{"id":"1b65f6ba-77d9-4328-89e2-983304bfe9c5","arxiv_id":"1908.08894","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In the Kondo regime, molecular vibrations that would ruin destructive interference can be counteracted by tuning the orbital energy splitting, restoring near-zero conductance that survives finite temperature and bias voltage.","lead":"This paper calculates current through a model of a two-level molecule where quantum interference, electron repulsion, and molecular vibrations act together. It shows that tuning the energy gap between the molecule's orbitals can restore a nearly blocked current even when vibrations are present, which supports proposals for molecular interference transistors.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"NCA is the sole numerical engine for the full two-level Anderson-Holstein model, and its 10^-6 conductance null is not benchmarked; an exact NRG check is needed before the central robustness claim is accepted.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: all quantitative results—the restored conductance null, its depth below 10^-6 G0, and its persistence over temperatures and voltages up to a few T_K—come from NCA, and NCA is not benchmarked against an exact method for the full two-level Anderson-Holstein model. This is not a complaint about disagreement with consensus; it is a correctness risk in the specific quantity that the central claim depends on. NCA is known to be a large-N expansion whose quantitative accuracy at small N is not guaranteed, and the phonon coupling further breaks the SU(4) symmetry whose emergent restoration is the proposed mechanism. The T=0 Fermi-liquid formula gives independent support for the zero-bias null in the purely electronic case, but the paper extends it to the phonon case through an effective-description argument, and the asymmetric case is explicitly left without proof. The right response is not rejection, because the model and mechanism are plausible and the zero-temperature analytic result provides real support, but the conditions proposed by the reader—an exact benchmark and released implementation details—are exactly what would settle the concern. Hence the verdict remains conditional, with no adjustment.","tokens_in":15958,"tokens_out":8413,"duration_ms":99710,"concrete_test":"Use NRG to compute the equilibrium zero-bias conductance of the full Hamiltonian Eq. (1) for the asymmetric parameters of Fig. 5 (Omega=0.1, lambda=sqrt(10)*10^-2, Delta_L=0.075, Delta_R=0.0025, E_d=-0.4, delta=delta_res=0.007) at T ~ T_K^SU4/20. If the NRG conductance is not below 10^-6 G0, or if the null occurs at a substantially different delta, the central robustness claim is not established. For the finite-bias part of the claim, complement with a time-dependent MPS or PT-MPO calculation at |eV_b| = 4T_K.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that at a fixed delta=delta_res the conductance stays below 10^-6 G0 over |V_b| < 4T_K and T < T_K for both symmetric and asymmetric leads—is produced entirely by the noncrossing approximation (NCA) applied to the full model with Kondo physics, electron-vibration coupling, and two-level interference. NCA is validated only against a one-level Anderson-Holstein model and a two-level model without phonons; those benchmarks do not establish that NCA reproduces the near-degeneracy of the two Kondo resonances that is the physical origin of the restored null. Because NCA is a controlled large-N expansion while the model has only spin SU(2) symmetry (and no orbital symmetry in the asymmetric case), the absolute depth of the conductance dip is not controlled. The zero-temperature Fermi-liquid formula Eq. (3) supports the T=0 null only after an approximate 'integrate out phonons' step, and it says nothing about the claimed finite-bias/temperature plateau; the asymmetric-restoration result in Sec. III E has no analytic support at all, as the paper itself states. Thus the headline claim depends on an unbenchmarked approximation in exactly the regime where its magnitude matters.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies resonant transport through a two-orbital molecular junction with opposite-parity levels, infinite Coulomb repulsion, and a Holstein phonon coupled to one level. In the absence of vibrations the model has a zero-bias destructive-interference null at level degeneracy delta=0; the paper shows with the noncrossing approximation (NCA) that the electron-vibration coupling lifts this null, but that tuning the level splitting to a specific value delta_res restores the conductance to values below 10^-6 G0. The restoration is interpreted as the emergence of approximate SU(4) symmetry at low energies, and the authors argue that the restored null is robust over a voltage window |V_b| up to a few T_K and temperatures up to about T_K, for both symmetric and strongly asymmetric lead couplings. Analytic support is given by a Fermi-liquid formula for the zero-temperature conductance and by a unitary transformation that maps the delta=0 interference model to a multi-channel Anderson model; an appendix discusses the relevance to annulene molecules.","tokens_in":16237,"tokens_out":14160,"duration_ms":147050,"significance":"If correct, the central claim is significant: it identifies a many-body mechanism, emergent low-energy SU(4) symmetry, that makes destructive quantum interference robust against both vibrations and moderate bias/temperature, in contrast to the fragile null of noninteracting junctions. The analytic results (Eq. (3), Appendix C, and the symmetry argument for the asymmetric lambda=0 case) are clean and provide a useful foundation. The numerical NCA results are internally consistent, and the comparison with the noninteracting limit is instructive. However, the headline quantitative statements—the 10^-6 conductance floor and the width of the plateau—are produced entirely by NCA for the full model and are not yet benchmarked against an exact or controlled method; the significance of the numerical part is therefore conditional. I do not regard the numerical search for delta_res as circular, because the nontrivial claim is the width of the restored-interference window rather than the existence of a zero, but the paper should state this more explicitly.","major_comments":[{"comment":"Section II states that the transport is calculated within the noncrossing approximation (NCA), and the central quantitative claims in Sections III.C and III.E (G<10^-6 G0 over |eV_b|<4T_K and T<T_K) rest entirely on NCA for the full two-level Anderson-Holstein model. The validation cited in Section II covers only simpler limits: one Kondo level with phonons and two interfering levels without phonons. Because NCA is not a controlled expansion for this model (the symmetry is only SU(2) in the asymmetric case), the absolute depth of the conductance null and the width of the restored-interference plateau are not established. I request an independent check for at least the zero-temperature/zero-bias limit (for example, NRG or a comparable method combined with the Friedel sum rule), or a clear statement of the NCA uncertainty and a corresponding softening of the 10^-6 claim.","section":"II and III.E"},{"comment":"The finite-bias robustness of the restored null in the symmetric-leads case is demonstrated in Fig. 3 for a phonon-free model with renormalized parameters (lambda=0, Delta_1=0.941 Delta_2, delta=0.001499), not for the full Hamiltonian (1) at lambda>0. The argument that the full model behaves like this proxy relies on the similarity of the spectral densities in Fig. 2, but the quantitative statement about the plateau is not directly computed for the full model. Please provide direct G(V_b,T) results for the full model at delta=delta_res, or explicitly present Fig. 3 as a proxy and restrict the quantitative claims accordingly.","section":"III.C"}],"minor_comments":[{"comment":"The integrand notation '(-d[f_L(omega)-f_L(omega)]/d omega)' appears to contain a typo; the second Fermi function should presumably be f_R, and the derivative should be defined with respect to the bias voltage. Please check and correct.","section":"Eq. (4)"},{"comment":"The symbol 'greaterorsimilar' should be a standard inequality (for example, ≳); as printed it is corrupted.","section":"III.A"},{"comment":"The phrase 'the model tales the same form as Eq. (A4)' should read 'takes'.","section":"Appendix A"},{"comment":"The sentence 'As shown by the numeric results. this continues to be true' has a punctuation and capitalization error after 'results.'","section":"III.E"},{"comment":"The sentence about triangles ('Triangles correspond to lambda=0 displaced in delta_res=0.005') is unclear and should explain what is displaced and why.","section":"Fig. 1 caption"},{"comment":"There are several typographical errors, including 'references therin' in Ref. 15 and 'explicitely' in Appendix B; these should be corrected.","section":"References and Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The main gate for acceptance is methodological: the 10^-6 robustness claim needs an independent benchmark of NCA for the combined interference+Kondo+phonon model, at least in the equilibrium limit. The paper is within the journal's scope and the analytic parts are solid; I would not reject it, but I would not accept it without the benchmark or a clearly softened quantitative claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is a plausible theory paper with a clean core idea, and its central quantitative robustness claim is not yet proven. The authors study a two-level Anderson-Holstein model in the Kondo regime and show that a vibration-induced loss of destructive interference can be undone by tuning the level splitting delta, restoring the conductance null over a wide bias/temperature window. That is genuinely new relative to Ref. 29 (no phonons) and to Haertle et al.'s noninteracting treatments.\n\nThe paper does several things well. The T=0 Fermi liquid formula G=(e^2/h) sum_sigma sin^2[pi(<n2sigma - n1sigma>)] is clean, and Appendix C derives the conductance for arbitrary couplings via a basis rotation in a simple way. The contrast with the noninteracting case in Fig. 3 is convincing: the noninteracting null dies over a scale set by the hybridization, while the interacting null survives over several T_K. The emergent-SU(4) interpretation is physically reasonable and supported by the spectral densities.\n\nThe soft spots are real but not fatal. All finite-bias and finite-T results come from NCA, and NCA is not benchmarked against an exact method for this combined problem. The model has only spin SU(2) symmetry, so the 10^-6 plateau is not protected by a large-N limit. The paper validates NCA on the separate ingredients, but that does not guarantee the near-degeneracy of the spectral densities that produces the null. The T=0 analytic result itself depends on an integrate-out-phonons step that is not rigorous, though it is plausible. The asymmetric-coupling restoration has no analytic support, as the authors state. Also, delta_res is chosen by searching for G=0, so its existence is not a prediction; the nontrivial claim is the width of the null in V and T, which is exactly the part that relies on NCA. No code is released, which makes independent checking harder than it should be.\n\nThat said, the mechanism does not require NCA to be exact in every detail; the emergent SU(4) argument carries weight. But the quantitative headline, conductance below 10^-6 G0 for |V_b|<4T_K, should be treated with caution until at least one NRG or equivalent check exists.\n\nThis paper is for people working on quantum interference in molecular junctions and Kondo physics. I would send it to peer review: the idea deserves referee time, with acceptance conditioned on an NRG check for at least one symmetric and one asymmetric parameter set, and on releasing the calculation details. Without that, the authors should downplay the 10^-6 plateau and present it as a suggestive prediction.","headline":"A plausible mechanism for restoring destructive interference in the Kondo regime, but the claimed 10^-6 conductance plateau over a wide V/T window rests on an unbenchmarked NCA and should be verified before being cited as a quantitative result.","tokens_in":16759,"tokens_out":4986,"would_cite":true,"duration_ms":49343,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["73.63.-b","72.10.Fk","85.65.+h"],"model":"deepseek-v4-flash","headline":"By tuning the energy difference between two quasi-degenerate molecular levels, destructive interference can be restored even when molecular vibrations are present, as long as the junction sits in the Kondo regime.","keywords":["destructive quantum interference","Kondo effect","molecular junction","electron-vibration interaction","SU(4) symmetry","noncrossing approximation","quantum interference transistor","Anderson-Holstein model"],"falsifier":"A numerically exact calculation of the same two-level Anderson-Holstein model at zero temperature, using a method beyond the noncrossing approximation, should produce a conductance minimum near the restoring splitting; if the minimum is larger than about $10^{-4}$ G0, or if the conductance rises above $10^{-3}$ G0 for bias voltages around half the Kondo temperature, the central claim is false. Alternatively, a break-junction experiment with an odd-occupancy molecule that shows no sharp conductance dip when the molecule is stretched through the restoring splitting would falsify the predicted many-body quantum interference effect transistor.","tokens_in":15761,"feed_emoji":"⚛️","tokens_out":8232,"duration_ms":72539,"temperature":0.7,"pith_summary":"The paper studies a molecular junction in which two nearly degenerate electronic states of opposite parity give destructive interference, shutting off the current. It asks whether molecular vibrations, which usually destroy this interference, can be overcome when strong electron-electron repulsion puts the junction in the Kondo regime. The central claim is yes: by tuning the energy gap between the two levels, the conductance can be restored to below $10^{-6}$ of the quantum of conductance, even though vibrations are present, and this near-zero conductance survives for bias voltages and temperatures up to a few times the Kondo temperature. Changing the gap by more than the Kondo temperature then switches the conductance back up by more than three orders of magnitude, toward $2e^{2}$/h. If true, this gives a robust many-body version of the quantum interference effect transistor that works at low bias and low power.","feed_headline":"Retune two levels to restore zero conductance in a molecule","feed_subtitle":"In the Kondo regime, tuning the level gap restores near-zero conductance despite vibrations.","key_machinery":"The central object is SU(4) symmetry of the two-level Anderson model with opposite-parity couplings: for zero vibration coupling and equal lead couplings, a unitary transformation maps the interference model onto the SU(4) impurity Anderson model, in which perfect destructive interference forces the zero-bias conductance to zero. The electron-vibration interaction, treated within the noncrossing approximation for Keldysh Green's functions, breaks this symmetry by renormalizing the level energy and hybridization. The mechanism that restores interference is the approximate emergent SU(4) symmetry at low energies: when the splitting is tuned to the restoring value, the two spectral densities become nearly identical near the Fermi energy and the level occupancies equalize, making the Fermi-liquid conductance expression vanish. This identity explains the symmetric case and also provides the definition of the Kondo temperature used throughout the paper.","core_discovery":"The authors show that in the Kondo regime, the main low-energy effect of electron-vibration coupling is a renormalization of one level's energy and hybridization, which breaks the SU(4) symmetry that protects perfect destructive interference. When the bare level splitting is tuned to a compensating value, the two low-energy spectral densities become almost identical near the Fermi level, so an emergent SU(4) symmetry is approximately restored. At zero temperature the conductance then follows a Fermi-liquid formula that vanishes when the two level occupancies are equal; tuning the splitting to equalize the occupancies restores total destructive interference. The same restoration is found numerically for strongly asymmetric couplings to the leads, where no symmetry argument applies. The resulting conductance remains below $10^{-6}$ G0 for bias voltages and temperatures up to about four times the Kondo temperature, and rises by more than three orders of magnitude when the splitting moves away by more than the Kondo temperature.","pith_inferences":["If the emergent-symmetry mechanism is generic, similar restoration of destructive interference should occur in other multilevel systems where vibrations renormalize one level more than the others, such as double quantum dots with valley or pseudospin degrees of freedom.","The equal-occupancy condition suggests a practical experimental protocol: measure the two level occupancies with charge sensing while tuning the splitting, and the zero-conductance point should coincide with equal occupations.","A testable extension is to include a second phonon mode or a finite vibration lifetime; the restoration may persist only while temperature and bias remain below the vibration energy, so devices would need low-temperature operation.","The effect might survive a finite rather than infinite Coulomb repulsion, but the paper does not treat that case, so whether the restored zero-conductance window narrows or shifts would require an additional calculation."],"forward_implications":["A molecular transistor based on destructive interference can operate in the Kondo regime with a conductance contrast of more than six orders of magnitude, controlled by a single gate or stretch parameter.","The restored zero-conductance state survives finite bias and temperature up to a few times the Kondo temperature, making it more robust than its noninteracting counterpart, which loses perfect destructive interference for bias voltages of order the hybridization.","The effect works for strongly asymmetric lead couplings, a common situation in real molecular junctions, so it is not limited to idealized symmetric contacts.","Vibrations do not simply destroy interference; at low temperatures their main effect is a renormalization that can be compensated by tuning the level splitting, so quasi-degenerate levels remain useful for quantum interference effect transistors.","A concrete experimental signature is a sharp dip in the conductance as the level splitting is tuned through the restoring value, with conductance below 10^-6 G0 for bias and temperature up to about four times the Kondo temperature."],"supporting_citations":[{"why":"introduces the two-level interference model with electron-vibration coupling that this paper extends to the Kondo regime","marker":"[27]"},{"why":"shows that two interfering levels in the Kondo regime have zero conductance at zero splitting and increasing conductance with splitting","marker":"[29]"},{"why":"provides the noncrossing approximation treatment of one magnetic level coupled to phonons, validating the vibrational renormalization used here","marker":"[32]"},{"why":"reproduces the temperature scaling of Kondo satellite peaks observed in experiments, supporting the noncrossing approximation for Kondo plus electron-vibration interaction","marker":"[33]"},{"why":"gives the noncrossing approximation method used for the Keldysh Green's functions in the conductance calculation","marker":"[42]"},{"why":"formulates the noncrossing approximation for nonequilibrium transport, used in the numerical calculations","marker":"[43]"},{"why":"provides the current expression and conductance formula used to evaluate the differential conductance","marker":"[8]"},{"why":"shows that SU(4) symmetry can emerge at low energies even when couplings differ, the basis for the restoration argument","marker":"[30]"},{"why":"gives another example of emergent SU(4) symmetry, supporting the restoration mechanism","marker":"[31]"},{"why":"provides the generalized Friedel sum rule relating spectral density to occupation, used for the zero-temperature Fermi-liquid conductance formula","marker":"[56]"}],"fun_headline_variants":["Vibration-proof quantum interference: retune levels to suppress conductance","Kondo regime: retune level gap to restore zero conductance despite vibrations","Tuning two levels restores perfect destructive interference in vibrating molecules","Emergent symmetry restores zero conductance in vibrating molecular junctions","Zero-conductance switch: retune levels in vibrating molecules"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The numerical results rely entirely on the noncrossing approximation, which is benchmarked only on simpler models and never on the full Kondo-plus-vibration-plus-interference problem; if that approximation exaggerates how similar the two spectral densities become, the predicted restoration of near-zero conductance could vanish.","fun_headline_variants_meta":{"raw":{"variants":["Vibration-proof quantum interference: retune levels to suppress conductance","Kondo regime: retune level gap to restore zero conductance despite vibrations","Tuning two levels restores perfect destructive interference in vibrating molecules","Emergent symmetry restores zero conductance in vibrating molecular junctions","Zero-conductance switch: retune levels in vibrating molecules"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001237,"raw_usage":{"total_tokens":5026,"prompt_tokens":843,"completion_tokens":4183,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":459,"completion_tokens_details":{"reasoning_tokens":4106}},"tokens_in":459,"tokens_out":4183,"duration_ms":29002,"temperature":1.0,"reasoning_tokens":4106,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:26:37.115634+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A numerically exact calculation of the same two-level Anderson-Holstein model at zero temperature, using a method beyond the noncrossing approximation, should produce a conductance minimum near the restoring splitting; if the minimum is larger than about $10^{-4}$ G0, or if the conductance rises above $10^{-3}$ G0 for bias voltages around half the Kondo temperature, the central claim is false. Alternatively, a break-junction experiment with an odd-occupancy molecule that shows no sharp conductance dip when the molecule is stretched through the restoring splitting would falsify the predicted many-body quantum interference effect transistor.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"introduces the two-level interference model with electron-vibration coupling that this paper extends to the Kondo regime"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"shows that two interfering levels in the Kondo regime have zero conductance at zero splitting and increasing conductance with splitting"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the noncrossing approximation treatment of one magnetic level coupled to phonons, validating the vibrational renormalization used here"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"reproduces the temperature scaling of Kondo satellite peaks observed in experiments, supporting the noncrossing approximation for Kondo plus electron-vibration interaction"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the noncrossing approximation method used for the Keldysh Green's functions in the conductance calculation"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"formulates the noncrossing approximation for nonequilibrium transport, used in the numerical calculations"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the current expression and conductance formula used to evaluate the differential conductance"},{"cited_title":"Phys.: Condens","cited_arxiv_id":null,"evidence_quote":"shows that SU(4) symmetry can emerge at low energies even when couplings differ, the basis for the restoration argument"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives another example of emergent SU(4) symmetry, supporting the restoration mechanism"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the generalized Friedel sum rule relating spectral density to occupation, used for the zero-temperature Fermi-liquid conductance formula"}],"review_version":1}