{"id":"29ca6beb-12f7-47b7-b9bb-d4412cad50c3","arxiv_id":"2505.18462","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Pulsed pumping of a hybrid alkali-noble-gas comagnetometer suppresses technical noise by about 38.5% and lowers low-frequency magnetic-field response by about 51.1%, supported by a three-phase analytical model.","lead":"A K-Rb-21Ne atomic comagnetometer is operated with pulsed rather than continuous laser pumping, and the authors report that this dynamic polarization scheme suppresses pump-light-induced polarization noise by about 38.5% while reducing low-frequency magnetic field response by up to 51.1%. They present a three-phase model of the spin dynamics and argue the approach could push such sensors beyond the standard quantum limit.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The suppression percentages rest on a fixed-frequency Phase III fit whose frequency-drift bias is untested; a chirp-model comparison is needed.","rationale":"The reader's weakest_assumption identifies the same load-bearing concern: the fixed-frequency approximation in Phase III. My stress-test agrees and sharpens it. Eq. (2) and Eq. (3) treat the precession frequency as constant, but the paper's own Phase III model has Q(t) varying while P_e_z decays, so the frequency necessarily changes within the selected fitting window. The paper's evidence for 'slow' variation is one RMSE value, which does not bound the frequency drift or rule out correlated bias in the fitted amplitudes. Since the suppression percentages are ratios of fitted A_sin/A_cos terms at different duty ratios, a duty-ratio-dependent bias in those fits would directly corrupt the headline numbers. This is a falsifiable concern: a synthetic-data test or a chirp-model refit would settle it. The paper otherwise reports a plausible experiment with independent evidence: the Earth-rotation response, the linear B_n0-versus-duty-ratio relation, and the consistency of Q^(III) across duty ratios. Those support the qualitative picture, but the quantitative suppression claims need the proposed check before being taken as definitive. The reader's CONDITIONAL verdict therefore remains appropriate; I do not see grounds to strengthen or weaken it beyond that.","tokens_in":17064,"tokens_out":4219,"duration_ms":38039,"concrete_test":"Generate synthetic Phase III waveforms using the measured R_e1^(III), R_e2^(III), Q^(III) and the model's time-dependent Q(t) for each duty ratio, with known input Ω_y and B_y amplitudes; fit them with Eq. (3). If the recovered A_sin/A_cos deviate from the inputs by more than about 10%, or if the deviation changes materially between 40% and 60% duty ratio, the reported suppression percentages are biased by the fixed-frequency approximation. Alternatively, if the raw data are shared, refit the Phase III records with f(t)=f0+βt and test whether β is statistically nonzero and whether the fitted A_sin/A_cos shift by more than 10%; if they do, the central comparisons need to be re-derived with a chirp-aware analysis.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims—38.5% polarization-noise suppression and 51.1% low-frequency field suppression—are extracted from A_sin/A_cos amplitudes obtained by fitting Eq. (3) to a 20–35 ms window of Phase III. Eq. (3) assumes a strictly constant frequency f = γ_e(B_z+B_n0)/Q^(III), but the paper's own model makes Q^(III) time-dependent because the longitudinal electron polarization continues to decay through the window. The only justification is the qualitative statement that the frequency 'changes slowly,' supported by a single RMSE (4.01e-5 V) for one representative fit. No bound on the actual frequency drift is given, and no comparison is made with a fit that includes a time-dependent phase, e.g. f(t)=f0+βt or ∫f dt. If f varies within the window, the eight-parameter fit can trade frequency drift against A_sin, A_cos, and φ_Ω. Because the decay trajectory at Phase III start depends on duty ratio, any such bias would shift the suppression percentages differentially across the five duty ratios—exactly the ratios that define the headline results. The 'complete analytical solution' is also only complete after this linearization; the fitting-window choice (20–35 ms) is not independently justified. Therefore the numerical suppression values are not yet established at the claimed precision.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports a K-Rb-21Ne SERF comagnetometer operated with pulsed pump light at 11 Hz, with duty ratios varied from 40% to 60%. The authors propose a three-phase evolutionary model of the coupled electron-nuclear spin dynamics, present a magnetic-field compensation procedure based on the pump-on steady state, and characterize the response to inertial rotation, low-frequency magnetic fields, and pump-light intensity fluctuations during the pump-off phase. The headline quantitative results are an averaged 38.5% suppression of pump-light-induced polarization noise and an averaged 51.1% suppression of the low-frequency magnetic-field response when the duty ratio is increased from 40% to 60%.","tokens_in":17355,"tokens_out":6451,"duration_ms":51826,"significance":"If the quantitative claims are correct, the pulsed-pump scheme is a useful technical addition to SERF comagnetometry: it reduces sensitivity to pump-light fluctuations and low-frequency magnetic fields without abandoning self-compensation. The experimental work is substantial: the setup includes a K-Rb-21Ne cell, AOM-based pump modulation, and a full three-axis compensation procedure, and the responses are measured at five duty ratios. The paper's honesty about limiting factors (magnetic-field stability near the cell, electric heating, and the small Phase III scale factor) is commendable. The main weakness is that the central suppression percentages are extracted from fits to a linearized damped-oscillation model whose key assumption—slowly varying precession frequency—is asserted but not quantitatively validated. The 'complete analytical solution' is actually an approximate piecewise model with many fitted parameters, and the manuscript does not provide code or processed data tables, only data available upon request.","major_comments":[{"comment":"The suppression percentages (38.5% and 51.1%) are derived from the fitted amplitudes A_sin and A_cos in Eq. (3), which assumes a strictly constant precession frequency f = gamma_e(B_z+B_n0)/Q^(III). The paper's own model makes Q^(III) time-dependent through the decaying longitudinal electron polarization, and the stated value Q^(III)=10.27 corresponds to P_e^z ~ 18%, which is not 'near zero.' The only justification for the linearization is the qualitative statement that the frequency 'changes slowly,' plus one RMSE value (4.01e-5 V) for a single fit. This is insufficient: a fit with a time-dependent phase (e.g., f(t)=f0+beta t) or fits in sliding sub-windows should be compared to Eq. (3) to bound the frequency drift over the 20–35 ms window. Because the decay trajectory at the start of Phase III depends on the duty ratio, any frequency-drift bias would enter A_sin and A_cos differentially across the five duty ratios and could shift the headline suppression values.","section":"II.D, Eq. (2), and II.E, Eq. (3)"},{"comment":"The claim of a 'complete analytical solution' for the spin dynamics is overstated. The three-phase model is a piecewise approximation: Phase I uses an equivalent mean pumping rate, Phase II uses the steady state, and Phase III explicitly 'can be approximated as a linear evolution process' (Methods). The resulting expressions contain many fitted parameters (Q, R_e1/Q, R_e2/Q, f, B_n0, A, B, C, phi, A_sin, A_cos, k_e) and are not closed-form in the usual sense. Recommend rephrasing to 'approximate analytical solution' and stating the validity conditions quantitatively.","section":"Abstract and METHODS, Eq. (8)"},{"comment":"The 51.1% suppression figure is described only as an 'average' over the low-frequency magnetic field responses, but the averaging rule is not given. It is not specified how the ratio is computed across the tested frequency range (0.005–1 Hz), across the Bx and By axes, or across the five duty ratios, nor is an uncertainty or confidence interval provided. The figure shows outliers attributed to 'system drift' without quantitative criteria. Please provide a table of the per-frequency, per-axis, per-duty-ratio response coefficients used to compute the 51.1% average, and report the standard error of the mean.","section":"II.F, Fig. 6"},{"comment":"The 38.5% suppression of pump-light fluctuations is reported as an 'overall average of all collected data,' but only two duty ratios (50% and 55%) are shown, and no statistical uncertainty or number of measurements is given. The definition of the 'relative response coefficient' also needs clarification: it is the slope of light-intensity fluctuation against A_cos or P_e^(II)_x, divided by the scale factor with respect to Omega_y, but the units of the light-intensity fluctuation (photodiode voltage) and the normalization must be stated explicitly. Please report the per-frequency values and their errors.","section":"II.G, Fig. 7"}],"minor_comments":[{"comment":"Consider plotting the residuals of the fit; one RMSE value for a single representative trace does not convey the fit quality across duty ratios.","section":"Fig. 4(a)"},{"comment":"The notation {R_e2, R_e2, R_e1} is nonstandard; please define it as a diagonal matrix acting on the polarization vector.","section":"METHODS, Eqs. (6)-(8)"},{"comment":"Report the fitted slope k and its uncertainty for the linear relation B_n0 versus duty ratio.","section":"Inset of Fig. 4"},{"comment":"The differential measurement with points 180 degrees apart is described too briefly; specify how the two points are combined to remove bias.","section":"II.E"},{"comment":"Define all symbols in Eq. (5), including the units of S, K_Omega, and the noise terms.","section":"II.H, Eq. (5)"}],"recommendation":"major_revision","confidential_remarks":"I support publication after major revision. The experimental apparatus and measurement campaign are solid, but the two headline suppression percentages are not yet established at the claimed precision because of the unquantified frequency drift in Phase III and the unspecified averaging procedures. The authors should also temper the 'complete analytical solution' claim. No concerns about authorship or scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The paper reports a working K-Rb-21Ne SERF comagnetometer run with pulsed pump light, and that's genuinely new relative to the two earlier pulsed-pumping papers (single-species Rb-Xe and Rb-Ne). The hybrid pumping is a sensible design choice for polarization homogeneity, and the three-phase model—though advertised as 'complete'—is a useful approximate tool. The experimental work is real: Earth-rotation calibration, response curves at five duty ratios, and direct measurements of the pump-light-modulation response and low-frequency magnetic-field response. If you work on comagnetometers, this is worth reading.\n\nThe main soft spot is exactly the one the stress-test flags. The headline suppression percentages (38.5% and 51.1%) are extracted from fitting Eq. 3, a fixed-frequency damped sinusoid, to a 20–35 ms window in Phase III. But the paper's own model has Q(III) time-dependent because the longitudinal electron polarization keeps decaying through that window. The authors justify the fixed-frequency approximation with one sentence and one RMSE value. That's not adequate. If the frequency drifts inside the window, the fit can trade drift against A_sin, A_cos, and the phase, and since the decay trajectory at Phase III entry depends on duty ratio, the bias wouldn't necessarily cancel in the suppression ratios. They need to show a chirp fit (for instance, linear frequency ramp) gives the same amplitudes or bound the frequency drift directly.\n\nOther soft spots are minor. The 'complete analytical solution' is an approximation with many fitted parameters; the suppression percentages come without error bars or per-point scatter; the sensitivity improvement is limited to below 0.1 Hz and the paper says so. The data availability statement is 'on request,' no code or repository, which is common but not great. The abstract's talk about overcoming the quantum noise limit is speculative; it's not demonstrated.\n\nAll that said, the central claim—that pulsed pumping can reduce pump-related technical noise in a hybrid SERF comagnetometer—is plausible and supported by the direction of the measurements. I don't think the paper is wrong; I think it needs a harder look at the frequency-drift assumption and honest error analysis on the headline numbers.\n\nWho's this for? Anyone in atomic magnetometry or comagnetometry, especially those working on pulsed-pumping schemes. It deserves a serious referee. I'd send it to review, but I'd insist the authors add the chirp comparison and confidence intervals before accepting.","headline":"A useful experimental extension of pulsed-pumping SERF comagnetometry to a hybrid K-Rb-21Ne system, with credible direct measurements but headline suppression numbers that need error bars and a chirp check before I'd take them at face value.","tokens_in":17931,"tokens_out":2979,"would_cite":true,"duration_ms":26438,"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":"A K-Rb-21Ne SERF comagnetometer can be run in a pulsed, dynamically polarized mode that suppresses pump-light polarization noise by 38.5% and low-frequency magnetic-field response by 51.1% when the duty ratio is raised from 40% to 60%…","keywords":["atomic spin sensor","spin-exchange relaxation-free","dynamically polarized comagnetometer","pulsed pump light","three-phase evolutionary model","self-compensation","K-Rb-21Ne","damped oscillation"],"falsifier":"Measure the instantaneous frequency of the Phase III damped signal across the 20-35 ms window using a short-time Fourier transform or Hilbert transform while holding the duty ratio fixed. If the instantaneous frequency changes by more than the frequency resolution within that window, the fixed-frequency linear assumption fails and the fitted A_sin and A_cos values are biased. A complementary check is to integrate the full Bloch equations with a density-matrix solver and compare the recovered amplitudes with the analytical fixed-frequency fit; disagreement indicates model bias.","tokens_in":1952,"feed_emoji":"🧲","tokens_out":2444,"duration_ms":78631,"temperature":0.7,"pith_summary":"The paper claims that a spin-exchange relaxation-free (SERF) comagnetometer can be operated with pulsed pump light instead of continuous pumping, and that this mode shields the measurement from pump-light fluctuations while retaining sensitivity to rotation and magnetic fields. The dynamics are split into three phases: nuclear polarization settles to a value proportional to the duty ratio, electron spins reach a pump-on steady state, and after pump-off they perform a damped free precession that can be treated as a fixed-frequency linear process once the electron polarization has decayed. On this basis the authors report an average 38.5% suppression of polarization noise and an average 51.1% suppression of low-frequency magnetic-field response when the duty ratio is increased from 40% to 60%. A sympathetic reader would care because technical noise from pump light is a known bottleneck in ultra-precise magnetometers, and the pulsed scheme offers a way to reduce that noise without giving up self-compensation.","feed_headline":"Pulsed pump light cuts comagnetometer noise by 38.5%","feed_subtitle":"Pulsed pumping also suppresses low-frequency magnetic-field response by 51.1% while keeping self-compensation.","key_machinery":"The load-bearing object is the three-phase evolutionary model for the coupled alkali-noble-gas spin ensemble. In Phase I, nuclear spin polarization is described by exponential polarization and depolarization with a constant relaxation rate, giving a steady longitudinal nuclear polarization proportional to the duty ratio, $P_n^{{(I)}}$_z = D_r $P_n^{{(DC)}}$_z. Phase II is the pump-on steady state of the electron spins, obtained from the Bloch equations with the real pump rate; it supplies the compensation point and the initial conditions for free precession. Phase III is the pump-off damped oscillation, modeled after the electron polarization has decayed to near zero as a fixed-frequency linear process, fit with the formula $P_x^{{(III)}}$(t) = exp(-$R_2^{{(III)}}$ t / $Q^{{(III)}}$) [A_sin sin(2π f t + φ_Ω) + A_cos cos(2π f t + φ_Ω)] + B exp(-$R_1^{{(III)}}$ t / $Q^{{(III)}}$) + C. The amplitudes A_sin and A_cos are the readouts that separate the two transverse axes, and the assumption of slow frequency drift in the chosen 20-35 ms window is what makes the analytical fit possible.","core_discovery":"The paper's central claim is that a SERF comagnetometer can be driven by pulsed pump light rather than continuous pumping, and that the resulting dynamic polarization is analytically tractable and technically advantageous. In the K-Rb-21Ne setup, K electron spins are pumped along z and transfer polarization to Rb and then to 21Ne; once the pump is switched off at 11 Hz, the electron spins depolarize within tens of milliseconds while the nuclear polarization remains nearly fixed, leaving the electron spins to undergo damped free precession around the residual field. The authors divide this cycle into three phases and derive analytical solutions for each: nuclear polarization settles to a value proportional to the duty ratio (Phase I), electron spins reach a DC-like steady state while the pump is on (Phase II), and after pump-off the damped oscillation in a frequency-stable window (Phase III) is fit with a fixed-frequency sinusoid whose sine and cosine amplitudes separate the x- and y-axis responses. With this model they report an average 38.5% suppression of pump-light-induced polarization noise and an average 51.1% suppression of the low-frequency magnetic-field response when the duty ratio goes from 40% to 60%, and they show that the same compensation point works in both pump-on and pump-off phases.","pith_inferences":["The same three-phase split would likely transfer to other hybrid alkali-noble-gas sensors with widely separated electron and nuclear relaxation times, so the model may describe Rb-Xe comagnetometers and pulsed atomic clocks that measure free precession.","Because raising the duty ratio suppresses low-frequency magnetic-field response but also lowers the rotation scale factor, an optimal duty ratio should exist for a given magnetic-field stability budget; the paper does not optimize this trade-off.","The pump-off measurement window creates a natural stroboscopic geometry, so combining the scheme with squeezed or entangled probe light could in principle reach below the standard quantum limit, though the paper only mentions such extensions as future work.","The reported 38.5% and 51.1% values are averages over the tested duty-ratio and frequency ranges; a direct extension would map these suppression percentages as functions of pump modulation frequency and duty ratio and compare them with the model's predictions at other operating points."],"forward_implications":["Operating the comagnetometer in the pump-off phase reduces the response to pump-light intensity fluctuations: the reported average suppression is 38.5% relative to the pump-on steady state.","Raising the duty ratio from 40% to 60% increases the nuclear longitudinal polarization in proportion to the duty ratio and suppresses the low-frequency magnetic-field response by an average 51.1%, so dynamic polarization preserves self-compensation.","The fitted amplitudes A_sin and A_cos separate the responses to rotations about two orthogonal axes, enabling dual-axis inertial rotation measurement during free precession.","In the low-frequency band below about 0.1 Hz the pulsed scheme improves sensitivity compared with the pump-on steady state; above 0.1 Hz sensitivity is limited by background noise and magnetic-field instability near the cell.","Using the fitting parameters instead of a single time point of the damped signal reduces the noise floor by about 36.3%."],"supporting_citations":[{"why":"Establishes the hybrid K-Rb pumping approach and its benefit for polarization homogeneity and nuclear spin relaxation, motivating the cell design.","marker":"[1]"},{"why":"Shows the feasibility of dual-axis inertial rotation and magnetic-field measurement from free precession in a pulsed SERF scheme, the approach this paper extends with a three-phase model.","marker":"[35]"},{"why":"Reports a 87Rb-21Ne pulsed-pumping comagnetometer with dual-axis measurements and pump/probe pointing-noise suppression, providing the baseline that hybrid pumping and three-phase modeling improve.","marker":"[36]"},{"why":"Supplies the relation between the slowing-down factor and electron polarization used to extract the approximately 18% longitudinal polarization in Phase III.","marker":"[37]"},{"why":"Provides the coupled Bloch-equation framework and self-compensation concept for DC-mode SERF comagnetometers that Phases I and II generalize.","marker":"[38]"},{"why":"Gives the DC-mode steady-state rotation response used to validate the Phase II response to Earth's rotation.","marker":"[42]"},{"why":"Defines the pump-light intensity-modulation procedure and relative response coefficient used to quantify the 38.5% polarization-noise suppression.","marker":"[43]"},{"why":"States the DC-mode steady-state solution that the dynamic model must reduce to in the single-phase limit, used to check the calculation.","marker":"[58]"}],"fun_headline_variants":["Dynamic polarization cuts magnetometer noise 38.5%","Pulsed pumping quells noise and field response in SERF","Comagnetometer dynamic spin state tames technical noise","Pump pulsing suppresses noise 38.5%, field 51.1%"],"cache_read_input_tokens":19968,"weakest_assumption_plain":"The Phase III fits assume that after the electron polarization has decayed to near zero, the precession frequency changes slowly enough over the 20-35 ms fitting window that the damped oscillation can be treated at fixed frequency; if the frequency drifts within that window, the fitted amplitudes A_sin and A_cos are biased in a way that would distort the reported suppression percentages.","fun_headline_variants_meta":{"raw":{"variants":["Dynamic polarization cuts magnetometer noise 38.5%","Pulsed pumping quells noise and field response in SERF","Comagnetometer dynamic spin state tames technical noise","Pump pulsing suppresses noise 38.5%, field 51.1%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000527,"raw_usage":{"total_tokens":2552,"prompt_tokens":962,"completion_tokens":1590,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":1517}},"tokens_in":578,"tokens_out":1590,"duration_ms":12066,"temperature":1.0,"reasoning_tokens":1517,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:30:36.959539+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the instantaneous frequency of the Phase III damped signal across the 20-35 ms window using a short-time Fourier transform or Hilbert transform while holding the duty ratio fixed. If the instantaneous frequency changes by more than the frequency resolution within that window, the fixed-frequency linear assumption fails and the fitted A_sin and A_cos values are biased. A complementary check is to integrate the full Bloch equations with a density-matrix solver and compare the recovered amplitudes with the analytical fixed-frequency fit; disagreement indicates model bias.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the hybrid K-Rb pumping approach and its benefit for polarization homogeneity and nuclear spin relaxation, motivating the cell design."},{"cited_title":"Hedges, A","cited_arxiv_id":null,"evidence_quote":"Shows the feasibility of dual-axis inertial rotation and magnetic-field measurement from free precession in a pulsed SERF scheme, the approach this paper extends with a three-phase model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports a 87Rb-21Ne pulsed-pumping comagnetometer with dual-axis measurements and pump/probe pointing-noise suppression, providing the baseline that hybrid pumping and three-phase modeling improve."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the relation between the slowing-down factor and electron polarization used to extract the approximately 18% longitudinal polarization in Phase III."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the coupled Bloch-equation framework and self-compensation concept for DC-mode SERF comagnetometers that Phases I and II generalize."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the DC-mode steady-state rotation response used to validate the Phase II response to Earth's rotation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the pump-light intensity-modulation procedure and relative response coefficient used to quantify the 38.5% polarization-noise suppression."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the DC-mode steady-state solution that the dynamic model must reduce to in the single-phase limit, used to check the calculation."}],"review_version":1}