{"id":"e59e42e3-a1ad-4578-852b-274553f74d43","arxiv_id":"2608.01916","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The Yb+(E3)/Sr optical frequency ratio is measured with 4.3e-18 fractional uncertainty, a 3.5x improvement over the previous best, using a transportable Sr clock.","lead":"Physicists measured the frequency ratio between a ytterbium ion clock and a strontium lattice clock to a fractional uncertainty of 4.3 parts in 10^18, the most precise value yet for this pair. The strontium clock is transportable and was moved between measurement campaigns, demonstrating that such clocks can stay accurate at the 10^-18 level.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Global-νE1 lattice shift evaluation has a model discrepancy: new vs per-campaign lattice shifts differ by up to 4.5e-18, exceeding the claimed 4.3e-18 total uncertainty.","rationale":"The experiment and main comparison are credible, with internal consistency checks, transparent campaign handling, and agreement with several previous measurements. However, the central uncertainty claim hinges on the Sr4 lattice light shift evaluation. The Supplemental Material itself reveals that the adopted global-νE1 method and the previous per-campaign method disagree by up to 4.5e-18, which is larger than the stated total uncertainty. This is not an external-consensus disagreement; it is an internal model dependence that is not included as a systematic uncertainty. The reader identified the same general weakness (reproducibility of νE1 across transport), but the sharper issue is the method-to-method discrepancy. Therefore, acceptance should be conditional on either reconciling the two evaluations or adding a model-difference uncertainty to the budget.","tokens_in":11992,"tokens_out":7690,"duration_ms":88474,"concrete_test":"Recompute the C1–C3 ratios with the per-campaign interleaved two-trap-depth lattice light shift evaluation of Ref. [20] (same raw data, Yb1 and gravity corrections unchanged) and combine with the same covariance-weighted average. If the resulting overall R shifts by more than 1e-18, or if the per-campaign evaluation has an uncertainty that does not cover the global-νE1 result, the global-νE1 assumption is not validated; at minimum add a systematic model uncertainty equal to half the difference (≈2.3e-18) and re-evaluate the 4.3e-18 claim.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The 4.3e-18 claim depends critically on the Sr4 lattice light shift. The paper replaces per-campaign interleaved evaluations [20] with one global E1 magic frequency νE1 = 368554463.4(1.7) MHz assumed stable across transports (Supplemental Material, 'Determination of the E1 magic frequency'). The Supplemental states that this approach and the per-campaign approach differ by up to 4.5e-18 in the lattice light shift. That is larger than the quoted total uncertainty of 4.3e-18 and larger than the new lattice-shift uncertainties (1.3–2.8e-18). The global νE1 is derived from only three internal measurements with χ2_red = 1.6 (p = 0.16); the analysis multiplies errors by sqrt(χ2), but a common-mode bias (e.g., atomic-temperature determination, as the authors themselves suspect) would shift the weighted mean without increasing its error. If the per-campaign evaluation is also valid, the method choice alone moves R by up to 4.5e-18, so the uncertainty budget needs a model-difference term that is not present.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a determination of the optical frequency ratio between the 171Yb+ E3 transition at 642 THz and the 87Sr clock transition at 429 THz, using the stationary Yb1E3 single-ion clock and the transportable Sr4 lattice clock at PTB. The measurement spans four campaigns over nearly two years, with Sr4 occasionally transported and operated off-site. The ratio is measured phase-coherently via an optical frequency comb referenced to a common ultrastable laser. The final result is R = 1.495 991 618 544 900 588 1(65), with a fractional uncertainty of 4.3 × 10^-18, a factor 3.5 improvement over the previous best. The campaign results are statistically consistent (daily reduced chi-squared 1.19), and the uncertainty budget combines statistics, Yb1E3 systematics, Sr4 systematics, and gravitational redshift. The Sr4 systematic is dominated by the lattice light shift, now evaluated using a global E1 magic frequency νE1 = 368 554 463.4(1.7) MHz that is assumed reproducible across transports.","tokens_in":12295,"tokens_out":3730,"duration_ms":39347,"significance":"If the quoted uncertainty is sound, this measurement is the most accurate interspecies frequency comparison involving a transportable clock to date, and one of only a few below 5 × 10^-18 that meet the roadmap requirement for optical redefinition of the second. The paper's strengths are the direct phase-coherent measurement chain, the explicit per-campaign uncertainty budget, the covariance treatment of correlated systematics, and the demonstration of long-term reproducibility after transportation. Those features make the work valuable for chronometric geodesy and for linking clock clusters. The central uncertainty claim, however, rests critically on the Sr4 lattice light shift evaluation, whose model choice introduces a possible bias that is not currently included in the uncertainty budget.","major_comments":[{"comment":"Table I quotes a total fractional uncertainty of 4.3 × 10^-18, dominated by the Sr4 systematic contribution of 2.6 × 10^-18. The lattice light shift is evaluated using the global νE1 value. The supplement states explicitly that 'the lattice light shifts differ by no more than 4.5 × 10^-18 between this approach and the per-campaign approach [20]'. This difference is larger than the quoted total uncertainty and larger than the individual lattice-light-shift uncertainties (1.3–2.8 × 10^-18). Because the per-campaign method was used in the previous work and is not demonstrated to be invalid, the choice between the two methods changes the central value by more than the claimed uncertainty. The paper needs either a quantitative justification for preferring the global method, or a conservative systematic term covering the model difference. The statement 'at the level of the lattice light shift","section":"Supplemental Material, 'Determination of the E1 magic frequency in 87Sr'; Table I"},{"comment":"The global νE1 is derived from only three internal measurements with χ2_red = 1.6 (p = 0.16). The authors scale the uncertainties by sqrt(χ2_red), but this does not protect against a common-mode bias in the analysis, which they themselves identify as a plausible explanation (e.g., atomic temperature determination). Such a common-mode bias would shift the weighted mean νE1 and hence the lattice light shift for all campaigns without increasing the reported error. A sensitivity analysis or an explicit additional uncertainty for this bias is needed to support the 4.3 × 10^-18 claim.","section":"Supplemental Material, 'Determination of the E1 magic frequency in 87Sr'"}],"minor_comments":[{"comment":"Typographical issue: 'below5 × 10−18' is missing a space; should read 'below 5 × 10−18'.","section":"Abstract"},{"comment":"The axis labels and tick labels in the figure are garbled, e.g., '4/20230 3/20241 0/20240 2/2025'. Please ensure the final figure has clean month/year labels.","section":"Figure 1"},{"comment":"The legend/figure text repeats 'Ref. [29], 2025' twice; the labels should distinguish the two measurements. Also, the sentence about point colors before 2026 is difficult to parse.","section":"Figure 2"},{"comment":"The symbol eR is used in the table header but is not defined in the main text until later; consider defining it at first use.","section":"Table I"}],"recommendation":"major_revision","confidential_remarks":"The core measurement approach is sound and the dataset is valuable. The main concern is the lattice light shift model-dependence, which exceeds the quoted total uncertainty. I believe this is fixable by adding a model-difference term or providing a rigorous argument for excluding the per-campaign evaluation, but without that, the headline uncertainty cannot be considered established. I would not recommend reject because the underlying data and methods are credible and the issue is confined to the systematic evaluation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Know this: the paper reports the most accurate measurement to date of the 171Yb+(E3)/87Sr frequency ratio, at 4.3e-18 fractional uncertainty, a 3.5x improvement, using a transportable Sr lattice clock against a stationary Yb+ ion clock over four campaigns. The result is likely right, and the transportable clock's reproducibility at the 10^-18 level is a real milestone.\n\nWhat's new: a factor-3.5 gain in the ratio, and the first interspecies comparison below 5e-18 involving a transportable clock. The measurement is a phase-coherent comb comparison against a common ultrastable laser, so no free parameters are fitted to force the answer. The uncertainty budget is internally consistent: statistical 2.1e-18, Yb+ 2.7e-18, Sr4 2.6e-18, gravity 0.5e-18, combined 4.3e-18. Campaign C0 is downweighted correctly via the covariance matrix; chi2_red of daily means is 1.19, so the scatter is consistent. The paper also connects the result to known inconsistencies with SYRTE-Sr2, which is useful.\n\nThe soft spot: the Sr4 lattice light shift is evaluated with a single global E1 magic frequency instead of per-campaign interleaved measurements. The supplemental states that the two approaches differ by up to 4.5e-18 in the lattice shift, which is larger than the quoted 4.3e-18 total. The authors say the difference is at the level of the lattice shift uncertainties, but the new uncertainty on that shift is only 1.3-2.8e-18, so a 4.5e-18 model difference is not fully covered. The global value is an average of three measurements with chi2_red=1.6; inflating by sqrt(chi2) doesn't capture a common-mode bias like atomic temperature, which the authors themselves suspect. This deserves an explicit model-difference term or a sensitivity check in the total budget.\n\nThe math and data otherwise look solid. The citation pattern is appropriate; prior work is credited and the comparison to CIPM evaluations is transparent.\n\nWho this is for: optical clock frequency metrology, especially groups working on transportable clocks or the second's redefinition. It deserves a serious referee; the lattice shift question should be pushed in review, but the central claim is supported and the paper is written honestly.","headline":"A new best value for the Yb+ E3 / Sr ratio at 4.3e-18 from a transportable Sr clock, with a caveat about the lattice shift model difference.","tokens_in":12819,"tokens_out":2284,"would_cite":true,"duration_ms":23687,"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 most accurate interspecies frequency ratio involving a transportable clock is measured at 4.3 × 10^-18, demonstrating reproducible 10^-18-level operation.","keywords":["frequency ratio","optical lattice clock","transportable clock","ytterbium ion clock","E3 transition","interspecies comparison","lattice light shift","redefinition of second"],"falsifier":"Re-determine νE1 independently at a later campaign (or after a transport) and compare it to 368 554 463.4 MHz. If the difference exceeds the 1.7 MHz uncertainty, the fixed-magic-frequency assumption fails and the reported 4.3 × 10^-18 total uncertainty would need revision. A direct interleaved two-trap-depth measurement at a remote site would also test the prediction.","tokens_in":11898,"feed_emoji":"🕐","tokens_out":4492,"duration_ms":46030,"temperature":0.7,"pith_summary":"The paper reports the most accurate frequency ratio yet measured between two different atomic species when one of the clocks is transportable. The ratio between the ytterbium ion's octupole transition and strontium's clock transition is determined with fractional uncertainty 4.3 × 10^-18, improving on the previous best by a factor of 3.5. Four measurement campaigns across nearly two years, including physical transport and off-site operation of the strontium lattice clock, give statistically consistent results. This demonstrates that a transportable optical lattice clock can reproduce its frequency at the 10^-18 level, a requirement for its use in chronometric geodesy and as a transfer standard for inter-institute comparisons.","feed_headline":"Transportable clock hits 4.3e-18 in interspecies comparison","feed_subtitle":"Four campaigns over two years show reproducible 10^-18 operation, enabling gravity mapping and clock transfer without fiber links.","key_machinery":"The measurement chain connects each clock's laser to a common ultrastable reference laser via an optical frequency comb, allowing a phase-coherent comparison of the two clock transitions. The transportable strontium clock's lattice light shift is evaluated using a measured E1 magic frequency νE1 = 368 554 463.4(1.7) MHz, assumed reproducible across campaigns, with the lattice laser detuning measured per campaign. The ytterbium ion clock operates on the electric-octupole transition with a systematic uncertainty of 2.7 × 10^-18. A covariance-based weighted average combines the four campaigns, treating systematic errors as fully correlated.","core_discovery":"The authors determine the optical frequency ratio νYb+/νSr = 1.495 991 618 544 900 588 1(65) with fractional uncertainty 4.3 × 10^-18. This is the most accurate interspecies clock comparison involving a transportable clock and one of the few measurements meeting the roadmap requirement of below 5 × 10^-18 for the redefinition of the SI second. The result is derived from a weighted average of four campaigns, whose means differ by no more than 3 × 10^-18 from the overall average, demonstrating reproducible operation of the transportable clock after transport and operational breaks. The authors also update the lattice light shift evaluation of the strontium clock using a single E1 magic frequen","pith_inferences":["If the E1 magic frequency reproducibility holds for future transports, transportable clocks could omit per-campaign lattice shift calibrations, shortening remote measurement campaigns.","The consistency across four campaigns suggests that operational magic intensity techniques may be robust enough for other lattice clock species.","A future measurement of νYb+/νSr after a longer idle period, or with a second transportable clock, would test whether the 1.7 MHz reproducibility bound is a hard floor or can be lowered further."],"forward_implications":["The transportable clock can act as a transfer standard for inter-institute clock comparisons without optical fiber links.","The measured ratio provides a benchmark that can link the two existing clusters of sub-5×10^-18 clock comparisons, e.g., by measuring the 27Al+/171Yb+ ratio.","The result supports the conclusion that a 2022 comparison suffered an uncontrolled ~10^-16-level frequency error, affecting current recommended frequency values.","Demonstrated reproducibility enables chronometric leveling at the centimeter uncertainty level."],"fun_headline_variants":["Transportable clock sets 4.3e-18 interspecies ratio","Interspecies clock test hits 4.3e-18 with transportable unit","Two-year clock comparison yields 4.3e-18 precision","4.3e-18 ratio from transportable clock comparison"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The strontium clock's lattice light shift is computed from a single measured E1 magic frequency that is assumed not to change when the clock is transported or over time; all campaigns rely on this fixed value rather than per-campaign lattice shift determinations.","fun_headline_variants_meta":{"raw":{"variants":["Transportable clock sets 4.3e-18 interspecies ratio","Interspecies clock test hits 4.3e-18 with transportable unit","Two-year clock comparison yields 4.3e-18 precision","4.3e-18 ratio from transportable clock comparison"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001307,"raw_usage":{"total_tokens":5200,"prompt_tokens":813,"completion_tokens":4387,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":557,"completion_tokens_details":{"reasoning_tokens":4309}},"tokens_in":557,"tokens_out":4387,"duration_ms":32884,"temperature":1.0,"reasoning_tokens":4309,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T18:15:02.866700+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-determine νE1 independently at a later campaign (or after a transport) and compare it to 368 554 463.4 MHz. If the difference exceeds the 1.7 MHz uncertainty, the fixed-magic-frequency assumption fails and the reported 4.3 × 10^-18 total uncertainty would need revision. A direct interleaved two-trap-depth measurement at a remote site would also test the prediction.","supporting_citations":[],"review_version":1}