{"id":"6d814d32-d5df-4216-937b-73adef1ab343","arxiv_id":"2506.04320","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"First measurement of the 1[5/2]5/2 state's lifetime (37.9 µs) and branching ratios in 172Yb+, plus a tenfold improvement in the 2D5/2 to 2S1/2 branching ratio.","lead":"This paper reports the first direct measurement of the lifetime and decay paths of a high-lying excited state in a single ytterbium ion, finding it lives about 38 microseconds, far longer than expected. The results provide a benchmark for atomic structure theory and a new tool for qubit control and detection in trapped-ion quantum computers.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Potential unmodeled decay of 1[5/2]5/2 to the 2F5/2 state could bias the reported branching ratios; not constrained by the b_F7/2=0 result.","rationale":"The reader correctly identified that an unmodeled decay path from 1[5/2]5/2 could shift the extracted branching ratios, but concluded this was not load-bearing because the measured b_F7/2 is consistent with zero. That conclusion overlooks channels such as 4f13 6s2 2F5/2, which lies below the 1[5/2]5/2 state and would decay rapidly by E1 into the 2D manifold, thereby evading the dark-state constraint while still contaminating the branching-ratio analysis. The lifetime claim (τ = 37.9(9) μs) is robust to this concern because the decay time constant is set by the total decay rate regardless of the final channel, and any fast cascade would not change the measured exponential lifetime. The branching ratios, however, are extracted from steady-state populations and closure, so they would be biased if b_F5/2 is non-negligible. The proposed fluorescence search (or an extended OBE including 2F5/2) would settle whether this systematic lands. The reader's overall CONDITIONAL verdict remains appropriate, now with an additional concrete condition: bounding or including the 2F5/2 decay channel. I therefore keep the verdict unchanged but with a sharper and more specific condition.","tokens_in":20082,"tokens_out":34008,"duration_ms":281417,"concrete_test":"Excite the 1[5/2]5/2 state with the existing 434 nm π pulse and, during the decay wait, monitor fluorescence on the 2F5/2 -> 2D5/2 (≈620 nm) and 2F5/2 -> 2D3/2 (≈572 nm) transitions using the same collection optics; an upper limit below 0.001 on the inferred direct branch would resolve the concern. Alternatively, add the 2F5/2 level to the QuTiP OBE with its known E1 decay rates and refit the Fig. 3 datasets; if the best-fit direct branch b_F5/2 is consistent with zero, the reported direct branching ratios stand.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The branching-ratio extraction assumes the 1[5/2]5/2 state decays only to 2D3/2, 2D5/2, 2S1/2, and 2F7/2. However, the 4f13 6s2 2F5/2 level (odd parity, J=5/2) lies roughly 6900 cm^-1 below the 1[5/2]5/2 state (E ~ 47,379 cm^-1), so an M1 or E2 decay 1[5/2]5/2 -> 2F5/2 is energetically allowed. Once populated, 2F5/2 decays rapidly by E1 (ΔE ~ 16,000-17,500 cm^-1) to 2D5/2 and 2D3/2, well within the 500 μs wait used before detection. This channel would therefore appear as additional population in the 2D manifold and would not be seen as a dark branch. The fit result b_F7/2 = 0.0000(2) only constrains dark branches; it does not constrain a fast-cascade channel through 2F5/2. The closure relation b_D5/2 = 1 - b_D3/2 - b_S1/2 - b_F7/2 forces the branching ratios to sum to unity, so any nonzero b_F5/2 would be reabsorbed into the inferred b_D3/2 and b_D5/2, biasing both by approximately b_F5/2 times the 2F5/2 cascade fractions and making the reported direct branching ratios effective rather than direct. Because the quoted uncertainties are ±0.002, a branch as small as 0.001-0.004 would shift the central values beyond the stated errors. The paper does not discuss or bound b_F5/2, and the detection scheme (370 nm + 935 nm) is blind to 2F5/2 fluorescence; the appendix error budget also omits this channel. Thus the main branching-ratio claims may be systematically biased.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experimental characterization of the 4f^13 5d6s ^1[5/2]_5/2 state in single 172Yb+ ions. Using heralded state preparation, coherent drives at 410, 411, 434, and 435 nm, and time-resolved detection of the 2S1/2 and 2D3/2 populations, the authors measure the state lifetime τ = 37.9(9) μs, the branching ratios to 2D3/2, 2D5/2, 2S1/2, and 2F7/2 as 0.359(2), 0.639(2), 0.0023(16), and 0.0000(2), respectively, and the Landé g-factor of the 1[5/2]5/2 state. They also report a 2D5/2→2S1/2 branching ratio of 0.188(3) and a 2D5/2 lifetime of 7.3(3) ms. The unexpectedly long lifetime is presented as evidence against the assumed 0.12 admixture of the 4f^13 5d^2 configuration in the 1[5/2]5/2 state.","tokens_in":20479,"tokens_out":12192,"duration_ms":101997,"significance":"If the results stand, this is the first precision characterization of a high-lying 'bracket' state in Yb+ that is relevant both as an atomic-structure benchmark and for metastable-state quantum information protocols. The experimental methodology is careful in several concrete ways: the lifetime is corroborated by two independent fit strategies that agree within 1%; systematic errors are propagated by Monte Carlo sampling of experimentally characterized noise sources; truncation analyses are reported for each fit; and the important external parameters (τ_D3/2, τ_D5/2, b_D5/2→F7/2) are propagated with their uncertainties. The paper also provides a full Lindblad master-equation model implemented in QuTiP and explicit pulse sequences, which aids reproducibility. These strengths make the measurement program credible; however, the branching-ratio extraction rests on a completeness assumption about the decay channels of the 1[5/2]5/2 state that is not discussed, and one of the quoted branching ratios is not statistically significant at the claimed level.","major_comments":[{"comment":"The OBE model used to fit all branching-ratio data assumes that the 1[5/2]5/2 state decays only to 2D3/2, 2D5/2, 2S1/2, and 2F7/2. In the known Yb+ level structure, the 4f13 6s2 2F5/2 level lies below the 1[5/2]5/2 state, so an M1 or E2 branch to 2F5/2 is energetically allowed. A nonzero branch would not appear as a dark branch: 2F5/2 decays by E1 into the 2D5/2 and 2D3/2 manifolds within the 500 μs wait used in the experiment, and the closure relation would fold this additional 2D population into the inferred b_D3/2 and b_D5/2. The measured b_F7/2 = 0.0000(2) constrains only dark channels and does not constrain this fast-cascade channel. Because the quoted errors in Table I are at the 0.002 level, a branch of order 0.001–0.004 would shift the central branching ratios beyond the stated uncertainties. The manuscript should provide a bound on b_F5/2 (e.g., an estimated M1/E2 rate or a dedicated search) or explicitly add this channel to the error budget.","section":"Branching ratio measurements (Fig. 3e) and Appendix Table I"},{"comment":"The fitted value b_S1/2 = 0.0023(16) is only 1.4 standard deviations from zero. The sentence in the text stating that the comparison 'suggests a non-zero value for the branching ratio of this E2 decay' overstates the evidence at the quoted precision. The appropriate statement at this significance is an upper limit (b_S1/2 < 0.007 at 3σ) or a result compatible with zero. This matters because the abstract reports b_S1/2 = 0.0023(16) on the same footing as the well-determined 0.359(2) and 0.639(2), and because b_D5/2 = 0.639(2) is obtained by closure and therefore inherits the significance of b_S1/2.","section":"Branching ratio measurements (Fig. 3e)"}],"minor_comments":[{"comment":"The caption reports Ω410/2π = 7790 kHz and Ω434/2π = 2771 kHz, whereas the main text (Fig. 3e) states Ω434/2π ≈ 7800 kHz and Ω410/2π ≈ 2800 kHz; the assignment appears swapped and should be corrected.","section":"Supplemental Material, Fig. 7 caption"},{"comment":"The term 'omg architecture' is used with inconsistent capitalization and styling across the abstract and main text; please unify the notation.","section":"Abstract and main text"},{"comment":"Reference [47] is incomplete, listing only a URL; please provide the full NIST Atomic Spectra Database citation with title, version, and access date.","section":"Reference [47]"},{"comment":"The notation b_D5/2|m=−5/2> is defined only implicitly in the text; please define it explicitly as the Clebsch-Gordan-weighted branching ratio before first use so that it is not misread as a conditional probability.","section":"Equation (1)"},{"comment":"The error budget lists τ_D3/2, τ_D5/2, and b_D5/2→S1/2 as inputs, but the paper does not state whether the values of τ_D5/2 and b_D5/2→S1/2 used in the branching-ratio fits were determined from datasets independent of the branching-ratio datasets; if the same data were used in both places, possible correlations should be addressed in the Monte Carlo propagation.","section":"Appendix, Table I"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of the journal and the experimental effort is substantial. The main risk to the central branching-ratio claims is the unmodeled 2F5/2 decay channel; this should be addressed with a bound or a clear statement that the quoted branching ratios are conditional on neglecting that channel. The b_S1/2 result also needs to be reframed as an upper limit rather than a nonzero measurement. Both issues are fixable within the scope of a revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is the first direct measurement of the lifetime, branching ratios, and g-factor of the 4f13 5d6s 1[5/2]5/2 state in 172Yb+, and the headline number—tau = 37.9(9) microseconds—is a genuine surprise for a dipole-allowed state. The work is careful and the error budget is more thorough than usual in this area. It deserves a serious referee.\n\nWhat it does well: the lifetime comes from two independent fitting methods that agree within 1%; the branching-ratio measurements use coherent population dynamics plus high-power optical pumping, with QuTiP OBE simulations and Monte Carlo sampling of systematics. The new 2D5/2 -> 2S1/2 branching ratio, 0.188(3), is a real improvement over the literature. The g-factor measurements are straightforward and consistent with expectations. The paper also frames the result correctly as a benchmark for core-valence correlation theory, which is exactly what Yb+ needs.\n\nThe soft spots. First, the abstract presents b_S1/2 = 0.0023(16) as a measured branching ratio when it is only ~1.4 sigma from zero. The text is more careful (\"suggests a non-zero value\"), but readers will quote the abstract. That needs to be qualified or presented as an upper bound.\n\nSecond—and this is the one that would make me push for revision—the branching-ratio extraction closes the decay budget over 2D3/2, 2D5/2, 2S1/2, and 2F7/2, and treats b_F7/2 = 0.0000(2) as proof that no dark branch is missing. But there is a physically allowed decay to the 4f13 6s2 2F5/2 state (odd, J=5/2, lower in energy), which would cascade rapidly to 2D3/2 and 2D5/2. That channel is invisible to the b_F7/2 search because it does not leave the population dark; it shows up as excess 2D population. The closure relation then reabsorbs it into b_D3/2 and b_D5/2, biasing both by up to the unknown branch fraction. The quoted uncertainties are +/-0.002, so even a 0.001-0.004 branch would shift the central values. The paper does not discuss or bound this channel. The authors should either estimate its rate or build it into the OBE model and show it does not move the results.\n\nMinor: no data or analysis code is attached, which makes independent verification harder; for a precision atomic measurement like this I'd want at least the decay curves and fit scripts.\n\nBottom line: the lifetime is robust and likely correct; the branching ratios are probably close, but the 2F5/2 question needs a real answer before I'd treat the reported numbers as precision values. This is a good paper for PRA; send it to a referee who knows Yb+ structure and ask for a quantitative bound on the 2F5/2 branch.","headline":"A credible first measurement of the 1[5/2]5/2 lifetime in Yb+ with real atomic-structure payoff, but the branching-ratio extraction needs a bound on a decay to 4f13 6s2 2F5/2 that the paper never mentions.","tokens_in":21168,"tokens_out":9857,"would_cite":true,"duration_ms":84695,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["32.70.Cs","32.70.Fw","37.10.Ty"],"model":"deepseek-v4-flash","headline":"A high-lying ytterbium-ion state, 1[5/2]5/2, is shown to live 37.9(9) μs and to decay mostly to the two 2D levels, with branching ratios measured to parts-per-thousand precision.","keywords":["ytterbium ion","atomic lifetime","branching ratio","optical Bloch equations","trapped ion spectroscopy","Landé g-factor","metastable state","omg architecture"],"falsifier":"Detect the 410 nm and 434 nm photons emitted by a single ion during $1[5/2]_{5/2}$ decay and histogram their arrival times; a lifetime that disagrees with $37.9(9)\\,\\mu$s, or a photon budget inconsistent with the reported branching ratios, would falsify the claim. Alternatively, repump the $^2F_{7/2}$ population directly with the 760 nm transition and look for any signal above $b(^2F_{7/2}) = 0.0000(2)$.","tokens_in":19791,"feed_emoji":"⚛️","tokens_out":11559,"duration_ms":90879,"temperature":0.7,"pith_summary":"Single $^{172}$Yb$^+$ ions are a workhorse platform for quantum computing and clocks, but the electronic structure of Yb$^+$ is too complex for existing calculations to predict reliably. The paper reports the first precision characterization of a high-lying excited state, the $1[5/2]_{5/2}$ state, by driving coherent transitions from the metastable $^2D_{3/2}$ and $^2D_{5/2}$ levels and fitting the resulting population dynamics. Its central result is that this dipole-allowed state lives $\\tau = 37.9(9)\\,\\mu$s, much longer than the naive expectation, and decays almost exclusively to the $^2D_{3/2}$ and $^2D_{5/2}$ levels (branching ratios $0.359(2)$ and $0.639(2)$), with a small direct branch to $^2S_{1/2}$ ($0.0023(16)$) and none to $^2F_{7/2}$ within resolution. If correct, this changes the picture of how the state mixes with the $4f^{13}5d^2$ configuration and opens practical uses for high-lying states in the metastable-state ('omg') architecture for trapped-ion computing.","feed_headline":"Yb+ state measured: lifetime 37.9 μs, near-closed decays","feed_subtitle":"The near-closed decay pattern revises Yb+ structure theory and opens faster trapped-ion qubit control.","key_machinery":"The central object is the $1[5/2]_{5/2}$ state, a high-lying odd-parity level of Yb$^+$ that is reached by 410 nm and 434 nm lasers from the metastable $^2D_{3/2}$ and $^2D_{5/2}$ states. The argument is carried by a Lindblad-form optical Bloch equation (the master equation for coherent drive plus spontaneous decay) that includes the Zeeman sublevels of these states; fits of its numerical solution to measured population curves yield the lifetime and branching ratios. Supporting machinery includes heralded state preparation in a single Zeeman sublevel, a 935 nm 'clearout' pulse that turns $^2D_{3/2}$ population into measurable $^2S_{1/2}$ fluorescence, and low-power versus high-power pumping regimes that isolate different branching ratios.","core_discovery":"The paper claims that the $4f^{13}5d6s\\,^1[5/2]^o_{5/2}$ state in $^{172}$Yb$^+$ is not the short-lived, leaky level the naive mixing picture implies. Instead it has spontaneous-emission lifetime $\\tau = 37.9(9)\\,\\mu$s, branching ratios $b(^2D_{3/2}) = 0.359(2)$, $b(^2D_{5/2}) = 0.639(2)$, $b(^2S_{1/2}) = 0.0023(16)$, and $b(^2F_{7/2}) = 0.0000(2)$, plus a Landé $g$-factor ratio placing $g$ near $0.983$--$0.986$ depending on the assumed value for $^2S_{1/2}$. The long lifetime makes visible Rabi oscillations on the 410 nm and 434 nm transitions from the $^2D$ manifold, and the near-closed $^2D_{3/2}$--$1[5/2]_{5/2}$--$^2D_{5/2}$ loop means population can cycle through the state without leaking to $^2F_{7/2}$. The same experiments improve the $^2D_{5/2}\\to{}^2S_{1/2}$ branching ratio to $0.188(3)$, an order-of-magnitude improvement in uncertainty. The authors emphasize that no current numerical method predicts these values, so the results question the assumed $0.12$ mixing of $4f^{13}5d6s$ with $4f^{13}5d^2$ and provide a benchmark for future calculations.","pith_inferences":["A direct photon-counting measurement of the 410 nm and 434 nm fluorescence during decay, rather than the steady-state population closure used here, would independently confirm the branching ratios and would be sensitive to any weak unmodeled decay channel.","The same pump-and-probe protocol could be applied to other odd-parity $4f^{13}5d^2$ states of Yb$^+$ that the paper names as candidates for broader cycling transitions, giving a quick survey of which states are closed enough for mid-circuit measurement.","If the branching to $^2F_{7/2}$ is genuinely zero, the $1[5/2]_{5/2}$ state may enable a leakage-free optical cycle between the two $^2D$ states, which could circumvent a shelving-fidelity limit set by the $^2D_{5/2}\\to{}^2D_{3/2}$ decay.","A dedicated ab initio calculation that isolates the $1[5/2]_{5/2}$ state would directly test whether the $0.12$ configuration-mixing value needs revision; the paper leaves that calculation as future work."],"forward_implications":["If the measured branching ratios hold, the $^2D_{3/2}$--$1[5/2]_{5/2}$--$^2D_{5/2}$ system is nearly closed, so scattering on these transitions can pump population between the two $^2D$ states without leaking to $^2F_{7/2}$.","Because the lifetime is $37.9\\,\\mu$s, coherent Rabi oscillations and resolved-sideband operations at 410 nm and 434 nm are feasible, turning these transitions into usable qubit-manipulation tools in the omg architecture.","A near-zero $^2F_{7/2}$ branch means $1[5/2]_{5/2}$ can serve as a fast dissipative initialization path into the metastable-state qubit, replacing the slow $^2D_{5/2}\\to{}^2F_{7/2}$ natural decay.","The improved $^2D_{5/2}\\to{}^2S_{1/2}$ branching ratio of $0.188(3)$ sharpens models of shelving and decay in Yb$^+$ by an order of magnitude.","The lifetime and branching-ratio data give atomic-structure theory a concrete target: a calculation that reproduces these values would validate a treatment of core-valence correlation in Yb$^+$."],"supporting_citations":[{"why":"Supplies the fixed $^2D_{5/2}$ lifetime, the $^2D_{5/2}\\to{}^2F_{7/2}$ branching ratio used as an input, and the comparison value for the remeasured $^2D_{5/2}$ lifetime.","marker":"[36]"},{"why":"Supplies the fixed $^2D_{3/2}$ lifetime used in the optical-Bloch-equation fits.","marker":"[44]"},{"why":"Prior single-ion spectroscopy that measured the absolute frequencies of the 410 nm transition, providing the frequency baseline this work excites.","marker":"[31]"},{"why":"Source of the $0.12$ configuration mixing between $4f^{13}5d6s$ and $4f^{13}5d^2$ that sets the naive lifetime expectation being questioned.","marker":"[32]"},{"why":"Describes the omg metastable-state architecture that motivates the practical relevance of high-lying states like $1[5/2]_{5/2}$.","marker":"[28]"},{"why":"Provides the 760 nm repumping and electron-shelving method used to reinitialize population from $^2F_{7/2}$ after each experimental run.","marker":"[30]"},{"why":"Supplies the numerical integrator used for the optical-Bloch-equation fits that extract the lifetime and branching ratios.","marker":"[38]"}],"fun_headline_variants":["Yb+ state's 37.9 μs lifetime enables visible Rabi oscillations","Near-closed Yb+ decay loop measured, improving qubit control","Yb+ branchings: D5/2 0.639, D3/2 0.359, S1/2 0.0023","Yb+ excited state shows no leak to F7/2, long lifetime"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The extraction of every reported number presumes the optical-Bloch-equation model includes the complete set of decay channels and Zeeman sublevels, so an unmodeled decay path from $1[5/2]_{5/2}$, or an error in the literature values adopted for the two $^2D$ lifetimes and the $^2D_{5/2}\\to{}^2F_{7/2}$ branching, would shift the quoted branching ratios.","fun_headline_variants_meta":{"raw":{"variants":["Yb+ state's 37.9 μs lifetime enables visible Rabi oscillations","Near-closed Yb+ decay loop measured, improving qubit control","Yb+ branchings: D5/2 0.639, D3/2 0.359, S1/2 0.0023","Yb+ excited state shows no leak to F7/2, long lifetime"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001519,"raw_usage":{"total_tokens":6268,"prompt_tokens":1308,"completion_tokens":4960,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":924,"completion_tokens_details":{"reasoning_tokens":4857}},"tokens_in":924,"tokens_out":4960,"duration_ms":31452,"temperature":1.0,"reasoning_tokens":4857,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:45:03.515420+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Detect the 410 nm and 434 nm photons emitted by a single ion during $1[5/2]_{5/2}$ decay and histogram their arrival times; a lifetime that disagrees with $37.9(9)\\,\\mu$s, or a photon budget inconsistent with the reported branching ratios, would falsify the claim. Alternatively, repump the $^2F_{7/2}$ population directly with the 760 nm transition and look for any signal above $b(^2F_{7/2}) = 0.0000(2)$.","supporting_citations":[{"cited_title":"Taylor, M","cited_arxiv_id":null,"evidence_quote":"Supplies the fixed $^2D_{5/2}$ lifetime, the $^2D_{5/2}\\to{}^2F_{7/2}$ branching ratio used as an input, and the comparison value for the remeasured $^2D_{5/2}$ lifetime."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the fixed $^2D_{3/2}$ lifetime used in the optical-Bloch-equation fits."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior single-ion spectroscopy that measured the absolute frequencies of the 410 nm transition, providing the frequency baseline this work excites."},{"cited_title":"Ralchenko.NIST atomic spectra database","cited_arxiv_id":null,"evidence_quote":"Source of the $0.12$ configuration mixing between $4f^{13}5d6s$ and $4f^{13}5d^2$ that sets the naive lifetime expectation being questioned."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the 760 nm repumping and electron-shelving method used to reinitialize population from $^2F_{7/2}$ after each experimental run."},{"cited_title":"Johansson, P","cited_arxiv_id":null,"evidence_quote":"Supplies the numerical integrator used for the optical-Bloch-equation fits that extract the lifetime and branching ratios."}],"review_version":1}