{"id":"1b5503d6-a602-41af-ac4f-93e5027ea3ff","arxiv_id":"2607.18808","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Radiation-pressure acceleration and heating during the probe pulse explain the discrepancy between in-situ coherence spectra of magnetically trapped 87Rb and time-of-flight temperatures.","lead":"Using light-scattering coherence measurements on a rubidium cloud inside a magnetic trap, the authors show that the probe laser itself accelerates and heats the atoms, explaining why the measured spectral width disagrees with time-of-flight thermometry. A simple model that adds these probe effects brings the two measurements into agreement, supporting coherence spectroscopy as an in-situ thermometer.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Model validated only against window-difference data, so static broadening mechanisms remain untested — absolute spectra needed.","rationale":"The reader's weakest assumption focused on the T_r/3 heating rate being unstated and the TOF temperature being used as an input rather than extracted. Both are valid, but the most load-bearing issue is more specific: the paper's only quantitative model-data comparison is on the difference between two integration windows (Fig. 3). Difference measurements are insensitive to any static or slowly varying broadening source, including the very magnetic-trap effects the paper claims to rule out. The gradient-independence and trap-on/off checks address gradient-dependent mechanisms, but not a constant offset. Therefore the central explanatory claim — that the full discrepancy is due to probe dynamics — cannot be assessed from the presented data. The model itself is physically reasonable and parameter-free after fixing T; the saturation dependence of the differences is encouraging. The fix is straightforward: report and compare absolute linewidths and center frequencies, or overlay measured and simulated spectra. Until that is done, the 'reliable in situ probe' claim remains conditional. This does not change the reader's verdict, but sharpens the condition.","tokens_in":9487,"tokens_out":8623,"duration_ms":88915,"concrete_test":"Re-analyze the stored g1(τ) data: for each saturation setting, fit Eq. (4) to the 100 µs and 300 µs windows separately, then compare the absolute f0 and Δf1 to the model's predictions (using T=85 µK and the same s, tp) without subtracting the 100 µs values. The concern is settled if the model reproduces both absolute quantities within the fit uncertainties (roughly ±5 kHz) for all s; if deviations exceed that, an unmodeled static broadening mechanism is present and the central claim needs revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that probe-induced radiation pressure and heating fully explain the discrepancy between the measured g1 linewidth (Δf1 = 332±5 kHz at 300 µs, §'Probe-Driven Atomic Dynamics') and the thermal expectation from Eq. (5) (192±11 kHz). But the quantitative validation in Fig. 3 compares only the changes between the 100 µs and 300 µs integration windows — Δf0 and Δf1 — for both experiment and model. Any time-independent contribution to the linewidth (e.g., residual magnetic-field inhomogeneity, multiple scattering, or an incorrect initial temperature) cancels exactly in these differences. The model with T=85 µK could therefore match Fig. 3 while predicting the wrong absolute linewidth by tens of kHz, leaving the 'fully explained' claim unsupported. The paper never shows measured and simulated spectra or absolute fit parameters side by side; Fig. 4 shows simulated spectra only. Since the abstract and conclusion generalize to a 'reliable in situ probe,' this missing absolute comparison is the load-bearing gap.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports in-situ temporal coherence measurements of light scattered by a 87Rb cloud in a quadrupole magnetic trap. A Gaussian fit to the heterodyne g^(1) spectrum yields a width (332±5 kHz) and a central-frequency shift that deviate from the simple thermal expectation of Eq. (5) (192±11 kHz) for the independently measured TOF temperature of 85±10 μK. The authors attribute this discrepancy to radiation-pressure-induced acceleration and heating during the probe pulse. They support this by comparing, for two acquisition windows (100 μs and 300 μs), the window-difference frequency shift and linewidth broadening against a simple model that adds a Doppler shift and a T_r/3 heating per scattering event. They also report that these quantities show no dependence on the magnetic-field gradient. The paper claims that, once probe-induced dynamics are accounted for, coherence spectroscopy is a quantitative in-situ probe of velocity distributions in trapped ensembles.","tokens_in":9717,"tokens_out":2502,"duration_ms":29149,"significance":"If the claimed agreement is robust, the result would be valuable: it would demonstrate that light-scattering correlation spectroscopy can be applied to atoms in a conservative magnetic trap, and that probe back-action can be modeled and subtracted. The work extends prior diffusing-wave spectroscopy and cold-atom coherence studies to a new trapping geometry. The paper has concrete strengths: the use of two integration windows within the same experimental run and the variation of the magnetic-field gradient are good experimental controls, and the model is simple and parameter-free apart from the assumed heating rate and the externally supplied initial temperature. Machine-checkable code is not included, but the model is sufficiently transparent that it can be reproduced. The main limitation is that the quantitative validation is performed only on differences between windows, while the central claim concerns the absolute linewidth and the capability to extract velocity distributions in situ.","major_comments":[{"comment":"The quantitative comparison in Fig. 3 uses only the differences Δf0 and Δf1 between the 100 μs and 300 μs acquisition windows. Any time-independent contribution to the measured linewidth (residual magnetic inhomogeneity, multiple scattering, an incorrect initial temperature, or an unmodeled static broadening) cancels in these differences. The central claim that probe-induced acceleration and heating 'fully explain' the measured absolute width is therefore not actually tested. The authors should plot the measured and simulated absolute spectra or absolute fit parameters (Δf1 and f0 for each window) side by side, or at least provide a table of absolute values for the same conditions as Fig. 3.","section":"Probe-Driven Atomic Dynamics, Fig. 3"},{"comment":"The model assumes a heating rate of T_r/3 per scattering event, but no derivation or reference is given. The standard result for momentum diffusion from random recoil kicks would need to be stated explicitly, and the factor 1/3 (rather than, say, 2/3 or a full recoil temperature) must be justified. Since the broadening in the model depends directly on this parameter, an unsupported choice weakens the claim of quantitative agreement. Please derive T_r/3 from the scattering process or cite the relevant treatment, and show the sensitivity of the predicted Δf1 to this parameter.","section":"Eq. (7) and subsequent model"},{"comment":"The paper states that measurements were made for four different magnetic-field gradients and after 2 ms of time of flight, and that 'no clear dependence' was observed, but no data or plot for this null test is shown. This claim is used to exclude magnetic-trap-related broadening, which is a necessary part of the argument that the discrepancy arises from probe back-action. The authors should show these results (e.g., Δf1 and f0 versus ΔB∥) or include a representative comparison in the main text or supplement.","section":"Probe-Driven Atomic Dynamics, magnetic-field-gradient test"},{"comment":"The paper concludes that coherence spectroscopy is a 'reliable in situ probe of velocity distributions,' but the initial temperature T=85±10 μK is taken from TOF measurements and used as an input to the model, rather than being extracted from the coherence data. The current analysis therefore demonstrates that the model can reproduce the observed trends when T is known, but not that the measurement alone can determine T. To support the claimed inversion capability, the authors should either perform an inversion test (treating T as a free parameter and showing that the fit returns the TOF value) or explicitly state this limitation in the conclusion.","section":"Perspectives and Conclusion"}],"minor_comments":[{"comment":"The text says the dominant contribution is the |F=2,m_F=2> to |F'=3,m_F'=2> transition, but the saturation intensity is computed for the stretched-state σ± transition with a Clebsch-Gordan factor 1/3. Clarify which transition is used for the saturation parameter and how the polarization selection affects the Clebsch-Gordan coefficients.","section":"Experimental setup"},{"comment":"The expression for g_BN^(2)(τ) has a term proportional to |g_sc^(1)(τ)| cos(2π f_BN τ). The sign convention and the relative phase of the local oscillator are not discussed. Please state the phase convention or note that only the amplitude around f_BN is used.","section":"Eq. (3)"},{"comment":"The figure shows a 'clear frequency shift and a reduced HWHM' for the shorter probe duration, but the text refers to a broadening with increasing probe duration. Consider rewording the caption to be consistent: the 100 μs window has a narrower line than the 300 μs window.","section":"Fig. 2 caption and text"},{"comment":"The model sums Gaussian spectra after each scattering event, but the description does not specify whether the initial velocity distribution is Maxwell-Boltzmann at T=85 μK, whether the Doppler shift is applied to the whole spectrum or only to the scattered photon, and how the random direction of spontaneous emission is treated. A short equation or algorithm outline would make the model reproducible.","section":"Probe-Driven Atomic Dynamics"},{"comment":"The paper would benefit from a statement of the number of experimental repetitions and how the error bars in Fig. 3 were obtained (e.g., fit uncertainty only or shot-to-shot scatter). This is important for judging the claimed 'agreement within error bars.'","section":"General"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short take: this is the first g1 coherence measurement on atoms confined in a purely magnetic trap, and the paper gives a clean, parameter-free explanation of why the coherence linewidth disagrees with TOF thermometry: the probe itself accelerates and heats the cloud. The gradient-independence check is genuinely useful evidence that the trap isn't the culprit. Credit where due: the experiment is demanding (day-long acquisitions), the single-scattering conditions are checked, and the model has essentially no fitted constants—T0 comes from TOF, s from laser intensity, N_ex from the standard scattering rate. That's reproducible, falsifiable work in spirit, and the saturation-dependent shift/broadening trend in Fig. 3 is convincing.\n\nThe soft spots are real but addressable. The main one is exactly what the stress-test flags: the quantitative comparison in Fig. 3 uses only differences between the 100 and 300 µs windows. Any static contribution to the linewidth—residual field inhomogeneity, multiple scattering, an incorrect input temperature—cancels in those differences. The model with T=85 µK could fit the trend while missing the absolute linewidth by tens of kHz. The paper never shows a measured and simulated spectrum side by side; Fig. 4 is simulated only. So the phrase 'fully explained' is stronger than what the data shown can support. The fix is straightforward: show the absolute f0 and Δf1 at fixed s along with the model curve, or overlay one real spectrum with the simulation.\n\nMinor but worth saying: the T_r/3 heating per scattering event is asserted, not derived, and the dynamical model is sketched rather than specified equation by equation. No data or code are released, which makes the check harder. These are not fatal; they're the usual gap between a short letter and a methods paper.\n\nFor who this is for: people working on non-destructive probing of cold atoms, coherence spectroscopy, and far-from-equilibrium dynamics would get real use from this. The context-setting around TOF limitations (interaction energy conversion during expansion) is well made. The claim that this is a 'reliable in situ probe' is plausible but should be treated as provisional until the absolute comparison appears.\n\nDeserves peer review, yes. I'd send it out. And I'd ask the authors for the absolute comparison and the heating-rate derivation before accepting. Would I cite it? Probably, as the first magnetic-trap coherence measurement, with a caveat about the validation. Reading group: maybe—the model-vs-difference issue is a useful teaching case.","headline":"First coherence measurement in a purely magnetic trap, with a plausible radiation-pressure/heating model, but the quantitative case rests on difference data, so the 'fully explained' claim is not yet closed.","tokens_in":10223,"tokens_out":1774,"would_cite":true,"duration_ms":17663,"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 discrepancy between in-situ coherence spectra and time-of-flight temperatures in a magnetically trapped cold cloud is fully explained by radiation pressure and heating from the probe itself, establishing coherence spectroscopy as a quan","keywords":["temporal coherence spectroscopy","first-order correlation function","cold atoms","magnetic quadrupole trap","radiation-pressure acceleration","recoil heating","in-situ thermometry","time-of-flight"],"falsifier":"Measure the g^(1) spectrum with the probe switched to the red side of resonance: radiation pressure then pushes atoms opposite the probe direction, so the accumulated Doppler shift should change sign with the same magnitude predicted by Eq. (7), while the heating-induced broadening stays the same. If the sign or magnitude of the shift does not follow N_ex k_L (1 - cosθ) v_r, the acceleration mechanism is not the full explanation. Alternatively, invert the model for several probe durations and saturation values and check that the recovered starting temperature is identical every time; any syste","tokens_in":9443,"feed_emoji":"⚛️","tokens_out":9603,"duration_ms":85395,"temperature":0.7,"pith_summary":"The paper demonstrates that temporal coherence spectroscopy—measuring the first-order correlation function g^(1)(τ) of light scattered by atoms—can probe atomic motion inside a purely magnetic quadrupole trap, a configuration where such measurements had not previously been reported. The measured scattered-light spectrum is broader and shifted relative to the prediction based on the independently measured time-of-flight (TOF) temperature of 85 µK. The authors show that this mismatch does not come from the magnetic field: varying the trap gradient leaves the spectrum unchanged. Instead, a simple dynamical model in which each scattered photon accelerates the atom by one recoil (shifting subsequent emitted frequencies) and heats it by T_r/3 per event reproduces the observed frequency shift and linewidth broadening for two probe durations across the full range of saturation parameters. This establishes coherence spectroscopy as a quantitative in-situ probe of velocity distributions in trapped ensembles, complementing destructive TOF imaging and opening the way to time-resolved studies of nonequilibrium dynamics.","feed_headline":"Probe push explains the cold-atom coherence-TOF temperature gap","feed_subtitle":"Modeling radiation-pressure push and heating makes coherence spectroscopy quantitative in magnetic traps.","key_machinery":"The central object is the first-order temporal coherence function g^(1)(τ) of the scattered light, measured through a heterodyne Hanbury Brown–Twiss setup in which a local oscillator derived from the probe is injected into the second port of a fiber beam splitter; the beat-note correlation g^(2)_BN(τ) contains g^(1) as a fringe at the beat frequency f_BN and g^(2) near zero frequency. In the low-saturation single-scattering regime the g^(1) spectrum is Gaussian, with HWHM Δf_T = (k_L/π)√(ln2 k_B T/m)(1 - cosθ), linking linewidth directly to temperature. The load-bearing mechanism is the probe-back-action model: each photon scattered by an atom adds one recoil velocity v_r (shifting the next","core_discovery":"The central discovery is that the apparent disagreement between coherence spectra and TOF-based temperatures for a cloud of 87Rb atoms in a quadrupole magnetic trap originates not from the trap but from the measurement itself. While the probe is on, radiation pressure accelerates the cloud in the probe direction—each of the N_ex scattered photons adds a recoil velocity v_r, so the Nth scattered photon is Doppler-shifted by -N_ex k_L (1 - cosθ) v_r—and each scattering event also heats the atoms by T_r/3, broadening their velocity distribution. Summing the per-event Gaussian spectra over the pulse reproduces the measured shift of the g^(1) center frequency and the broadening of its width as fu","pith_inferences":["A natural inversion the paper does not perform: treating the measured frequency shift itself as a readout of scattered-photon number would let the probe serve as its own calibration for scattering rate, since the shift is predicted to grow linearly with N_ex.","The model's heating rate T_r/3 is assumed rather than derived; a decisive test would be to invert the model across several probe durations and saturation values and check that the recovered initial temperature is constant—if not, one of the two back-action rates is mis-specified.","In the quantum-degenerate regime the single-scattering and dilute assumptions will break down; using g^(2), which is insensitive to radiation-pressure acceleration, could isolate heating and avoid the Doppler-ramp complication.","A red-detuned probe should reverse the sign of the accumulated Doppler shift while keeping broadening unchanged; that asymmetry is a signature a future experiment could check to confirm the mechanism."],"forward_implications":["Coherence spectroscopy now works as a quantitative in-situ thermometer inside a purely magnetic trap, without needing destructive time-of-flight expansion.","Apparent discrepancies between in-situ and TOF temperatures can be fully accounted for by probe-induced acceleration and heating; the magnetic field itself does not perturb the temperature measurement.","The dynamical model provides a calibration rule: for a given saturation parameter and probe duration, the frequency shift and broadening are predictable, so the initial atom temperature can in principle be extracted.","Since the method is minimally destructive and works inside the trap, it enables time-resolved studies of thermalization and nonequilibrium dynamics in confined cold gases, complementing TOF imaging.","Because g^(2) is measured without a local oscillator, it is largely immune to radiation-pressure acceleration and offers a cleaner (though noisier) channel for studying heating alone."],"fun_headline_variants":["Probe's radiation pressure skews cold-atom coherence spectra","Coherence-TOF gap traced to probe-induced recoil and heating","Radiation-pressure recoil explains coherence-TOF temperature mismatch","Probe self-effect: scattered light alters atomic motion in traps","Model accounts for probe push and heating in coherence spectroscopy"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The whole analysis rests on the assumption that the probe's only back-action is one recoil Doppler shift and a fixed T_r/3 heating increment per scattering event, and that the cloud's initial temperature is exactly the TOF value—if either rate is wrong or another broadening mechanism contributes, the agreement between model and data could be fortuitous.","fun_headline_variants_meta":{"raw":{"variants":["Probe's radiation pressure skews cold-atom coherence spectra","Coherence-TOF gap traced to probe-induced recoil and heating","Radiation-pressure recoil explains coherence-TOF temperature mismatch","Probe self-effect: scattered light alters atomic motion in traps","Model accounts for probe push and heating in coherence spectroscopy"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000294,"raw_usage":{"total_tokens":1522,"prompt_tokens":693,"completion_tokens":829,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":437,"completion_tokens_details":{"reasoning_tokens":743}},"tokens_in":437,"tokens_out":829,"duration_ms":6940,"temperature":1.0,"reasoning_tokens":743,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T14:15:20.043770+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the g^(1) spectrum with the probe switched to the red side of resonance: radiation pressure then pushes atoms opposite the probe direction, so the accumulated Doppler shift should change sign with the same magnitude predicted by Eq. (7), while the heating-induced broadening stays the same. If the sign or magnitude of the shift does not follow N_ex k_L (1 - cosθ) v_r, the acceleration mechanism is not the full explanation. Alternatively, invert the model for several probe durations and saturation values and check that the recovered starting temperature is identical every time; any syste","supporting_citations":[],"review_version":1}