and an uncertainty budget that are not present in the preprint; without the missing data and supplementary model, the zero-field frequency cannot be independently verified.","rationale":"The reader's conditional verdict is reasonable and I do not propose to change it. My stress-test identifies the same broad area of concern—the zero-field extrapolation and systematic budget—but the most load-bearing issue is not an obvious error in the Zeeman model; it is that the entire uncertainty analysis and the data are absent from the preprint. The paper explicitly promises a Supplemental Material with details on Zeeman shifts, systematic corrections, and uncertainties, and a DataverseNL link 'to be inserted by authors.' Without these, an independent reviewer cannot verify that the 8.8-Hz statistical zero-field uncertainty, the 5.4-Hz AC-Zeeman correction, or the 3.0-Hz AC-Stark correction are correctly evaluated. The reader's specific worry about a 10-20 Hz Zeeman error seems less likely given the stated 1e-8 theoretical accuracy, but the missing material prevents ruling it out. The hybrid rotational-scaling claim is a secondary interpretation and is robust to moderate correlation deviations as the authors show, so it is not the weakest point. Thus the conditionality of the verdict is appropriate, and no change is needed.","tokens_in":10198,"tokens_out":9198,"duration_ms":125230,"concrete_test":"Obtain the promised DataverseNL dataset and Supplemental Material. Reproduce the Fig. 3 fit of line centers versus bias field using the published model, then rerun the fit with (i) the quadratic Zeeman coefficient left free instead of fixed to the theoretical value, and (ii) a 1% linear scaling of the Be+-derived magnetic-field axis. If the zero-field intercept shifts by more than 0.011 kHz in either test, the 10-ppb claim is not robust to the Zeeman-model and field-calibration assumptions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reported zero-field frequency, 1,050,804.511(11) kHz, is obtained by fitting measured line centers versus bias field to the theoretical HD+ Zeeman curve, then applying several systematic corrections. The paper states the Zeeman theory is known to several parts in 10^8 and that the leading systematic uncertainties are the AC-Zeeman shift (5.4 Hz), AC-Stark shift (3.0 Hz), and MW-field AC-Zeeman shift (1.8 Hz). However, the full Zeeman model, the derivation of the 24.09 kHz/G^2 coefficient, the magnetic-field offset fit, and the systematic-correction calculations are all relegated to the Supplemental Material, which is absent from the preprint. The dataset promised in the data-availability statement is also missing. If, for example, the quadratic Zeeman coefficient or the B-field scale inferred from Be+ spectroscopy were wrong by a few percent, the zero-field intercept would shift by several hertz—comparable to the claimed 11-Hz total uncertainty. Because the data and analysis code are not available, this cannot be checked, so the central 10-ppb claim is currently unverified.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This Letter reports the first measurement of the total hyperfine interval in the (v=0, L=3) manifold of HD+ using microwave spectroscopy of sympathetically cooled HD+ ions in a Paul trap. The authors overcome the low occupancy of the relevant hyperfine state by combining intermittent Majorana depolarization with blackbody-driven rotational redistribution, achieving an effective population amplification of 25. They observe the first-order-field-insensitive component J=5, M_J=0 → J=4, M_J=−1 and, after fitting the line centers versus bias field to the theoretical HD+ Zeeman curve, obtain a zero-field transition frequency of 1,050,804.511(11) kHz (10 ppb). This differs from the ab initio prediction, 1,050,802.78(89) kHz, by 1.9σ. A hybrid prediction constructed from the L=0 experimental spin coefficients of König et al. and theoretical L-scaling of the Fermi-contact coefficients agrees with the measured value, which the authors interpret as support for the correctness of the rotational scaling of the hyperfine theory. The main text gives an uncertainty budget and a plausible high-level description, but the central Zeeman model, the systematic-correction derivations, and the analysis/data are relegated to a Supplemental Material and a data repository that are not included in the preprint.","tokens_in":10467,"tokens_out":9777,"duration_ms":119535,"significance":"If the result is correct, it is a significant experimental advance: it is the most precise HD+ hyperfine measurement to date, 81 times more accurate than the theoretical prediction, and it sharpens the existing tension between mQED hyperfine theory and experiment. The population-amplification technique is novel and potentially transferable to other hydride ions. The hybrid analysis is a clever consistency test, and the authors are transparent about the correlation assumption behind it. The main weakness is that the load-bearing details — the Zeeman model, the 24.09 kHz/G^2 coefficient, the B-field offset fit, and the systematic corrections — are not present in the manuscript itself, so the central 10-ppb claim cannot currently be independently verified.","major_comments":[{"comment":"The zero-field frequency is the central result. It is obtained by fitting measured line centers as a function of bias field to the theoretical HD+ Zeeman curve, with an overall field offset as a free parameter. The quadratic Zeeman coefficient 24.09 kHz/G^2, the derivation and uncertainty of the Zeeman curve, the propagation of the 10-mG field noise and the fitted −10(6) mG offset into the 8.8-Hz statistical uncertainty, and the 1.2-Hz quadratic-Zeeman systematic are all relegated to Supplemental Material [19], which is not included with the preprint. The data in Fig. 3 span only 1–2 G; the extrapolation to zero field therefore depends sensitively on the assumed curvature. A deviation of a few percent in the 24.09 kHz/G^2 coefficient would shift the intercept by kilohertz, i.e., far above the claimed 11-Hz total uncertainty. The paper's statement that the Zeeman effect is known to severa","section":"Control over Zeeman shifts / Fig. 3"},{"comment":"The data availability statement is a placeholder ('{public DataverseNL handle to be inserted by authors}'), and the promised measurement data, fit model, and analysis are not present. For a 10-ppb claim that disagrees with theory by 1.9σ and relies on a multi-step analysis (Welch's t-test, slow-signal correction, Lorentzian fits, Zeeman fit, multiple systematic corrections), the underlying data and code are necessary for a referee to confirm the central result. Please provide the actual data repository and, at minimum, the fitted line centers with their uncertainties and residuals as a table in the supplement.","section":"Data availability"},{"comment":"The a-posteriori calibration uses the theoretical HD+ Zeeman curve, which is derived within the same hyperfine theory that the measurement is designed to test. The manuscript should quantify the sensitivity of the quadratic Zeeman coefficient and the inferred field offset to the hyperfine parameters, and should demonstrate that a 1.9σ error in the hyperfine intervals does not alter the zero-field intercept by more than a fraction of 1 Hz. This analysis is absent from the main text and presumably resides in the missing supplement.","section":"Control over Zeeman shifts / Fig. 3"}],"minor_comments":[{"comment":"The final paragraph states the transition as '(F, S, J) = (1,2,4)→(1,2,5)', but the rest of the paper uses '(1,2,5)→(0,1,4)'. Please correct this inconsistency.","section":"Summary / last paragraph"},{"comment":"The caption gives the zero-field theoretical value as 1,050,802.79(89) kHz, while the text and abstract quote 1,050,802.78(89) kHz. Please harmonize the rounding.","section":"Fig. 3 caption"},{"comment":"The line is called 'field-insensitive', but it has a quadratic Zeeman shift of 24.09 kHz/G^2, which is about 95 kHz at 1.99 G. Consider calling it 'first-order field-insensitive' or 'linearly field-insensitive' to avoid confusion.","section":"Zeeman section"},{"comment":"The statement that the ratios E_i(0,3)/E_i(0,0) have 'zero uncertainty' is categorical; the text immediately acknowledges the correlation assumption, but it would be clearer to state that this holds exactly only under the hypothesis of perfectly correlated theoretical uncertainties. The 0.98-robustness test is helpful, but the wording should reflect the conditional nature.","section":"Hybrid prediction"}],"recommendation":"major_revision","confidential_remarks":"The paper appears to be from a leading group and the result is likely important. My recommendation is driven by the absence of the supplemental material and the placeholder data-availability statement, not by any identified error in the main text. I would be willing to review a revised version that includes the Zeeman model, systematic-correction derivations, and the actual data/fit model."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuine, careful measurement of a new HD+ hyperfine interval, with a clever new technique, and the main thing holding it back is that the supporting material isn't in the preprint. The claim is believable, but not independently checkable right now.\n\nWhat's new: first measurement of the total hyperfine interval in v=0,L=3 of HD+, at 10 ppb, 81x better than theory. The technique that combines intermittent Majorana depolarization with blackbody-driven rotational redistribution is a real technical contribution, and it's generalizable to other sympathetically cooled hydride ions. The uncertainty budget is laid out in enough detail: 8.8 Hz statistical from the Zeeman fit, leading systematics 5.4, 3.0, 1.8, 1.2 Hz, total 11 Hz. The 1.9 sigma deviation from theory is a legitimate result, and the consistency with Koenig et al.'s L=0 measurement (different trap, different group) is encouraging. The hybrid prediction is properly caveated: they test correlation coefficient 0.98 and it still holds. So the physics story is coherent.\n\nSoft spots, in order: (1) The supplemental material and data are absent from the arXiv version. The Zeeman model, the derivation of the 24.09 kHz/G^2 coefficient, the Be+ calibration procedure, and the systematic-correction calculations are all in the SM. The stress-test point is fair: they calibrate the field a posteriori using the theoretical HD+ Zeeman curve, and if that curve had an error of order 10-20 Hz at 1-2 G, the zero-field intercept would shift by several Hz--comparable to the claimed 11 Hz. Their statement that the Zeeman effect is known to several parts in 10^8 is a strong one; it's probably right, but without the derivation and the code, a referee can't verify it. This is a verification problem, not evidence of error. (2) The hybrid prediction relies on the perfect-correlation assumption between E_i(0,0) and E_i(0,3). They acknowledge it and test robustness, which is good--but note that the hybrid is presented as a consistency test, not as independent input. That's fine; the paper doesn't overclaim. (3) Minor: the data-availability statement says the data will be on DataverseNL, but the handle is a placeholder. Not an issue for the preprint, but should be in place before publication.\n\nCitation pattern is fine: all relevant prior work is cited, no obvious gaps. The paper is honest about the limitations and the circularity of the hybrid.\n\nBottom line: this deserves a serious referee. The measurement is important for the HD+ hyperfine puzzle and the CODATA adjustment. The main thing I'd ask for is the missing supplement and dataset at revision. I'd probably cite it when it appears.\n\nRecommendation: accept for peer review, with a request to make the supplementary material and data available. I'd take it to a reading group if the group does precision spectroscopy.","headline":"A solid, novel precision measurement of a new HD+ hyperfine interval, with the main caveat being missing supplementary material rather than any apparent flaw in the logic.","tokens_in":10992,"tokens_out":2845,"would_cite":true,"duration_ms":31270,"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":"The first direct microwave measurement of the total hyperfine interval in the (v=0,L=3) state of HD+ yields 1,050,804.511 ± 0.011 kHz, 81 times more precise than theory and 1.9σ higher.","keywords":["HD+","hyperfine structure","microwave spectroscopy","Paul trap","Majorana depolarization","blackbody redistribution","Fermi-contact interaction","molecular QED"],"falsifier":"Measure the same J=5→J=4 hyperfine transition at several bias fields between 1 and 2 G using an independently calibrated magnetic field (for example, fully characterized Be+ co-magnetometry), and check whether the zero-field extrapolation still gives 1,050,804.511 kHz; if the theoretical Zeeman curve is off by more than about 10 Hz at these fields, the extrapolated frequency will move away from the reported value. Alternatively, a future direct microwave measurement of the v=9,L=3 total hyperfine interval that disagrees with the predicted −6.8(1.2) kHz deviation would undercut the rotational-s","tokens_in":10108,"feed_emoji":"⚛️","tokens_out":5034,"duration_ms":49393,"temperature":0.7,"pith_summary":"This paper reports the first direct microwave measurement of the total hyperfine interval in the (v=0, L=3) rovibrational state of the HD+ molecular ion. The measured value, 1,050,804.511 kHz with a 10-parts-per-billion relative uncertainty, is 81 times more precise than the best ab initio prediction and exceeds it by 1.9 standard deviations, deepening an unresolved tension between molecular quantum-electrodynamics theory and experiment. To observe a transition that starts from a state occupied by only 0.24% of the ions, the authors amplify the usable population by a factor of 25 through a combination of intermittent Majorana depolarization and blackbody-driven rotational redistribution. They further show that the result is consistent with a recent Penning-trap measurement of the L=0 hyperfine structure once theoretical rotational scaling is assumed, supporting the idea that the hyperfine error is not in the rotational dependence of the Fermi-contact interactions.","feed_headline":"HD+ hyperfine interval measured to 10 ppb","feed_subtitle":"New microwave measurement is 81 times more precise than theory and leaves a 1.9-sigma gap to solve.","key_machinery":"The key mechanism is a population-amplification scheme: intermittent Majorana depolarization randomly re-orients the spin state of the HD+ ensemble, while blackbody radiation repopulates rotational levels, together raising the occupancy of the target magnetic sublevel from 0.24% to about 6%. The measured transition, J=5, M_J=0 → J=4, M_J=-1, has a small quadratic Zeeman shift, and its residual field dependence is removed by fitting line centers at several bias fields to the theoretical Zeeman curve. This yields a zero-field interval with an 8.8-Hz statistical uncertainty, later corrected for systematic shifts.","core_discovery":"The central claim is that the total hyperfine interval in the (v=0,L=3) state of HD+ is 1,050,804.511 ± 0.011 kHz, determined by detecting a single field-insensitive magnetic subcomponent of the J=5→J=4 transition at 1.0508 GHz in a linear Paul trap. The paper argues that this measurement, at 10 ppb fractional uncertainty, is the most precise hyperfine benchmark for HD+ to date and provides a critical test of molecular quantum-electrodynamics theory. The measured value sits 1.9σ above the ab initio prediction of 1,050,802.78 ± 0.89 kHz, while a hybrid prediction built from the measured L=0 coefficients scaled by theoretical rotational ratios agrees with the new result to within 0.06σ. The au","pith_inferences":["If the rotational-scaling hypothesis is correct, then comparing the L=3 and L=0 measurements isolates the L-dependent part of the hyperfine discrepancy and may locate the missing physics in terms that do not scale with rotation, such as nuclear-structure or higher-order QED effects tied to the deuteron.","The same population-amplification approach could be adapted to other hydride molecular ions with REMPD-accessible vibrational levels, potentially extending high-precision hyperfine benchmarks to species such as H2+ or other deuterated ions.","A direct measurement of the v=9,L=3 hyperfine interval, as the authors propose, would test the predicted −6.8 kHz deviation and could determine whether the discrepancy scales with vibrational quantum number, offering a further discriminator between theories.","The reliance on the theoretical Zeeman curve for the zero-field extrapolation could be checked by independently calibrating the magnetic field with fully characterized Be+ spectroscopy, which would turn the current a posteriori calibration into a model-independent measurement."],"forward_implications":["The measured hyperfine interval provides a benchmark for mQED hyperfine calculations, 81 times more accurate than the current theory, which should stimulate improved calculations of the missing contributions.","Combined with the L=0 Penning-trap results, the measurement supports the hypothesis that the rotational scaling of the Fermi-contact coefficients E4 and E5 is correctly predicted by theory, so the theoretical error is roughly L-independent rather than rotationally growing.","The measurement, together with previous two-photon spectroscopy of the v=0→v=9 transition, implies a −6.8(1.2) kHz deviation from theory for the total hyperfine interval in v=9, L=3, a prediction the authors plan to test directly.","The demonstrated population-amplification technique can be applied to any microwave or optical transition in sympathetically cooled molecular ions with appreciable 300-K rotational population, enabling spectroscopy of rare quantum states."],"supporting_citations":[{"why":"supplies the theoretical hyperfine coefficients and errors that the measurement is compared against","marker":"[21]"},{"why":"provides the measured L=0 spin coefficients E4 and E5 used in the hybrid prediction","marker":"[24]"},{"why":"gives the ab initio total hyperfine interval f_theo and its 0.89 kHz uncertainty","marker":"[35]"},{"why":"defines the quantum-number coupling scheme and the CODATA extraction context for HD+ hyperfine data","marker":"[13]"},{"why":"the prior two-photon spectroscopy that revealed the hyperfine discrepancies motivating this work","marker":"[5]"},{"why":"provides the Zeeman-shift theory and the systematic-error budget underlying the final uncertainty","marker":"[19]"},{"why":"supplies the magnetic-field dependence formalism used for the zero-field extrapolation","marker":"[31]"}],"fun_headline_variants":["HD+ hyperfine interval pinned down to 10 ppb","Rotationally excited HD+ hyperfine measured at 10 ppb","HD+ hyperfine result challenges theory by 1.9σ","Trap trick amplifies HD+ signal, nails hyperfine to 10 ppb"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The paper assumes the theoretical Zeeman shift of HD+ is accurate to a few parts in 10^8; this theory is used both to set the bias field scale and to extrapolate line positions measured at 1–2 G down to zero field, so any error of order 10–20 Hz in the Zeeman model would shift the reported 1,050,804.511 kHz value.","fun_headline_variants_meta":{"raw":{"variants":["HD+ hyperfine interval pinned down to 10 ppb","Rotationally excited HD+ hyperfine measured at 10 ppb","HD+ hyperfine result challenges theory by 1.9σ","Trap trick amplifies HD+ signal, nails hyperfine to 10 ppb"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000673,"raw_usage":{"total_tokens":2964,"prompt_tokens":867,"completion_tokens":2097,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":611,"completion_tokens_details":{"reasoning_tokens":2019}},"tokens_in":611,"tokens_out":2097,"duration_ms":16519,"temperature":1.0,"reasoning_tokens":2019,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T00:31:07.723493+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same J=5→J=4 hyperfine transition at several bias fields between 1 and 2 G using an independently calibrated magnetic field (for example, fully characterized Be+ co-magnetometry), and check whether the zero-field extrapolation still gives 1,050,804.511 kHz; if the theoretical Zeeman curve is off by more than about 10 Hz at these fields, the extrapolated frequency will move away from the reported value. Alternatively, a future direct microwave measurement of the v=9,L=3 total hyperfine interval that disagrees with the predicted −6.8(1.2) kHz deviation would undercut the rotational-s","supporting_citations":[{"cited_title":"Haidar, V","cited_arxiv_id":null,"evidence_quote":"supplies the theoretical hyperfine coefficients and errors that the measurement is compared against"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the measured L=0 spin coefficients E4 and E5 used in the hybrid prediction"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the ab initio total hyperfine interval f_theo and its 0.89 kHz uncertainty"},{"cited_title":"Karr and J","cited_arxiv_id":null,"evidence_quote":"defines the quantum-number coupling scheme and the CODATA extraction context for HD+ hyperfine data"},{"cited_title":"Patra, M","cited_arxiv_id":null,"evidence_quote":"the prior two-photon spectroscopy that revealed the hyperfine discrepancies motivating this work"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the Zeeman-shift theory and the systematic-error budget underlying the final uncertainty"},{"cited_title":"Bakalov, V","cited_arxiv_id":null,"evidence_quote":"supplies the magnetic-field dependence formalism used for the zero-field extrapolation"}],"review_version":1}