{"id":"b295f66f-6fbc-4717-9220-36e4b1c4f71f","arxiv_id":"2501.07971","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Experiments show sub-hertz oscillations of the helium depth in an electron microchannel, attributed to superfluid film flow, which shift the electron Rydberg transition by tens of gigahertz.","lead":"Electrons trapped on superfluid helium in a microscopic channel show slow, roughly 7-second oscillations in their quantum transition frequency, caused by the helium surface rising and falling. The work identifies a new, very low-frequency source of noise for electrons-on-helium qubits, which matter for building scalable quantum computers.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The inferred helium-depth amplitude is internally inconsistent: with the stated δVB≈10 mV, z≈h, VB=168 mV, and VT=−0.4 V, the authors' own relation δVB=δz(VB−VT)/z gives δz≈26 nm, not 180–250 nm; until this is resolved, the surface-oscillation amplitude and its Sommer-Tanner support are uncertain.","rationale":"The reader's verdict (CONDITIONAL) is appropriate. The experiment appears to show a real, reproducible sub-hertz oscillation in the Rydberg-transition signal, and the simultaneous uim/ST correlation in Fig. 4(b) is strong evidence for a common physical origin. The proposed superfluid-film inertia mechanism is plausible and, when the stated geometry is substituted into ω = sqrt(gS/(LA)) without arithmetic error, actually yields f ≈ 0.13 Hz, close to the observed 0.14 Hz — so the mechanism is more consistent than the paper's reported 0.4 Hz suggests. However, the quantitative amplitude inference has a more serious internal inconsistency: the stated δVB ≈ 10 mV and δz ≈ 180–250 nm are mutually incompatible with the authors' own Eq. (2)–(3) relation for the fixed-electron-density case. Since the identification of the signal as helium-depth oscillation rather than, say, electron-density or potential oscillation rests on this amplitude and on the corresponding Sommer-Tanner comparison, this needs correction or clarification before the central claim can be taken at face value. The concrete check is straightforward and can be done from the already-measured data. I therefore do not propose changing the reader's CONDITIONAL verdict, but I flag that the more load-bearing concern is the amplitude conversion, not the frequency estimate as originally emphasized.","tokens_in":8188,"tokens_out":14464,"duration_ms":153608,"concrete_test":"Re-extract the temporal shift δVB directly from the Fig. 2(b) time traces (using the measured slope duim/dVB at the fixed operating point) and insert it into δz = δVB z/(VB − VT) with z ≈ h = 1.5 μm and the recorded VB and VT. If δVB ≈ 10 mV, then δz ≈ 26 nm and the 180–250 nm/20% amplitude claim is unsupported; if δVB ≈ 100 mV is recovered, correct the text. In parallel, compare the expected Sommer-Tanner modulation for each hypothesis: δz = 26 nm gives ~1.7% variation of C (and hence iST), while δz = 250 nm gives ~17%; the observed iST amplitude in Fig. 4(b) distinguishes the two and would also help decide whether the oscillations are in helium depth z or in electron density/potential.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is that helium depth oscillates by roughly 180–250 nm, shifting the Rydberg transition by about 10 mV in equivalent bias. In the paragraph following Eq. (3), the authors state δVB = δz(VB − VT)/z, then write: 'With δVB ≈ 10 mV and z ≈ h, we obtain an estimate δz ≈ 250, 226 and 180 nm.' Plugging in their own values — z ≈ h = 1.5 μm, VB = 168 mV (the operating point used for Figs. 3 and 4), VT = −0.4 V — gives δz = 10 mV × 1500 nm / 568 mV ≈ 26 nm. To obtain 180–250 nm would require δVB ≈ 68–95 mV, not 10 mV. The factor β in Eq. (3) cancels in the fixed-density derivative, so this is not rescued by α/β. This inconsistency is load-bearing because it underpins both the 'about 20%' depth-variation statement and the claimed order-of-magnitude agreement with the Sommer-Tanner signal amplitude. If the true amplitude is ~26 nm, the predicted ST modulation is only ~1.7%, whereas δz = 250 nm predicts ~17%, and the simultaneous uim/ST correlation in Fig. 4(b) could instead reflect electron-density or potential fluctuations triggered by the thermal excitation pulse rather than helium-depth oscillations. Good-faith note: independently recomputing the proposed oscillator frequency from the stated film geometry gives f ≈ 0.13 Hz, not the paper's 0.4 Hz, so the mechanism actually agrees much better with the observed 0.14 Hz if the arithmetic is corrected; this does not weaken the mechanism but indicates a separate numerical slip that should be fixed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports time-domain measurements of the Rydberg transition of electrons trapped on superfluid helium in a microchannel device. After a pulsed heating excitation, the image-charge signal oscillates with a period of about 7 s, and the Sommer-Tanner current shows correlated oscillations with a phase difference consistent with a common origin in variations of the helium depth. The authors attribute these oscillations to inertia of superfluid film flow connecting the microchannels to the bulk helium reservoir, and they estimate peak-to-peak depth variations of 180–250 nm from the voltage-equivalent shift of the Stark spectrum. A simple oscillator model is used to estimate the oscillation frequency, with a stated value of 0.4 Hz compared to the observed 0.14 Hz.","tokens_in":8621,"tokens_out":5853,"duration_ms":53211,"significance":"If the interpretation is correct, the paper identifies a new sub-hertz mechanism of helium surface fluctuations that could affect the coherence of electrons-on-helium qubits in microchannel devices. The simultaneous measurement of two independent signals (image-charge and Sommer-Tanner) with a clear correlation is a notable strength, and the proposed superfluid-film-inertia mechanism is falsifiable and testable by varying device geometry. The paper also extends the study of helium surface noise into the sub-hertz range, which was largely unexplored in prior work. However, the quantitative amplitude extraction contains a serious arithmetical inconsistency, and the frequency estimate contains a numerical error that, once corrected, actually improves the agreement with observation. These issues must be resolved before the quantitative conclusions can be accepted.","major_comments":[{"comment":"The relation δVB = δz(VB − VT)/z is stated, followed by 'With δVB ≈ 10 mV and z ≈ h, we obtain an estimate δz ≈ 250, 226 and 180 nm'. Using the authors' own operating values for the right-most resonance, VB = 168 mV and VT = −0.4 V, with z = h = 1.5 μm, this relation gives δz ≈ 26 nm, not 180–250 nm. This inconsistency is load-bearing because the 'about 20%' depth variation and the order-of-magnitude agreement with the Sommer-Tanner amplitude both rely on the 180–250 nm values. The authors should either recalculate δz with the correct voltages for each trace or explain what values of VB and δVB were used; if the correct amplitude is about 26 nm, the predicted ST modulation is only about 1.7%, and the stated agreement must be revised accordingly.","section":"Section 3, paragraph after Eq. (3)"},{"comment":"The oscillator frequency estimate contains a numerical error. Using ω = sqrt(gS/(LA)) with the stated film thickness of 30 nm covering five adapters of diameter 3.6 mm, L = 20.3 mm, and A = 1.2 mm^2, one obtains f ≈ 0.13 Hz, not 0.4 Hz. Correcting this arithmetic gives excellent agreement with the observed 0.14 Hz, and the 'satisfactory' agreement statement should be updated to reflect this.","section":"Section 3, last paragraph"},{"comment":"The caveat that the parallel-plate capacitor model is 'not adequate' for the low electron density of the right-most plot in Fig. 2(b) directly undermines the reliability of the δz values extracted for that trace, which is exactly the trace used for the quantitative comparisons in Figs. 3 and 4. The authors should quantify how this inadequacy affects the extracted δz and the subsequent ST amplitude comparison.","section":"Section 3, paragraph following Eq. (3)"}],"minor_comments":[{"comment":"The caption contains a long run of corrupted symbols ('/s48 /s50 ...'); the original source should be checked and the caption restored to readable text.","section":"Figure 4 caption"},{"comment":"The journal name 'L. Low Temp. Phys.' should be 'J. Low Temp. Phys.'","section":"Reference 15"},{"comment":"The word 'dependance' should be 'dependence'.","section":"Section 3, first paragraph"},{"comment":"The word 'suprfluid' should be 'superfluid'.","section":"Section 3, last paragraph"},{"comment":"The word 'microchanel' should be 'microchannel'.","section":"Conclusion"},{"comment":"The statement that the liquid level is determined by 'the balance of gravitational force and the surface tension' would be clearer if the meniscus curvature assumption underlying Eq. (1) were explicitly mentioned in the main text.","section":"Section 2, Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The core experimental observation appears solid, and the proposed mechanism is plausible, but the quantitative analysis contains significant arithmetic errors that must be corrected before publication. There is no indication of any ethical concerns; the errors appear to be genuine slips that can be fixed within the scope of the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look: the experiment clearly sees weakly damped, ~7 s oscillations in both image-charge and Sommer-Tanner signals, correlated in phase and amplitude, and triggered by a thermal pulse. Prior noise spectra in this field stopped above 1 Hz, so the sub-hertz regime is genuinely new. The interpretation that the helium depth in the microchannel is oscillating is plausible, and the simultaneous uim/ST correlation is decent supporting evidence.\n\nWhat is weaker is the quantitative packaging. The relation δVB = δz(VB − VT)/z gives about 26 nm if you use the stated operating point (VB = 168 mV, VT = −0.4 V), not the 180–250 nm quoted. The paper never tells you which VB values go into the three numbers, and if the right-most resonance in Fig. 2(a) is the one measured at 168 mV in Fig. 3, then the 180–250 nm amplitudes and the \"about 20%\" depth variation are off by a factor of about seven. That matters because the claimed order-of-magnitude agreement with the Sommer-Tanner signal amplitude depends on δz/z ≈ 17%; with ~26 nm the expected ST modulation is only ~1.7%, and the correlated signal could conceivably be driven by density or potential shifts from the excitation pulse rather than by helium depth. The observation itself still stands; the amplitude claim needs re-examination or clarification.\n\nAnother slip: the oscillator estimate ω = sqrt(gS/(LA)) with the stated 30 nm film and five 3.6 mm adapters gives about 0.13 Hz, not the paper's 0.4 Hz. That corrected number actually agrees much better with the observed 0.14 Hz, so this is a numerical mistake that helps the proposed mechanism rather than hurting it.\n\nAlso missing: error bars on the oscillation frequency and amplitude, a sensitivity analysis of the assumed film thickness and flow path, and raw data. The model is intentionally order-of-magnitude, and the authors explicitly ask for a proper theoretical framework, so I would not punish the paper for not having one.\n\nWho benefits: people building electrons-on-helium qubits or microchannel devices, and anyone doing low-frequency noise spectroscopy of superfluid films. It deserves a serious referee—the experiment is unusual and the sub-hertz channel is under-studied—but that referee should insist on resolving the amplitude arithmetic and adding the sensitivity analysis. I would not desk-reject it.","headline":"The paper reports a real, under-studied sub-hertz helium-level oscillation in a microchannel electron trap, but the amplitude and frequency arithmetic both need fixing before the quantitative story should be taken at face value.","tokens_in":9137,"tokens_out":5470,"would_cite":true,"duration_ms":55419,"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":"Sub-hertz oscillations of the superfluid helium surface—read out through the Rydberg transition of trapped electrons—are traced to the inertia of superfluid film flow inside the microchannel cell.","keywords":["electrons on helium","Rydberg transition","superfluid helium film","microchannel device","sub-hertz oscillations","Stark shift","image-charge detection","transport-current measurement"],"falsifier":"Shorten the film flow path by replacing the five cylindrical connectors with a single connector of half the total length, keeping the fill level fixed, and re-measure the oscillation period: the proposed model predicts $f \\propto 1/\\sqrt{L}$, so the 7 s period should shrink to about 5 s, and an unchanged period would rule out the film-inertia origin.","tokens_in":7989,"feed_emoji":"⚛️","tokens_out":10227,"duration_ms":92398,"temperature":0.7,"pith_summary":"This paper reports that the Rydberg transition of electrons trapped on liquid helium in a microchannel device oscillates in time with a period of about 7 seconds when the cell is given a small thermal pulse. The oscillations correspond to slow, weakly damped changes in the depth of the helium filling the channel, which shift the electron transition frequency through the Stark effect by about 10 mV in equivalent bias. The authors attribute the motion to the inertia of superfluid film flow connecting the microchannels to the bulk helium reservoir, and a simple oscillator estimate gives 0.4 Hz, in reasonable order-of-magnitude agreement with the observed 0.14 Hz. If the attribution is right, this sub-hertz surface motion is a distinct, potentially controllable mechanical noise source for electrons-on-helium qubits, separate from the cryostat vibrations studied before.","feed_headline":"Trapped electrons clock a 7-second helium surface slosh","feed_subtitle":"Depth swings of ~200 nm shift the Rydberg transition by ~10 mV: a slow noise source for electron-on-helium qubits.","key_machinery":"Two models carry the argument. First, the parallel-plate capacitor description of the charged helium surface in the channel, Eqs. (2)–(3), relates the perpendicular electric field on the electrons to the helium depth $z$ and gives the linear conversion $\\delta V_B = \\delta z (V_B - V_T)/z$ between a depth change and the equivalent bias shift of the Rydberg resonance; this turns the measured Stark shifts into physical depth amplitudes of 180–250 nm. Second, the superfluid film-inertia oscillator, $\\omega = \\sqrt{gS/(LA)}$, with $S$ the film cross-section, $A$ the total channel area, and $L$ the connector length, supplies the predicted sub-hertz frequency that the authors compare with the observed 7 s period.","core_discovery":"The central claim is that the temporal dynamics of the Rydberg transition of electrons in a microchannel device reveal a weakly damped oscillation of the liquid helium depth with a period of about 7 s (frequency about 0.14 Hz), and that this oscillation originates from the inertia of superfluid film flow inside the experimental cell. At the onset of the oscillations the helium depth varies by roughly 180–250 nm, shifting the Rydberg resonance by about 10 mV in equivalent bias voltage. The image-charge signal and the transport-current signal oscillate together with the expected $\\pi$ phase difference, showing that both readouts see the same helium-depth motion. The proposed mechanism, modeled as an inertia-dominated oscillator $\\omega = \\sqrt{gS/(LA)}$ with a 30 nm superfluid film covering five cylindrical connectors, yields about 0.4 Hz, which the authors call satisfactory agreement with the measured 0.14 Hz for such a simplified estimate.","pith_inferences":["A testable extension of the paper's mechanism is that the oscillation frequency should scale as the square root of the film cross-section divided by the connector length, so a cell with a shorter or wider film path should show a measurably different period while retaining the same qualitative damped response.","In our reading, the Rydberg-transition readout may be a more sensitive sub-hertz helium-surface monitor than resonator-frequency jitter, since it resolves depth changes of order 200 nm at frequencies below 1 Hz that earlier vibration studies did not probe.","The observed onset behavior—oscillations appearing immediately or half a period earlier depending on whether the excitation pulse heats or cools the cell—suggests the sign of the initial helium-depth displacement matters; a quantitative model of the thermal pulse's effect on film thickness could predict the phase and amplitude of the first oscillation cycle.","If the film is indeed the conduit, the damping rate should depend on the normal-fluid viscosity, so measuring the quality factor as a function of temperature could separate inertial film flow from thermal counterflow as the restoring mechanism."],"forward_implications":["Electrons-on-helium qubits operated in microchannel devices will be subject to slow helium-depth motion in the sub-hertz range, so Rydberg transition frequencies can drift by tens of millivolts in equivalent bias on a timescale of seconds.","Because the oscillations are triggered by small thermal pulses and persist for minutes to thousands of seconds, heat from filament charging or millimetre-wave excitation can leave a long-lived mechanical memory in the helium surface.","The $\\pi$ phase difference between the image-charge and transport-current oscillations indicates that the effect is a genuine change in helium depth rather than an electronic artifact of the detection circuit.","The frequency estimate depends on the geometry of the film path, so changing the connector length, the channel area, or the bulk helium level should shift the period in a predictable way and can serve as a direct test of the proposed origin.","Reducing thermal excitation of the cell or decoupling the microchannel liquid from the film-connected reservoir should suppress or eliminate this particular surface-fluctuation source."],"supporting_citations":[{"why":"Provides the capillary rise expression for the helium level in the channel and the earlier observation of cryostat-vibration-induced surface jitter that this work extends into the sub-hertz range.","marker":"[12]"},{"why":"Supplies the higher-frequency noise measurements showing that mechanical vibrations produce only about 1 nm surface displacement, the baseline the paper argues cannot explain the observed 180–250 nm oscillations.","marker":"[13]"},{"why":"Describes the resonant image-charge amplifier whose lock-in detection of the Rydberg transition is the primary measurement used in the time-trace experiments.","marker":"[20]"},{"why":"Gives the parallel-plate capacitor model equations for the perpendicular electric field and electron density that convert the measured Stark shifts into helium-depth variations.","marker":"[22]"},{"why":"Supplies the transport-current readout whose simultaneous measurement with the image-charge signal demonstrates the common origin of the oscillations.","marker":"[23]"},{"why":"One of the classical sources for the inertia-dominated oscillator frequency formula $\\omega = \\sqrt{gS/(LA)}$ applied to superfluid flow through narrow connections.","marker":"[25]"},{"why":"Provides the analytical treatment of slow weakly damped meniscus oscillations in a container connected to a helium bath by a fine slit or capillary.","marker":"[26]"},{"why":"Extends the oscillator model to superfluid film flow, the basis for estimating the 0.4 Hz frequency in the present cell geometry.","marker":"[27]"}],"fun_headline_variants":["Slow helium slosh shakes electron qubits every 7 seconds","7-second helium slosh shifts electron Rydberg states","Helium surface waves drive slow noise in electron qubits","Trapped electrons feel 7-second helium depth slosh","Weakly damped helium slosh disturbs Rydberg transition"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The frequency comparison assumes that the superfluid film connecting the microchannels to the bulk helium is a uniform roughly 30 nm thick layer coating all five cylindrical connectors, so that the oscillator formula $\\omega = \\sqrt{gS/(LA)}$ correctly describes the restoring flow; if the real film thickness, wetted area, or flow path differs substantially, the predicted 0.4 Hz would not match the observed 0.14 Hz even though the oscillation itself is well documented.","fun_headline_variants_meta":{"raw":{"variants":["Slow helium slosh shakes electron qubits every 7 seconds","7-second helium slosh shifts electron Rydberg states","Helium surface waves drive slow noise in electron qubits","Trapped electrons feel 7-second helium depth slosh","Weakly damped helium slosh disturbs Rydberg transition"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000163,"raw_usage":{"total_tokens":1216,"prompt_tokens":889,"completion_tokens":327,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":505,"completion_tokens_details":{"reasoning_tokens":243}},"tokens_in":505,"tokens_out":327,"duration_ms":3370,"temperature":1.0,"reasoning_tokens":243,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:29:09.794295+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Shorten the film flow path by replacing the five cylindrical connectors with a single connector of half the total length, keeping the fill level fixed, and re-measure the oscillation period: the proposed model predicts $f \\propto 1/\\sqrt{L}$, so the 7 s period should shrink to about 5 s, and an unchanged period would rule out the film-inertia origin.","supporting_citations":[{"cited_title":"Koolstra, G.Yang, and D","cited_arxiv_id":null,"evidence_quote":"Provides the capillary rise expression for the helium level in the channel and the earlier observation of cryostat-vibration-induced surface jitter that this work extends into the sub-hertz range."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the higher-frequency noise measurements showing that mechanical vibrations produce only about 1 nm surface displacement, the baseline the paper argues cannot explain the observed 180–250 nm oscillations."},{"cited_title":"Belianchikov, J","cited_arxiv_id":null,"evidence_quote":"Describes the resonant image-charge amplifier whose lock-in detection of the Rydberg transition is the primary measurement used in the time-trace experiments."},{"cited_title":"Zou and D","cited_arxiv_id":null,"evidence_quote":"Gives the parallel-plate capacitor model equations for the perpendicular electric field and electron density that convert the measured Stark shifts into helium-depth variations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the transport-current readout whose simultaneous measurement with the image-charge signal demonstrates the common origin of the oscillations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"One of the classical sources for the inertia-dominated oscillator frequency formula $\\omega = \\sqrt{gS/(LA)}$ applied to superfluid flow through narrow connections."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the analytical treatment of slow weakly damped meniscus oscillations in a container connected to a helium bath by a fine slit or capillary."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Extends the oscillator model to superfluid film flow, the basis for estimating the 0.4 Hz frequency in the present cell geometry."}],"review_version":1}