{"id":"c39cdc7f-9c11-4992-981c-132887742efd","arxiv_id":"2504.15409","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The phase shift observed in the experiment is a magnetic recoil phase, with all uniform gravitational contributions cancelling, so the experiment does not test the equivalence principle.","lead":"A commentary team reanalyzes a recent atom-interferometer experiment claimed to test the quantum equivalence principle. They show the measured phase is a magnetic recoil effect, not a gravitational one, and that it does not depend on a uniform gravitational field.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim treats the magnetic gradient as an independent parameter, but the paper's own assumption that the gradient holds the reference wave packet stationary against gravity requires Fmag = m_g g0; imposing this in Eq. (6) makes the phase depend on g0 and m_g.","rationale":"The reader correctly identified the assumption of perfect closure and negligible inertial phase shifts as the weakest point, but the concern is sharper than a mere residual-term estimate. The paper's own description of the experiment imposes a constraint on Fmag that reintroduces the gravitational field and the gravitational mass into the central phase formula. If Fmag = m_g g0, Eq. (6) becomes a measurement of m_g^2/m_i, so the conclusion that the phase does not depend on g0 and therefore does not test the equivalence principle is unsupported. The paper could perhaps be revised to a narrower claim about an idealized interferometer with independently calibrated Fmag, but as written the central conclusion does not follow. This moves the verdict from conditional to reject, because the identified issue directly undermines the claim about what the experiment observed.","tokens_in":2887,"tokens_out":22240,"duration_ms":209866,"concrete_test":"Re-derive Eq. (6) under the two stated conditions in the paper: (i) the chip is at rest in the lab and (ii) the reference wave packet remains at constant distance from the chip. These imply Fmag = m_g g0. Substitute this into Eq. (6) to obtain phi = -(m_g^2/m_i) g0^2 T^3/(3hbar). Then check whether this expression, rather than Eq. (6), is the phase whose T^3 scaling is compared in Ref. [1]. If the expression contains g0 and m_g, the claim that the observed phase is independent of g0 and of gravitational mass fails for the levitated configuration. As a complementary numerical check, estimate a_res from the experimental parameters in Ref. [1] and compare |a_res Fmag T^3/hbar| with |(Fmag^2/m_i) T^3/hbar|; if the ratio is not small, the neglected inertial term is itself the signal.","verdict_should_be":"REJECT","load_bearing_attack":"The conclusion depends on treating Fmag = µ∇B as an externally set parameter in Eq. (6). However, the Introduction states that the gradient 'keeps the reference wave packet at an approximately constant distance from the chip.' For a chip at rest in the lab, the reference packet remains stationary relative to the chip only if the magnetic force cancels the gravitational force, i.e., Fmag = m_g g0, not Fmag = m_i g0. Inserting this condition into Eq. (6) gives phi = -(m_g^2/m_i) g0^2 T^3/(3hbar), which depends on the uniform gravitational field and on the gravitational mass. Equivalently, a measurement of phi at known g0 and T with independently known m_i is a measurement of m_g^2/m_i, i.e., an equivalence-principle-relevant quantity. The paper's claim that the phase 'depends on the magnetic gradient and the inertial mass only, not on the uniform gravitational field g0' is therefore not the phase of the described experimental configuration; it is the phase in a hypothetical setup where Fmag is not constrained by levitation. The assumption of no residual motion between the chip and the reference packet is not an innocuous simplification but the step that removes the gravitational signal. The paper does not provide experimental numbers showing that the residual phase a_res Fmag T^3/hbar is small; it simply asserts neglect.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript is a comment on arXiv:2502.14535, which reports a matter-wave interferometer phase and interprets it as a test of the quantum equivalence principle. The authors calculate the interferometer phase in a uniform gravitational field using the action for a ballistic arm and a magnetically levitated reference arm. They obtain phi = -(mu grad B)^2 T^3 / (3 m_i hbar) and argue that this is a magnetic recoil phase shift that depends only on the magnetic gradient and the inertial mass, not on the uniform gravitational field. From this they conclude that the experiment does not test the equivalence principle. The algebraic derivation is checked and is internally consistent under the stated assumptions, but the central claim about the observed phase depends on how the magnetic gradient is related to the gravitational field.","tokens_in":3226,"tokens_out":10409,"duration_ms":93049,"significance":"If the central claim were correct, it would reclassify a recent experimental result as a magnetic recoil measurement rather than a gravitational observation, with implications for how quantum equivalence principle tests are interpreted. The manuscript provides a self-contained action-based derivation and does not rely on fitting parameters. However, the significance is severely undercut by a missing constraint: the experimental configuration requires the magnetic gradient to levitate the reference wave packet against gravity, which ties grad B to g0. Imposing that constraint makes the phase depend on g0 and m_g, so the conclusion that the observed phase is independent of the gravitational field is not supported for the actual experiment. The correct algebra alone does not settle the interpretational question.","major_comments":[{"comment":"The central claim that the phase observed in the experiment does not depend on the uniform gravitational field is not valid for the described experimental configuration. The Introduction states that the magnetic gradient 'keeps the reference wave packet at an approximately constant distance from the chip.' For a chip at rest in the lab frame, this condition requires the magnetic force to balance the gravitational force, i.e., F_mag = m_g g0. Substituting this constraint into Eq. (6) gives phi = -(m_g^2 / m_i) g0^2 T^3 / (3 hbar), which depends on the gravitational field and on the gravitational mass. The manuscript treats F_mag as an independent parameter, but in the experiment it is fixed by gravity. The statement that the phase 'depends on the magnetic gradient and the inertial mass only, not on the uniform gravitational field g0' is therefore only true in a hypothetical setup where the magnetic gradient is not adjusted to maintain levitation, which is not the experiment under discussion.","section":"Introduction and Eq. (6)"},{"comment":"The manuscript acknowledges that residual motion of the reference wave packet with respect to the chip would produce inertial phase shifts proportional to a_res F_mag T^3 / hbar, but it provides no estimate or bound for a_res. The neglect of these shifts is essential to the conclusion that the observed phase is exactly the magnetic recoil phase of Eq. (6). Without a quantitative assessment using the published experimental parameters of Ref. [1], the claim that the experiment measures no gravitational effect is not justified. The paper should either bound a_res from the experimental parameters or explicitly restrict the conclusion to an idealized limit of perfect levitation.","section":"Equivalence principle tests"}],"minor_comments":[{"comment":"The quantity F_mag is used before being defined; please define F_mag = mu grad B explicitly when it first appears.","section":"Eq. (4)"},{"comment":"The figure caption mentions arm trajectories for different values of g0, but the figure is not included in the text; please describe the trajectories in the text or include the figure.","section":"Figure 1"},{"comment":"The term 'inertial phase shifts' is used without a definition in the introduction; please clarify that these are phase shifts arising from motion of the wave packets relative to the chip or reference frame.","section":"Introduction"},{"comment":"The final paragraph on 'problematic formulations of the equivalence principle' is more philosophical than technical and could be shortened or removed without affecting the argument.","section":"Concluding remarks"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well written and the algebra is correct, but the main conclusion appears to be overstated because the levitation condition is not incorporated. The paper relies heavily on the authors' earlier work (Ref. [3]) and the novelty relative to that work may be limited if the central claim cannot be sustained. The editor may wish to consider whether the comment, after accounting for the levitation constraint, still offers a distinct and correct point about the experiment's interpretation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Peter and Chris have written a compact comment claiming that the phase in Dobkowski et al. is a magnetic recoil effect independent of the uniform gravitational field. The algebra in Eqs. (1)-(6) is correct given their assumptions, and the frame-independent gauge-phase demonstration is elegant. But the central claim does not survive contact with the experiment's actual operating point.\n\nThe problem is the levitation condition. The introduction states that the gradient keeps the reference wave packet at an approximately constant distance from the chip. In the lab frame, that means the magnetic force must cancel gravity: Fmag = mg g0 (up to a small residual a_res). The paper never imposes this constraint. Instead it treats Fmag as a free parameter in Eq. (6), concluding that phi depends only on the gradient and inertial mass. Once you enforce Fmag = mg g0, Eq. (6) becomes phi = -(mg^2/mi) g0^2 T^3/(3 hbar), which depends on the uniform gravitational field and on the gravitational mass. The phase is still a recoil phase in origin, but it is also a gravitational phase because the gradient is the support force against gravity. The paper's own assumption removes the g0-dependence, but only by not setting Fmag to the value the experiment requires.\n\nThe 'perfect closure' and 'ignore inertial phase shifts' assumptions are also asserted without any bound on a_res. That would be a minor issue if the levitation constraint were handled properly, but here it is load-bearing: the neglect of inertial shifts is not an innocuous simplification, it is the step that lets Fmag stay arbitrary. The paper does not provide experimental numbers to show that a_res Fmag T^3 / hbar is small.\n\nOn the positive side, the calculation is clean and the point that magnetic acceleration depends on particle properties while gravitational acceleration does not is worth stating. The self-citations to Refs. [3,4] are appropriate background; the derivation is self-contained. The paper is clearly written and would make a good discussion piece.\n\nBut as a comment on the original experiment, it likely fails. A chip at rest in the lab is not a freely falling frame, so a phase that depends on g0 is not a violation of the equivalence principle; it is exactly what you expect when one arm is supported against gravity. The original paper's interpretation may have its own problems, but this comment does not identify them.\n\nI would send this to peer review because it is a substantive comment and the tension between the levitation statement and Eq. (6) deserves a public referee exchange. But my honest expectation is that the main claim will not survive. I would not cite it in my own work.","headline":"The calculation is clean, but the levitation condition undermines the central claim: once the magnetic force is set to hold the reference arm against gravity, the phase depends on g0 after all.","tokens_in":3683,"tokens_out":5581,"would_cite":false,"duration_ms":52534,"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":"This comment re-derives the phase of the reported matter-wave interferometer and shows it is a magnetic recoil phase, independent of the uniform gravitational field, so the experiment cannot test the equivalence principle.","keywords":["atom interferometry","magnetic recoil phase shift","equivalence principle","matter waves","inertial mass","uniform gravitational field","magnetic gradient","gauge phase"],"falsifier":"Measure the same interferometer's phase after reorienting the apparatus by 180 degrees with respect to local gravity but keeping the magnetic gradient fixed: a phase change would contradict the claim that $\\phi$ is independent of $g_0$. Alternatively, compare two species with equal magnetic moments but different inertial masses; Eq. (6) predicts the phase ratio $m_{i,2}/m_{i,1}$, whereas a gravitational interpretation would predict sensitivity to the gravitational mass ratio.","tokens_in":2701,"feed_emoji":"⚛️","tokens_out":7908,"duration_ms":63774,"temperature":0.7,"pith_summary":"This comment re-analyzes the atom interferometer experiment in arXiv:2502.14535, which claimed to observe the quantum equivalence principle by measuring a phase proportional to the local gravitational acceleration. The authors show that, once the two interferometer arms are treated with their full actions and the magnetic force is kept distinct from gravity, all gravitational terms cancel between the arms. The resulting phase is a magnetic recoil phase, proportional to the square of the applied magnetic gradient divided by the inertial mass. Because the phase is independent of the uniform gravitational field, the measurement cannot serve as a local test of the equivalence principle.","feed_headline":"Phase is a magnetic recoil, not a gravity signal","feed_subtitle":"Uniform gravity cancels between the arms, leaving only a magnetic recoil phase, so no equivalence-principle test is made.","key_machinery":"The central object is the phase-shift identity $\\phi = -(\\mu \\nabla B)^2 T^3/(3 m_i \\hbar)$, obtained by evaluating the actions of the two arms and the gauge phase between the Newtonian and Einsteinian coordinate frames. The magnetic force $F_{\\rm mag} = \\mu \\nabla B$ acts only on the reference wave packet, while gravity acts identically on both arms; the recoil-phase form, momentum-difference squared divided by inertial mass, is what makes the phase independent of the uniform gravitational field. A similar calculation structure underlies earlier work showing that clocks and symmetric matter-wave interferometers do not observe uniform gravitational fields.","core_discovery":"Starting from the same geometry as the target experiment—a magnetically insensitive ballistic wave packet kicked upward, and a magnetically sensitive reference wave packet held at constant distance from a current-carrying chip—the authors compute the interferometer phase in a uniform gravitational field $g_0$, without assuming equality of inertial mass $m_i$ and gravitational mass $m_g$. The phase shift comes out to $\\phi = -(\\mu \\nabla B)^2 T^3 / (3 m_i \\hbar)$, with $\\mu \\nabla B$ the magnetic force on the reference arm and $2T$ the interferometer time. Every term involving $g_0$ cancels between the two arms, even though the individual actions and the gauge phase relating Newtonian and Einsteinian coordinates do depend on $g_0$. The authors conclude that the observed phase is a magnetic recoil phase, generated by the differential magnetic force between arms, and that the experiment therefore does not compare gravitational and inertial mass.","pith_inferences":["The same cancellation argument should apply to any proposed equivalence-principle test in which a uniform force acts identically on both arms: only differential forces produce a phase, so the experiment must be designed with a deliberately differential interaction.","A direct extension would be to measure the phase for two atomic isotopes with similar magnetic moments but different inertial masses; Eq. (6) predicts a phase ratio set by the inverse mass ratio, which would cleanly distinguish recoil from gravitational phase.","If the uniform-field independence holds, rotating the entire apparatus by 180 degrees relative to local gravity should leave the phase unchanged apart from field-direction-dependent magnetic terms, providing a simple laboratory check.","The recoil-phase nature suggests the interferometer could serve as a precision magnetometer or as a measurement of inertial mass for the trapped species, since the phase scales quadratically with the gradient and inversely with mass."],"forward_implications":["The measured phase can be used to characterize the magnetic gradient rather than the gravitational acceleration; it carries no gravitational-mass information.","The result does not falsify the equivalence principle; instead it confirms that uniform gravitational fields produce no local interferometric phase when both arms experience the same acceleration.","Any residual acceleration $a_{\\rm res}$ between the chip and the reference wave packet adds an inertial phase proportional to $a_{\\rm res} F_{\\rm mag} T^3/\\hbar$, which is how the apparatus could act as a test between the atoms and the chip.","Coordinate choices and gauge transformations may change the appearance of intermediate terms, but the physical phase is invariant; comparing a coordinate-dependent calculation to data cannot test the equivalence principle."],"supporting_citations":[{"why":"The target experiment whose measured phase is reinterpreted as a magnetic recoil phase.","marker":"[1]"},{"why":"Supplies the standard definition that uniform gravitational fields are locally unobservable.","marker":"[2]"},{"why":"Earlier analysis of clocks and symmetric interferometers that the present calculation extends.","marker":"[3]"},{"why":"Provides the notion of high-order inertial phase shifts and the recoil-phase scaling used here.","marker":"[4]"}],"fun_headline_variants":["Magnetic recoil, not gravity, drives phase","Uniform gravity cancels, magnetic recoil remains","No equivalence test: phase is recoil","Interferometer phase: magnetic recoil, not gravity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes the interferometer is exactly closed and that residual motion between the chip and the reference wave packet is negligible; if those leftover motions are not small, extra inertial phase shifts appear and the measured phase could include contributions beyond the magnetic recoil term.","fun_headline_variants_meta":{"raw":{"variants":["Magnetic recoil, not gravity, drives phase","Uniform gravity cancels, magnetic recoil remains","No equivalence test: phase is recoil","Interferometer phase: magnetic recoil, not gravity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000166,"raw_usage":{"total_tokens":1208,"prompt_tokens":852,"completion_tokens":356,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":468,"completion_tokens_details":{"reasoning_tokens":295}},"tokens_in":468,"tokens_out":356,"duration_ms":3308,"temperature":1.0,"reasoning_tokens":295,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:26:59.289653+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same interferometer's phase after reorienting the apparatus by 180 degrees with respect to local gravity but keeping the magnetic gradient fixed: a phase change would contradict the claim that $\\phi$ is independent of $g_0$. Alternatively, compare two species with equal magnetic moments but different inertial masses; Eq. (6) predicts the phase ratio $m_{i,2}/m_{i,1}$, whereas a gravitational interpretation would predict sensitivity to the gravitational mass ratio.","supporting_citations":[{"cited_title":"Nonequivalence of equivalence principles","cited_arxiv_id":null,"evidence_quote":"Supplies the standard definition that uniform gravitational fields are locally unobservable."},{"cited_title":"Matter waves and clocks do not observe uniform gravitational fields","cited_arxiv_id":null,"evidence_quote":"Earlier analysis of clocks and symmetric interferometers that the present calculation extends."},{"cited_title":"High-order inertial phase shifts for time-domain atom interferometers","cited_arxiv_id":null,"evidence_quote":"Provides the notion of high-order inertial phase shifts and the recoil-phase scaling used here."}],"review_version":1}