{"id":"38cdbf2d-065a-46ce-a834-6f421fa493c4","arxiv_id":"1908.07296","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"In unidirectionally coupled optomechanical cavities, quantum phase synchronization peaks away from resonance (synchronization blockade), while Gaussian quantum discord fails to reproduce that blockade.","lead":"Two optically coupled optomechanical oscillators are shown to synchronize differently depending on whether the optical link is bidirectional or unidirectional. The unidirectional case produces a quantum synchronization blockade, and Gaussian quantum discord does not track that blockade, which matters for experiments that use discord as a witness of synchronization.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (2c) drops the vacuum noise of the loss port, so the unidirectional input is non-canonical; the eta-scan that defines the synchronization blockade in Fig. 4(a) may not be physical.","rationale":"The reader's weakest assumption concerned the zero-delay cascaded model. I partially agree: Eq. (2c) is indeed the fragile step in the unidirectional claim. However, I identify a more concrete and more load-bearing defect in the same equation: the total omission of the vacuum noise from the loss port. This is not a numerical or convergence issue but a violation of the canonical commutation relation for the input field whenever eta < 1. Because the blockade is characterized by scanning eta, the missing vacuum term directly changes the noise budget of the right oscillator and can alter the location and even the existence of the finite-detuning peak. This concern is testable by adding the missing term and regenerating Fig. 4. The reader's verdict of CONDITIONAL remains appropriate: the central claim is plausible but not established until the cascaded input is made canonical and the tau > 0 case is checked. I therefore leave the verdict unchanged rather than escalating to rejection, because the error is concrete, fixable, and its impact can be settled by a well-defined numerical experiment.","tokens_in":8595,"tokens_out":16358,"duration_ms":178702,"concrete_test":"Recompute the <S_p>(delta/omega_mL, eta) map of Fig. 4(a) using the corrected cascaded input a_R^in(t) = sqrt(eta)(a_in^L - sqrt(2 kappa_L) a_L) + sqrt(1 - eta) v(t), with v an independent canonical vacuum operator, and verify that the zero-delay, no-loss case is recovered at eta = 1. If the delta = 0 minimum and finite-detuning maximum survive for all eta in [0,1], the blockade is robust; if they move, broaden, or disappear as eta approaches 0, the reported blockade is an artifact of the missing vacuum port. As a second branch, repeat with tau = 1/omega_mL as in the Fig. 2 caption to test the delay assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central unidirectional claim is read off from Fig. 4(a), a map of <S_p> versus (delta, eta). That map is generated from Eq. (2c), where the right cavity input is written as sqrt(2 kappa_R) sqrt(eta) (a_in^L - sqrt(2 kappa_L) a_L). Input-output theory for a lossy channel requires a_R^in(t) = sqrt(eta) a_out^L(t - tau) + sqrt(1 - eta) v(t), with v a vacuum field entering at the unused port. At tau = 0 this becomes sqrt(eta)(a_in^L - sqrt(2 kappa_L) a_L) + sqrt(1 - eta) v. The v term is absent, so for eta < 1 the effective input field has commutator eta delta(t - t') instead of the canonical delta(t - t'); at eta = 0 the right cavity has no optical input noise at all, which is unphysical for a cavity open to vacuum. Since eta is the control parameter in the blockade plot, the minimum at delta = 0 and the finite-delta maximum could be artifacts of this non-canonical noise. The zero-delay idealization is a second, related simplification; the paper never analyzes tau > 0 despite Fig. 2's caption listing tau = 1/omega_mL. The missing vacuum term is the more basic formal issue: without it the quantum noise budget of the right oscillator is incomplete, and both <S_p> and the discord comparison inherit the error.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates quantum phase synchronization and Gaussian quantum discord in two optically coupled optomechanical oscillators, comparing bidirectional and unidirectional coupling topologies. For the bidirectional configuration, the time-averaged synchronization measure and quantum discord both exhibit Arnold-tongue-like patterns centered at resonance. For the unidirectional configuration, the synchronization measure is reported to show a blockade-like behavior (suppressed at resonance, peaking at finite detuning), while quantum discord does not reproduce this anomaly. The analysis uses standard linearized quantum Langevin equations, a covariance-matrix approach, and the Mari et al. phase-synchronization measure, with time-averaged quantities computed after solving the classical limit-cycle dynamics.","tokens_in":8912,"tokens_out":8772,"duration_ms":84195,"significance":"If the reported synchronization blockade in the unidirectional configuration is physically correct, the paper offers a clear example where a phase-sensitive synchronization measure and a generic correlation measure behave differently across coupling topologies, which is of interest to the quantum synchronization and quantum information communities. The methods are standard: linearized Langevin dynamics, Gaussian covariance matrices, and a well-established synchronization metric. The paper also checks robustness at a thermal phonon occupation of n_th = 10. The central novelty, however, depends on the correct treatment of the unidirectional input-output model, which is presently incomplete.","major_comments":[{"comment":"The printed definition S_p(t) = 1/2⟨δp_-(t)⟩ - 1 is undefined because the phase-difference fluctuation operator δp_- has zero mean. The standard Mari et al. definition uses the variance, S_p = 1/(2⟨δp_-^2⟩) - 1 (or an equivalent form). This is the central figure of merit of the paper; please correct the equation.","section":"Section III, Eq. (5)"},{"comment":"The unidirectional input relation a_R^in(t) = √η a_out^L(t) with τ=0 omits the vacuum noise from the unused loss port. The physical input is a_R^in(t) = √η a_out^L(t) + √(1-η) v(t), where v is vacuum. Without this term, the effective input field has commutator η δ(t-t') instead of the canonical δ(t-t'), so for η<1 the right-cavity noise budget is incomplete. Since η is the control parameter in Fig. 4(a), the reported synchronization blockade and its η-dependence may be artifacts of the missing vacuum noise. Please redo the calculation with the full input-output relation and confirm that the blockade persists, including at η=1. Also clarify the inconsistency between the stated τ=0 assumption and the Fig. 2 caption listing τ=1/ω_mL, and assess the robustness of the results to finite τ.","section":"Section II, Eq. (2c), and Section IV, Fig. 4(a)"}],"minor_comments":[{"comment":"The expression C_{i,l}(t) = ⟨R_i(t)R_l(t)† + R_l(t)†⟩ is missing the second R_i; it should be the symmetrized form ⟨R_i R_l† + R_l† R_i⟩.","section":"Section III, correlation matrix definition"},{"comment":"The text states that τ is set to 0 for the rest of the paper, but the Fig. 2 caption lists τ = 1/ω_mL. This internal inconsistency should be resolved.","section":"Section II, text after Eq. (2) and Fig. 2 caption"},{"comment":"The captions for Figs. 3 and 4 do not list the parameter ranges used (e.g., ranges of λ/κ, δ/ω_mL, and η). For reproducibility, please provide these values or state them in the text.","section":"Section IV, Fig. 3 and Fig. 4 captions"},{"comment":"The phrase 'The blockade becomes maximum for detuned oscillators' is confusing: if 'blockade' denotes suppression of synchronization, the maximum blockade occurs at resonance. Please rephrase to state that synchronization is maximized at finite detuning.","section":"Section V, concluding paragraph"},{"comment":"The sentence 'Optically coupled optomechanical oscillators has turned out' uses a singular verb with a plural subject; it should be 'have turned out'.","section":"Introduction, first sentence"},{"comment":"For clarity, explicitly define a_out^L(t) in the input-output relation, since it appears implicitly in the unidirectional coupling term.","section":"Section II, Eq. (2c)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents a potentially interesting topology-dependent synchronization effect, but the central unidirectional claim rests on a non-canonical input-output model that omits the vacuum noise of the loss port. This is a fixable issue in principle, but the authors must rerun the simulations with the correct noise term and verify that the reported synchronization blockade survives. The Eq. (5) typo and the τ inconsistency in Fig. 2 suggest the manuscript would benefit from careful proofreading. If the blockade disappears after the correction, the paper's main novelty would be undermined."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new piece here is the comparison of bidirectional versus unidirectional optical coupling for quantum phase synchronization, plus the observation that Gaussian quantum discord does not track the synchronization blockade. That discord mismatch is a legitimate result if the underlying model holds. The paper also does some things well: it uses the standard linearized Langevin and covariance-matrix method, it checks thermal robustness at n_th=10, and it cites the earlier blockade literature (including the optomechanical blockade of Ref. [40]) rather than ignoring it.\n\nThe soft spots are serious, though, and one is load-bearing. Eq. (2c), the unidirectional input, omits the vacuum field entering at the loss port. For a lossy channel the right-cavity input should be sqrt(eta) a_out^L + sqrt(1-eta) v, and at tau=0 that becomes sqrt(eta)(a_in^L - sqrt(2 kappa_L) a_L) + sqrt(1-eta) v. The v term is absent, so for eta<1 the effective input commutator is eta delta(t-t') rather than delta(t-t'), and at eta=0 the right cavity has no optical input noise at all—unphysical for a cavity open to vacuum. Since eta is the control parameter in Fig. 4(a), the suppression at resonance and the finite-detuning peak could be artifacts of this non-canonical noise. This is not a minor typo; it changes the physics.\n\nThere are smaller but real problems. Eq. (5) as printed is undefined: S_p = 1/(2<δp_->) - 1 is missing the variance and reads as the inverse of the operator rather than the inverse of its second moment. The model sets inter-cavity delay tau=0, but the Fig. 2 caption lists tau=1/omega_mL, and no analysis of tau>0 is given. There is no code or convergence data. And while the abstract calls the blockade \"novel,\" the full text correctly notes that blockade has already been reported in optomechanical systems (Ref. [40]); what is new is only the unidirectional setup and the discord comparison.\n\nMy overall take: the central bidirectional result (Arnold tongue) is on solid ground and is not controversial. The unidirectional blockade claim is not trustworthy until the missing vacuum term is included and the eta-scan is redone. If the blockade survives, this is a useful contribution; if not, the paper becomes mostly a confirmation of known behavior. Either way, it deserves a serious referee, but the referee should ask for the fix and a rerun before publication.","headline":"The unidirectional input-output model drops the vacuum noise of the loss port, so the synchronization blockade in Fig. 4(a) may be an artifact; the rest is a reasonable but incremental numerical study.","tokens_in":9465,"tokens_out":1666,"would_cite":false,"duration_ms":18438,"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":"At zero propagation delay, unidirectionally coupled optomechanical oscillators synchronize best when they are detuned, not identical, while Gaussian quantum discord fails to flag this quantum synchronization blockade.","keywords":["quantum synchronization","optomechanical oscillators","quantum synchronization blockade","Gaussian quantum discord","Arnold tongue","cascaded quantum systems","continuous-variable quantum information"],"falsifier":"Recompute or measure the time-averaged synchronization $\\langle S_p\\rangle$ for the unidirectional cascade with a finite propagation delay $\\tau$ of order one mechanical period ($1/\\omega_{mL}$). If the minimum at $\\delta=0$ and the maximum at finite detuning disappear or invert, the claimed quantum synchronization blockade does not hold.","tokens_in":8350,"feed_emoji":"🔄","tokens_out":6959,"duration_ms":63461,"temperature":0.7,"pith_summary":"The paper shows that in two optically coupled optomechanical oscillators, the onset of quantum phase synchronization always accompanies a finite Gaussian quantum discord, but the shape of the synchronization region depends on the coupling topology. For bidirectional photon exchange, both synchronization and discord form a resonance-centered Arnold-tongue pattern in the detuning–coupling plane. For unidirectional (cascaded) exchange, the maximum synchronization moves away from resonance to finite detunings governed by the transmission loss, a quantum synchronization blockade, whereas the discord keeps its tongue shape. The paper concludes from this mismatch that Gaussian quantum discord is not a conclusive order parameter for quantum phase synchronization, despite being present whenever synchronization occurs.","feed_headline":"One-way coupling makes detuned oscillators sync best","feed_subtitle":"In cascaded optomechanical oscillators, quantum phase synchronization peaks off resonance; quantum discord misses this blockade.","key_machinery":"The load-bearing machinery is the zero-delay cascaded input–output boundary condition $a_{\\rm in}^R(t)=\\sqrt{\\eta}\\,a_{\\rm out}^L(t)$, which turns the right cavity into a driven replica of the left cavity's output and breaks the exchange symmetry that makes identical oscillators resonant partners. On top of this, the paper uses the phase-synchronization measure $S_p$, defined from the variance of the relative phase quadrature $\\delta p_-$, and the Gaussian quantum discord $D_G$ computed from the linearized fluctuation covariance matrix. These two observables, mapped over frequency detuning $\\delta/\\omega_{mL}$, coupling strength $\\lambda/\\kappa$, and transmission loss $\\eta$, produce the tongue and blockade diagrams.","core_discovery":"The central claim is that unidirectional optical coupling in a cascade of two optomechanical oscillators, described by the instantaneous input relation $a_{\\rm in}^R(t)=\\sqrt{\\eta}\\,a_{\\rm out}^L(t)$ with $\\tau=0$, produces a quantum synchronization blockade: identical mechanical oscillators ($\\delta=0$) synchronize less than moderately detuned ones, and the optimum $\\delta$ scales with transmission loss $\\eta$. In the same cascade, Gaussian quantum discord $\\langle D_G\\rangle$ does not exhibit the blockade and instead traces out a tongue-shaped region reminiscent of the bidirectional case. The paper also finds that in the bidirectional topology both $\\langle S_p\\rangle$ and $\\langle D_G\\rangle$ form quantum Arnold tongues centered at resonance, that the blockade survives at thermal occupation $n_{\\rm th}=10$, and that no logarithmic negativity (entanglement) is generated in either configuration.","pith_inferences":["A practical design consequence the authors do not spell out: in a unidirectional optomechanical link, one should intentionally detune the receivers to maximize phase synchronization, with the optimal detuning set by the transmission loss $\\eta$.","The failure of Gaussian discord to reproduce the blockade suggests that phase synchronization carries directional phase-space information that second-moment correlation measures discard; higher-order or non-Gaussian witnesses might recover the blockade.","A finite-delay ($\\tau>0$) version of the cascade is the natural next check: if the blockade survives delay, it becomes a reliable design tool; if not, it is an artifact of the instantaneous-input approximation.","Because no entanglement is produced, the synchronized states here are non-entangled but discordant; that makes them candidates for one-way quantum information tasks that need directional correlations without entanglement."],"forward_implications":["In bidirectionally coupled optomechanical arrays, synchronization and Gaussian discord can be used interchangeably as indicators: both give a resonance-centered tongue in detuning–coupling space.","In unidirectionally coupled chains, the maximum phase synchronization occurs between non-identical oscillators; the optimal detuning is set by the transmission loss $\\eta$.","Gaussian quantum discord is not a universal witness of quantum synchronization: it can fail to reveal a synchronization blockade, so experiments measuring only correlations may misread the synchronization phase diagram.","The synchronization blockade persists at mechanical occupations up to $n_{\\rm th}=10$, indicating the effect is not limited to ground-state mechanics, though thermal noise degrades the degree of synchronization.","No entanglement (negative logarithmic negativity) appears in either topology, so the correlations generated are of the discord-only, non-entangled type."],"supporting_citations":[{"why":"Provides the phase-quadrature figure of merit $S_p$ used to quantify quantum phase synchronization.","marker":"[32]"},{"why":"First reported synchronization blockade in nonlinearly coupled oscillators and gave the energy-mismatch explanation this paper applies to its unidirectional result.","marker":"[38]"},{"why":"Demonstrated synchronization blockade in optomechanical systems, providing the immediate precedent and terminology.","marker":"[40]"},{"why":"Supplies the definition of Gaussian quantum discord $D_G$ used to compare with synchronization.","marker":"[44]"},{"why":"Shows quantum noise can prevent synchronization of identical optomechanical oscillators, motivating the need to study coupled topologies.","marker":"[29]"},{"why":"Provides the logarithmic negativity criterion the paper uses to assess entanglement.","marker":"[45]"}],"fun_headline_variants":["Quantum sync blockade: off-resonance beats resonance in cascaded optomechanics","Unidirectional coupling flips sync: detuned oscillators sync better than identical","Quantum sync blockade: detuning beats matching in one-way optomechanical link","Cascade optomechanics: sync peaks off resonance, discord misses the blockade","In one-way optomechanical link, detuned sync beats resonance sync"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The unidirectional blockade rests on the assumption that light travels from the left to the right cavity instantly ($\\tau=0$); if a realistic propagation delay breaks this exact phase relation, the suppression at resonance and the off-resonance peak may not survive.","fun_headline_variants_meta":{"raw":{"variants":["Quantum sync blockade: off-resonance beats resonance in cascaded optomechanics","Unidirectional coupling flips sync: detuned oscillators sync better than identical","Quantum sync blockade: detuning beats matching in one-way optomechanical link","Cascade optomechanics: sync peaks off resonance, discord misses the blockade","In one-way optomechanical link, detuned sync beats resonance sync"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000734,"raw_usage":{"total_tokens":3282,"prompt_tokens":942,"completion_tokens":2340,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":558,"completion_tokens_details":{"reasoning_tokens":2239}},"tokens_in":558,"tokens_out":2340,"duration_ms":14452,"temperature":1.0,"reasoning_tokens":2239,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:20:56.643765+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute or measure the time-averaged synchronization $\\langle S_p\\rangle$ for the unidirectional cascade with a finite propagation delay $\\tau$ of order one mechanical period ($1/\\omega_{mL}$). If the minimum at $\\delta=0$ and the maximum at finite detuning disappear or invert, the claimed quantum synchronization blockade does not hold.","supporting_citations":[{"cited_title":"Bemani, A","cited_arxiv_id":null,"evidence_quote":"First reported synchronization blockade in nonlinearly coupled oscillators and gave the energy-mismatch explanation this paper applies to its unidirectional result."},{"cited_title":"Witthaut, S","cited_arxiv_id":null,"evidence_quote":"Demonstrated synchronization blockade in optomechanical systems, providing the immediate precedent and terminology."},{"cited_title":"Ying, Y.-C","cited_arxiv_id":null,"evidence_quote":"Supplies the definition of Gaussian quantum discord $D_G$ used to compare with synchronization."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the logarithmic negativity criterion the paper uses to assess entanglement."}],"review_version":1}