{"id":"be88f9fd-bdeb-4447-8d26-9a8b48fab2dc","arxiv_id":"2509.00526","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Geometric phase-induced destructive interference in conical intersection dynamics survives non-Markovian dissipation from vibrational and electronic baths, as shown by numerically exact LDR-HEOM simulations.","lead":"By simulating a model molecule with two crossing energy surfaces, the authors show that the destructive interference pattern produced by the geometric phase survives even when the molecule is strongly coupled to a dissipative environment. The work suggests that geometric phase effects, often ignored in condensed-phase simulations, may matter in realistic solvents.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Robustness for asymmetric initial states rests on numerics with no convergence tests; the analytic cancellation proof does not cover this case.","rationale":"The reader identified the reflection symmetry of paths and bath operators as the weakest assumption. I agree that the analytic proof depends on this symmetry, but the more load-bearing issue is that the paper explicitly extends the claim to asymmetric initial states (Fig. 8) where the symmetry argument is inapplicable. In that regime, the central claim rests entirely on numerically exact simulations, yet no convergence documentation is provided. This is an addressable evidentiary gap rather than a demonstrated flaw, so the appropriate verdict remains conditional: the paper's analytic argument supports the symmetric case, but the general robustness claim requires convergence verification or an extension of the analytic proof to asymmetric initial conditions. My concern overlaps partially with the reader's, but focuses on the asymmetric-case extrapolation and the missing numerical validation rather than on the symmetry assumption itself.","tokens_in":11147,"tokens_out":13042,"duration_ms":165169,"concrete_test":"Reproduce the asymmetric initial-state dynamics (Fig. 8, electronic bath λ=2ω1) with HEOM truncation levels L=4,6,8 and DVR grids Nx×Ny = 128×128, 256×256, 512×512. Compute the nuclear probability minimum on S1 at y=0, p1(x=0,y=0,tmax), for each setting. If the value changes by more than 20% between the two largest grids/truncations, the result is not converged and the robustness claim remains unverified. Additionally, for the symmetric initial state (y0=0), verify that the nodal line is exactly zero to machine precision; if not, the LDR-HEOM implementation is not exact and the central claim is suspect.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central analytic argument (Eqs. 26-27) is a pairwise cancellation: for every forward path Γ+ there is a mirror path Γ- with identical endpoints, identical x(t), and identical influence functional, so the two amplitudes cancel exactly. This requires the initial and final states to be symmetric under y→-y so that reflected paths share the same endpoints. The asymmetric initial state (y0=0.5) in Fig. 8 breaks this condition: the reflected path connects (x0,-0.5) to (xf,-yf), not the same endpoints, so the pairwise cancellation no longer applies. The paper's conclusion that the surviving pattern is 'protected by the topology of the CI, not only by the space reflex symmetry' is therefore not supported by the analytic proof; it relies entirely on HEOM simulations. No convergence tests, DVR grid sizes, hierarchy truncation levels, or time-step checks are reported. If the apparent nodal line in the asymmetric case is a numerical artifact of insufficient basis or truncation, the claim that geometric-phase interference is robust in generic (non-symmetric) dissipative environments is unestablished. Even in the symmetric case the proof yields an exact node, while in the asymmetric case one expects only a dip; the paper does not quantify how deep or stable this dip is.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies whether geometric-phase-induced destructive interference in the nuclear probability distribution, a hallmark of conical intersections, survives when the molecule is coupled to dissipative environments. The authors use a two-state, two-dimensional vibronic coupling model and propagate the system with a combination of the local diabatic representation (LDR) and the hierarchical equations of motion (HEOM). They consider both a vibrational bath (Q=x) and an electronic bath (Q=Π1), scanning coupling strength and temperature. The reported density plots show a nodal line along y=0 that persists under dissipation, including for an asymmetric initial wavepacket. The authors explain this by a Feynman path-integral argument: for reflection-symmetric path pairs around the conical intersection, the bath influence functional is identical for the two interfering amplitudes, so the geometric-phase phase factor still enforces destructive interference.","tokens_in":11428,"tokens_out":8592,"duration_ms":111117,"significance":"If the result holds, it is significant: it suggests that geometric-phase effects, usually neglected in condensed-phase dynamics, can survive strong non-Markovian dissipation. The path-integral cancellation argument is elegant and parameter-free for the symmetric case, and the LDR-HEOM combination is a reasonable methodological tool. The paper gives credit to the analytic mechanism and presents falsifiable numerical predictions. However, the strongest claim—robustness for asymmetric initial states and protection by topology rather than symmetry—goes beyond the analytic proof and rests on simulations for which no convergence details are reported; this limits the current verifiability of the central conclusion.","major_comments":[{"comment":"The conclusion that the surviving pattern is 'protected by the topology of the CI, not only by the space reflex symmetry' is not supported by the analytic argument. Equations (26)-(27) require the initial and final states to be invariant under y -> -y so that mirror paths share endpoints; for the y0=0.5 initial state in Fig. 8 this condition fails and no exact nodal line is expected. The robustness claim for this case therefore rests entirely on the HEOM results, but no convergence tests are given and the nodal feature is shown only as color maps. Please provide quantitative line cuts through y=0 (dip depth versus time and bath parameters) and convergence tests for this asymmetric case, or soften the topological-protection claim.","section":"Sec. III, Fig. 8 and Conclusion"},{"comment":"The simulations are repeatedly called 'numerically exact', but the manuscript reports no HEOM truncation level (L in Eq. (21)), no number of exponential terms K in Eq. (19), no DVR grid size, no time step, and no convergence tests. Since the title and abstract make a robustness claim, a reader cannot verify that the nodal line is not a numerical artifact. Please report these parameters and show at least one convergence check (e.g., dependence on L_max and grid spacing).","section":"Secs. II C and III"},{"comment":"The pairwise-cancellation proof is stated too broadly: 'as this applies to all pairs of paths' is not correct. W[Γ+]+W[Γ-] = 0 only when the two mirror paths together enclose the CI; paths whose closed loop does not encircle the CI (e.g., final xf left of the CI) need not cancel. The proof should either restrict the sum to encircling pairs or add an argument that non-encircling paths do not contribute at y=0. Relatedly, the 'only rigorously valid near the CI' caveat should be reconciled with its use to explain the numerical results over the whole wave packet.","section":"Sec. III, Eqs. (26)-(27)"},{"comment":"The extension of Eq. (26) to Q=Π1 is asserted rather than demonstrated. The equality of influence functionals for mirror paths holds here because, in the LDR basis, Π1 has path value δ_{α,1}, which is invariant under the reflection symmetry; if Π1 is interpreted in the raw diabatic basis, it is not y-invariant. Please state this explicitly to make the electronic-bath argument self-contained.","section":"Sec. III, electronic bath paragraph"}],"minor_comments":[{"comment":"There are typos: 'nuclears' should be 'nuclear', 'influce' should be 'influence', and 'envrionments' should be 'environments'.","section":"Abstract and Sec. I"},{"comment":"'reflex symmetry' should be 'reflection symmetry' (e.g., Sec. III and Fig. 8 discussion).","section":"Throughout"},{"comment":"The notation R± and F± is undefined; define these objects, since the equation is central to the cancellation argument.","section":"Eq. (27)"},{"comment":"Typo: 'nodel line' should be 'nodal line'.","section":"Fig. 4 caption"},{"comment":"Reference 22 is an arXiv preprint; please add a published version/DOI if available before final publication.","section":"References"},{"comment":"The corresponding-author email contains 'weatlake'; should be 'westlake'.","section":"Title page"}],"recommendation":"major_revision","confidential_remarks":"The paper has a solid core: the symmetric-case cancellation is a valuable exact statement, and the numerical setup is appropriate for the problem. My main concern is that the headline robustness claim for asymmetric initial states is supported only by unconverged-looking numerics, and the analytic proof is somewhat overstated. I would encourage the editor to require quantitative nodal-depth measures and convergence documentation before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, the takeaway: the paper makes a real first step—showing with numerically exact (in principle) LDR-HEOM that the geometric-phase nodal line in a two-mode model survives both vibrational and electronic non-Markovian baths. The path-integral argument for the symmetric case is clean: for reflection-symmetric path pairs, the influence functional factors out and the geometric phase cancellation goes through unchanged. That part holds up.\n\nThe simulations look consistent with that picture: nodal line persists across the parameter scans (coupling strength, temperature) for both bath types. The electronic bath construction is a bit more than a generic Q=Π1 coupling because of the reorganization term, but they handle it explicitly in Appendix A. The paper is honest about the near-CI caveat on the analytic proof.\n\nWhere it gets soft: the asymmetric initial state (Fig. 8) is the load-bearing part of the \"topological, not symmetry\" claim, but the analytic cancellation does not apply there—reflection partners don't share endpoints. So that claim rests entirely on the HEOM numerics, and the paper reports no convergence tests: no hierarchy truncation level, no grid size, no time step, no check that the nodal line is not a numerical artifact or merely a shallow dip. The phrase \"numerically exact\" is doing a lot of work without the usual supporting documentation. Also no code or data. That doesn't make the result wrong, but it makes the strongest claim unverifiable as written.\n\nThe self-citations to LDR are fine; it's a published method and they're using it transparently.\n\nBottom line: worth a serious referee. The symmetric-case result and the path-integral explanation are solid, and the question—whether geometric phase effects can be neglected in condensed-phase nonadiabatic dynamics—is worth a clear answer. But the referee should require convergence documentation and a quantification of the dip depth for the asymmetric case, plus a softened claim if the dip is not an exact node. I'd want those details before citing the robustness as a general result.","headline":"Solid demonstration that the geometric-phase nodal line survives symmetric dissipation; the asymmetric-case robustness claim needs convergence data before it can be fully trusted.","tokens_in":11872,"tokens_out":1715,"would_cite":false,"duration_ms":20456,"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":"Geometric phase interference survives quantum dissipation","keywords":["geometric phase","conical intersection","nuclear quantum interference","non-Markovian dissipation","hierarchical equations of motion","local diabatic representation","vibronic coupling","wave packet dynamics"],"falsifier":"Repeat the same two-state simulation with a bath that couples to the coupling coordinate y, or with an asymmetric bath that breaks the y→−y reflection; if the nodal line washes out at large coupling strength or temperature, the robustness is conditional on symmetry rather than intrinsic to the geometric phase.","tokens_in":11053,"feed_emoji":"🧪","tokens_out":3628,"duration_ms":44978,"temperature":0.7,"pith_summary":"The paper asks whether the destructive interference pattern created by a conical intersection's geometric phase can survive when the molecule is embedded in a dissipative environment. Using numerically exact simulations that combine a local diabatic representation with hierarchical equations of motion, it shows that the nodal line in the nuclear probability distribution remains intact for both vibrational and electronic baths, even under strong coupling and at elevated temperature. The key is a path-integral argument: for mirror-image pairs of paths around the conical intersection, the bath's influence functional cancels out of the interference, so the geometric-phase sign difference still enforces destructive interference. If this is right, geometric phase effects cannot be dismissed in condensed-phase chemistry, even when the environment strongly changes populations and relaxation rates.","feed_headline":"Geometric phase interference survives solvent dissipation","feed_subtitle":"Exact simulations show the nodal pattern from a conical intersection is not erased by strong vibrational or electronic noise.","key_machinery":"The central object is the local diabatic representation (LDR): a nuclear discrete-variable grid in which each grid point carries a fixed adiabatic electronic state evaluated at that geometry, so geometric phase enters through overlap matrices between neighbouring grid points while derivative-coupling singularities are avoided. Within LDR, path integrals give each electronic-nuclear path a dynamical action S and a geometric weight W[ξ(t)], and a Gaussian bath contributes a Feynman-Vernon influence functional F[Q(ξ(t)),Q(ξ′(t))]. The decisive step is that for reflection-symmetric path pairs surrounding the conical intersection, W[Γ+]+W[Γ−]=W[Γ−](1+e^{iπ})=0 and F is identical for the two paths","core_discovery":"The paper claims that, although dissipation alters nonadiabatic transitions and the nuclear density distribution, the destructive interference pattern induced by geometric phase stays robust for both non-Markovian vibrational and electronic environments. In the path-integral picture, two adiabatic paths surrounding a conical intersection acquire a relative phase of π through the geometric weight W[ξ(t)], and when their dynamical actions are equal the amplitudes cancel. For a bosonic bath with linear coupling Q⊗X, the influence functional F depends on the trajectory of Q; for the vibrational bath Q=x and for the electronic bath Q=Π1, mirror-image paths Γ+ and Γ− give the same F. The influence","pith_inferences":["The cancellation mechanism implies that any bath operator that is even under the same reflection symmetry that pairs the two interfering paths should be geometric-phase preserving; mapping which system-bath couplings violate this condition would delimit the effect.","Because the analytic argument requires reflection symmetry and is only rigorously local to a conical intersection, strongly asymmetric solvation or a bath coupled directly to the inter-system mode y could erase the node; a systematic scan over bath-coupling geometries would test that boundary.","The same path-pair reasoning should extend to other topological phase factors, such as multiple conical intersections, suggesting that Gaussian dissipation cannot generically destroy topological interference, only alter its surrounding dynamics."],"forward_implications":["Bath-modified reaction rates and electronic populations can coexist with an intact geometric-phase node, so population dynamics alone is not a reliable proxy for whether geometric phase interference survives.","Condensed-phase nonadiabatic simulations that omit the geometric phase will miss a persistent destructive-interference feature rather than a small correction.","Vibrational relaxation and electronic dephasing affect the nuclear density and electron population differently, yet both leave the y=0 nodal line intact.","The survival of the node under asymmetric initial conditions indicates that the protection is topological, not a consequence of symmetric initial-state preparation."],"supporting_citations":[{"why":"Supplies the local diabatic representation used to describe vibronic couplings without derivative-coupling singularities.","marker":"[35]"},{"why":"Provides the Strang-splitting Fourier-basis implementation of LDR that the simulations rest on.","marker":"[36]"},{"why":"Gives the Feynman-Vernon influence functional that enters the path-integral cancellation argument.","marker":"[34]"},{"why":"Provides the hierarchical equations of motion formalism used as the numerically exact open-system solver.","marker":"[37–43]"},{"why":"Defines the two-state two-dimensional vibronic coupling model that mimics the phenol-like conical intersection studied here.","marker":"[44]"},{"why":"Establishes the Berry phase factor of π acquired by adiabatic states on loops around a conical intersection.","marker":"[21]"},{"why":"Supports the identification of the geometric phase nodal line in nuclear wave packet dynamics.","marker":"[23]"}],"fun_headline_variants":["Dissipation fails to erase geometric phase interference","Geometric phase interference survives quantum dissipation","Conical intersection interference shrugs off solvent noise","Exact dynamics show geometric phase interference robust"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The cancellation argument assumes that the two interfering paths and the bath coupling operator are mirror images under a reflection symmetry, and it is strictly valid only near a conical intersection; away from that regime, the robustness rests on the numerical simulations rather than on proof.","fun_headline_variants_meta":{"raw":{"variants":["Dissipation fails to erase geometric phase interference","Geometric phase interference survives quantum dissipation","Conical intersection interference shrugs off solvent noise","Exact dynamics show geometric phase interference robust"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000226,"raw_usage":{"total_tokens":1240,"prompt_tokens":614,"completion_tokens":626,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":358,"completion_tokens_details":{"reasoning_tokens":571}},"tokens_in":358,"tokens_out":626,"duration_ms":8028,"temperature":1.0,"reasoning_tokens":571,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T13:29:30.612209+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the same two-state simulation with a bath that couples to the coupling coordinate y, or with an asymmetric bath that breaks the y→−y reflection; if the nodal line washes out at large coupling strength or temperature, the robustness is conditional on symmetry rather than intrinsic to the geometric phase.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the Feynman-Vernon influence functional that enters the path-integral cancellation argument."}],"review_version":1}