{"id":"88542ec0-dc64-415a-8945-eb6d582decc0","arxiv_id":"1908.04077","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"Inter-subsystem Ehrenfest identities for nuclear momentum, angular momentum, and kinetic energy are derived from the exact-factorization nuclear Schrödinger equation, all governed by one effective electromagnetic force operator.","lead":"This paper derives exact equations for how energy, momentum, and angular momentum flow between electrons and nuclei, using the exact factorization of the full wave function. The three equations share a single effective electromagnetic force operator, giving a compact language for electron-phonon energy transfer and molecular dynamics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The energy-transfer headline is not supported: Eq. (28) is exact only for the auxiliary kinetic energy \\tilde T_n = T_n - E_geo, and no expression or bound for dE_geo/dt is given, so the claimed electron–nuclear energy transfer is not actually characterized.","rationale":"The reader's weakest_assumption, smoothness of the exact-factorization potentials and vanishing surface terms, is a technical caveat that is likely patchable and does not threaten the central formal identities for square-integrable states. The E_geo gap is more load-bearing: it affects the paper's advertised central quantity, energy transfer between electrons and nuclei, and the text supports the needed smallness only by a private communication. The exactly solvable model cannot settle it because it is a nuclear-only model with imposed potentials, not an actual electron-nuclear wavefunction. The formal identities for momentum, angular momentum, and \\tilde T_n are credible, so the reader's CONDITIONAL verdict is appropriate. The concern is the same as the reader's rationale item (i), but it is not the reader's stated weakest assumption, hence partial agreement.","tokens_in":19943,"tokens_out":18576,"duration_ms":194654,"concrete_test":"Run an exact-factorization simulation of a nonadiabatic process, e.g., H2+ in a short laser pulse or the two-surface model of Agostini et al., J. Chem. Phys. 142, 084303, and compute T_n(t), \\tilde T_n(t), and E_geo(t) = T_n(t) - \\tilde T_n(t) using Eqs. (6) and (9). Compare dT_n/dt = d\\tilde T_n/dt + dE_geo/dt with the RHS of Eq. (28), Re⟨χ|∑_μ \\hat F_μ·\\hat v_μ|χ⟩. If |dE_geo/dt| exceeds, say, 10% of |d\\tilde T_n/dt| at any time, then Eq. (28) does not describe the actual electron-nuclear energy transfer, and the paper must either supply dE_geo/dt or explicitly restrict the energy-transfer conclusion to \\tilde T_n.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Equations (6) and (9) define E_geo = ⟨χ|∑_μ (1/2M_μ)(⟨∇_μΦ|∇_μΦ⟩ - A_μ²)|χ⟩ and T_n = \\tilde T_n + E_geo. The Heisenberg derivation leading to Eqs. (14) and (28) is applied to \\hat t_n = ∑_μ (1/2M_μ)(-i∇_μ+A_μ)², so the left side is d\\tilde T_n/dt, not dT_n/dt. For the true nuclear kinetic energy, dT_n/dt = d\\tilde T_n/dt + dE_geo/dt. The paper gives no formula for dE_geo/dt and no quantitative estimate; the only support is a personal communication [36] that E_geo is small 'in many cases'. Therefore the abstract and title claim that the identities characterize energy transfer between electrons and nuclei is not established. The momentum and angular momentum identities (26)–(27) are unaffected, since P_n and L_n have no E_geo-type correction. The numerical examples do not resolve this: the two-nucleus model has no electrons, so it cannot calibrate the size of dE_geo/dt in real molecules or at conical intersections. This is a scope/interpretation gap, not an inconsistency in the commutator algebra of (26)–(28) for \\tilde T_n.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript derives three inter-subsystem Ehrenfest identities from the exact-factorization nuclear time-dependent Schrödinger equation: one for the nuclear momentum of each nucleus, Eq. (26); one for the nuclear angular momentum, Eq. (27); and one for the total nuclear kinetic energy, Eq. (28). The central structural claim is that a single effective electromagnetic force operator F̂_μ = E_μ + B_μ × v̂_μ + D̂_μ, built from the exact-factorization scalar and vector potentials and their Berry curvatures, appears in all three identities. The authors also introduce a decomposition of the magnetic contribution into an intranuclear (classical Lorentz-like) term and an internuclear term with no classical analog, and they illustrate the momentum and energy identities with a two-nucleus one-dimensional model with analytically constructed potentials.","tokens_in":20256,"tokens_out":3340,"duration_ms":36016,"significance":"If the identities are correct as stated, they provide a compact, representation-independent characterization of momentum, angular momentum, and kinetic-energy exchange between electrons and nuclei in the exact-factorization framework, and they unify previously known trajectory-based force expressions with operator identities. The momentum and angular momentum identities (26) and (27) are well supported by the algebraic derivations and are a useful contribution. The energy identity, however, is exact only for the auxiliary quantity T̃_n = T_n − E_geo, not for the true nuclear kinetic energy T_n, so the manuscript's headline claim that it characterizes energy transfer between electrons and nuclei is not yet established. The numerical example is a consistency check on the algebra rather than a test of the energy-transfer interpretation.","major_comments":[{"comment":"The energy IEI, Eq. (28), governs T̃_n, not the exact nuclear kinetic energy T_n. As the authors note in Eq. (6), T_n = T̃_n + E_geo, with E_geo defined in Eq. (9), yet no formula for dE_geo/dt is provided and no quantitative estimate of its size is given; the only support for neglecting it is Ref. [36], a personal communication stating that E_geo is 'small in many cases'. Consequently, the abstract and title claims about energy transfer between electrons and nuclei are not supported for the exact T_n. The momentum and angular momentum identities are unaffected because P_n and L_n have no E_geo-type correction, but the energy identity must either be derived for T_n or the claims must be explicitly restricted to T̃_n with a clearly stated caveat.","section":"Eq. (6) and Eq. (28)"},{"comment":"The derivations assume that the exact-factorization potentials A_μ(R,t) and ϵ(R,t) are sufficiently smooth and that all boundary terms from integration by parts vanish. This assumption enters directly in the commutator algebra leading to Eqs. (10)–(13), (20), and the angular momentum derivation in the SI. It is not justified at nuclear configurations where χ has nodes or where the conditional electronic wavefunction Φ_R becomes singular, as can occur, for example, at conical intersections. Since all three identities rely on this assumption, the paper should either state the precise regularity conditions under which the identities hold or show that the identities extend to distributional settings with appropriate boundary-term cancellations.","section":"Eqs. (5), (10)–(13) and the associated integration by parts"},{"comment":"The numerical example contains no electronic degrees of freedom: it is a two-nucleus, one-dimensional model in which a Gaussian χ is prescribed and potentials A_1, A_2, and ϵ are reverse-engineered through Eqs. (31)–(34). The model therefore cannot calibrate the magnitude of E_geo or of dE_geo/dt in a real molecule, and it cannot demonstrate electron–nuclear energy transfer; it only verifies the internal algebraic consistency of the identities for a constructed χ. The caption and text present this as a validation of the IEIs, which is fair for the momentum identity, but the strength of the claim should be adjusted so that this example is not read as empirical or physical support for the energy-transfer interpretation.","section":"Exactly solvable model, Eqs. (29)–(34) and Fig. 1"}],"minor_comments":[{"comment":"The title and abstract contain a typographical spacing error in 'Ele ctrons'; this should be corrected.","section":"Title and Abstract"},{"comment":"The claim that E_geo is 'small in many cases' is supported only by a personal communication. Since this statement is now load-bearing for the energy identity, the authors should provide a published quantitative example or a derivation of a bound, rather than an unreferenced private communication.","section":"Ref. [36]"},{"comment":"The text introducing the operator Ĝ_μ states 'Derivation is omitted.' Because Ĝ_μ is used to explain why the individual-nucleus kinetic energy does not satisfy the simple IEI, the derivation (or a citation to a published derivation) should be included or the remark should be marked as a conjecture.","section":"SI, Section I.D, Eq. (S54)"},{"comment":"The statement that the angular-momentum IEI has been 'verified numerically (results not shown)' is a claim of verification without supporting data; either include the figure or describe the verification procedure in enough detail for the reader to reproduce it.","section":"SI, Section III.A"},{"comment":"The expression for ϵ in Eq. (31) is obtained by taking the real part of Eq. (29) and dividing by χ, which requires χ ≠ 0; the text should note that this construction is valid only where χ is nonzero, consistent with the smoothness assumptions flagged in the major comments.","section":"Eq. (31)"}],"recommendation":"major_revision","confidential_remarks":"The core algebraic content of the momentum and angular momentum identities appears sound, and the unification of the force operator is a worthwhile contribution. The main risk is that the paper's most prominent claim—energy transfer between electrons and nuclei—is not supported for the exact nuclear kinetic energy, and the gap is papered over by a personal communication. This is fixable by either deriving dE_geo/dt or carefully reframing the title and abstract. I would encourage the editor to require the authors to address the smoothness/boundary-term assumptions as well, since they are not merely technical in exact factorization at conical intersections. The paper is likely acceptable for the journal after these revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe momentum and angular momentum identities are the real content here: exact, representation-independent statements that the force operator \\hat F_mu appears in both. The kinetic-energy identity is also derived correctly, but for \\tilde T_n, not the true nuclear kinetic energy T_n. The abstract glosses over that, and the stress-test note is right that dE_geo/dt is not characterized. That said, the paper is transparent in the main text about the distinction; the problem is the packaging.\n\nWhat's genuinely new: the angular momentum IEI, the uniform force-operator structure across all three identities, and the clean decomposition of the effective magnetic field into intranuclear Berry curvature (classical Lorentz force) and internuclear curvature (the operator \\hat D, which has no classical analog). The proof that \\hat D does no net work is neat. The commutator algebra in the main text and SI is sound; I didn't find a step that fails. The exactly solvable two-nucleus model is a clever way to validate the subsystem identities, and it confirms the momentum and kinetic-energy equations, including the \\hat D correction.\n\nThe soft spots are in proportion: (i) The energy-transfer headline is not fully supported. Eq. (28) gives d\\tilde T_n/dt, and T_n = \\tilde T_n + E_geo. The paper leans on a personal communication to assert E_geo is usually small, but gives no formula or bound for dE_geo/dt. For practical claims about electron-phonon energy transfer, that matters. (ii) The \\hat G_mu correction in the SI is asserted without derivation. Minor, since it's for individual-nucleus kinetic energy and sums to zero. (iii) The model contains no electrons, so it tests the factorization algebra, not the actual electron-nuclear coupling. That's acceptable for an illustration, but it shouldn't be oversold. (iv) The smoothness assumptions are standard; a remark on node behavior would be responsible but not required.\n\nNet: this is a careful formal paper that deserves a serious referee. With a rewritten abstract that says \\tilde T_n explicitly, and ideally some estimate of E_geo for realistic systems, the energy claim would be honest. I'd cite the momentum and angular momentum identities, and would bring it to a reading group focused on exact factorization.\n\nRecommendation: send to peer review, with the expectation of a moderate revision focusing on the energy-identity framing.","headline":"Solid formal contribution: exact momentum and angular momentum IEIs, plus a kinetic-energy identity for \\tilde T_n that the abstract overstates.","tokens_in":20784,"tokens_out":3006,"would_cite":true,"duration_ms":29955,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["31.15.E-","71.10.-w","71.15.Mb"],"model":"deepseek-v4-flash","headline":"Exact subsystem Ehrenfest identities trace electron–nuclear momentum, angular momentum, and kinetic-energy transfer to a single effective electromagnetic force operator.","keywords":["exact factorization","subsystem Ehrenfest identities","electron-nuclear momentum transfer","Berry curvature","nonadiabatic dynamics","effective electromagnetic force","nuclear kinetic energy","electron-phonon energy transfer"],"falsifier":"Take a two-nucleus model in which the nuclear wave packet $\\chi$ develops a nodal line while the vector potentials remain smooth, compute $d\\mathbf P_1/dt$ directly from the time-dependent $\\chi$, and compare it with $\\mathrm{Re}\\langle\\chi|\\hat{\\bar{\\mathbf F}}_1|\\chi\\rangle$; any nonzero difference would show that the discarded surface terms do not vanish and would refute Eq. (26) in that regime.","tokens_in":19693,"feed_emoji":"🧲","tokens_out":8980,"duration_ms":82963,"temperature":0.7,"pith_summary":"This paper establishes exact equations of motion for the nuclear momentum, angular momentum, and the nuclear kinetic energy $\\tilde T_n$ obtained from the exact-factorization marginal wave function. Working from the nuclear Schrödinger equation supplied by the exact factorization of the electron–nuclear wave function, it shows that all three rates are governed by one effective electromagnetic force operator: $d\\mathbf P_\\mu/dt = \\mathrm{Re}\\langle\\chi|\\hat{\\bar{\\mathbf F}}_\\mu|\\chi\\rangle$, $d\\mathbf L_\\mu/dt = \\mathrm{Re}\\langle\\chi|\\mathbf R_\\mu\\times\\hat{\\bar{\\mathbf F}}_\\mu|\\chi\\rangle$, and $d\\tilde T_n/dt = \\mathrm{Re}\\langle\\chi|\\sum_\\mu \\hat{\\bar{\\mathbf F}}_\\mu\\cdot\\hat{\\mathbf v}_\\mu|\\chi\\rangle$. The force operator contains an electric part from the time-dependent scalar and vector potentials, a magnetic part from the intranuclear Berry curvature, and a purely quantum internuclear magnetic force with no classical analog. These identities give a compact, classical-looking language for quantifying how much energy and momentum electrons deposit into nuclei, which matters for electron–phonon relaxation, current-induced forces, and molecular-scale heat transfer. The paper verifies the identities in an exactly solvable two-nucleus model with nonzero internuclear Berry curvature.","feed_headline":"One force operator governs electron-nuclear transfer","feed_subtitle":"Exact identities reduce nuclear momentum, energy, and angular-momentum flow to a single electromagnetic force law.","key_machinery":"The load-bearing object is the effective electromagnetic force operator $\\hat{\\bar{\\mathbf F}}_\\mu$ built from the exact-factorization potentials. Its electric field $\\mathbf E_\\mu$ combines the scalar-potential gradient with the time derivative of the vector potential; its magnetic field is the intranuclear Berry curvature $C^{G'G}_{\\mu\\mu} = \\partial_{G'_\\mu}A_{G\\mu}-\\partial_{G\\mu}A_{G'_\\mu}$, which enters as $\\mathbf B_\\mu\\times\\hat{\\mathbf v}_\\mu$; and $\\hat{\\mathbf D}_\\mu$ collects the internuclear Berry curvatures $C^{G'G}_{\\nu\\mu}$ between different nuclei. Promoting the velocities in a trajectory-based force function to operators reproduces $\\hat{\\bar{\\mathbf F}}_\\mu$, so the identities connect the operator picture with the trajectory picture.","core_discovery":"The central discovery is that the exact nuclear Schrödinger equation of exact factorization, $i\\partial_t\\chi = [\\sum_\\mu (1/2M_\\mu)(-i\\nabla_{\\mathbf R_\\mu}+\\mathbf A_\\mu)^2+\\epsilon]\\chi$, already contains all electronic back-reaction in the form of an effective electromagnetic field. Replacing the full wave function with the marginal nuclear wave function $\\chi$ preserves the exact nuclear momentum and angular momentum, and their time derivatives equal expectation values of a single effective force operator $\\hat{\\bar{\\mathbf F}}_\\mu = \\mathbf E_\\mu + \\mathbf B_\\mu\\times\\hat{\\mathbf v}_\\mu + \\hat{\\mathbf D}_\\mu$, where $\\mathbf E_\\mu = \\partial_t\\mathbf A_\\mu - \\nabla_{\\mathbf R_\\mu}\\epsilon$, $\\mathbf B_\\mu$ is the intranuclear Berry curvature, and $\\hat{\\mathbf D}_\\mu$ is the internuclear Berry-curvature force. The same operator appears in the identities for momentum, angular momentum, and total nuclear kinetic energy; the magnetic parts do no net work. The internuclear term transfers momentum between individual nuclei but cancels in the total, so it has no classical analog.","pith_inferences":["In molecular transport or laser-driven dynamics, these identities could decompose electronic energy loss into per-nucleus electric and magnetic channels, separating ordinary Joule-like heating from geometric-phase forces.","The internuclear Berry-curvature term implies a testable prediction: momentum can appear to pass between two nuclei without any direct nuclear–nuclear force, mediated entirely by the electronic subsystem.","Because the paper's model has constant internuclear Berry curvature, an immediate next step is to test the identities in a model with spatially varying curvature, where the individual-nucleus kinetic-energy correction operator $\\hat G_\\mu$ no longer vanishes.","The same derivation should carry over to multicomponent mixtures or exciton–phonon systems, since only the TDSE form of the subsystem Hamiltonian is used."],"forward_implications":["The identities give a direct way to quantify energy transfer in electron–phonon systems, which the paper highlights as a primary application.","Nuclear momentum and angular momentum can be extracted exactly from the marginal nuclear wave function alone, without the full electron–nuclear wave function.","Intranuclear magnetic forces do no work, and the internuclear contributions cancel in the total nuclear kinetic energy while still redistributing momentum among individual nuclei.","Approximate factorizations based on adiabatic states yield the same inter-subsystem Ehrenfest identities for approximate quantities, extending the results beyond exact factorization.","The expectation value of the force operator can be evaluated by replacing it with the corresponding classical force function, linking these identities to trajectory-based nonadiabatic dynamics."],"supporting_citations":[{"why":"introduces the exact-factorization nuclear TDSE that is the starting point of all three identities.","marker":"[31]"},{"why":"establishes uniqueness and the exact density and current properties of the nuclear wave function chi.","marker":"[32]"},{"why":"shows the nuclear momentum computed from chi equals the full-system expectation value.","marker":"[34]"},{"why":"defines the geometric kinetic energy E_geo that separates tilde T_n from the true nuclear kinetic energy.","marker":"[35]"},{"why":"supplies the Berry-curvature concept used to identify the intranuclear magnetic field.","marker":"[33]"},{"why":"provides the trajectory-based force function whose operator promotion yields the effective force operator.","marker":"[37]"},{"why":"contains the detailed derivations of the angular-momentum IEI and the proof that internuclear magnetic forces do no net work.","marker":"[43]"}],"fun_headline_variants":["One force operator unifies all electron-nuclear transfer","Exact identities trace electron-nuclear flow to a single force","Berry curvature force orchestrates nuclear momentum and energy exchange","Single effective force law governs electron and nuclear exchange","Electron-nuclear transfer collapses into one effective electromagnetic operator"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivations assume the exact-factorization vector and scalar potentials are smooth enough that all boundary terms from integrations by parts vanish, which may fail where the nuclear wave function has nodes or the conditional electronic wave function becomes singular.","fun_headline_variants_meta":{"raw":{"variants":["One force operator unifies all electron-nuclear transfer","Exact identities trace electron-nuclear flow to a single force","Berry curvature force orchestrates nuclear momentum and energy exchange","Single effective force law governs electron and nuclear exchange","Electron-nuclear transfer collapses into one effective electromagnetic operator"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000247,"raw_usage":{"total_tokens":1546,"prompt_tokens":951,"completion_tokens":595,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":517}},"tokens_in":567,"tokens_out":595,"duration_ms":7391,"temperature":1.0,"reasoning_tokens":517,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:53:04.294759+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a two-nucleus model in which the nuclear wave packet $\\chi$ develops a nodal line while the vector potentials remain smooth, compute $d\\mathbf P_1/dt$ directly from the time-dependent $\\chi$, and compare it with $\\mathrm{Re}\\langle\\chi|\\hat{\\bar{\\mathbf F}}_1|\\chi\\rangle$; any nonzero difference would show that the discarded surface terms do not vanish and would refute Eq. (26) in that regime.","supporting_citations":[{"cited_title":"T.; Gross, E","cited_arxiv_id":null,"evidence_quote":"introduces the exact-factorization nuclear TDSE that is the starting point of all three identities."},{"cited_title":"T.; Gross, E","cited_arxiv_id":null,"evidence_quote":"establishes uniqueness and the exact density and current properties of the nuclear wave function chi."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"shows the nuclear momentum computed from chi equals the full-system expectation value."},{"cited_title":"K.; Maitra, N","cited_arxiv_id":null,"evidence_quote":"defines the geometric kinetic energy E_geo that separates tilde T_n from the true nuclear kinetic energy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the Berry-curvature concept used to identify the intranuclear magnetic field."},{"cited_title":"Nevertheless, the effect of ˆD1 is only of secondary importance in our model, and much less than the electromotive force E1","cited_arxiv_id":null,"evidence_quote":"contains the detailed derivations of the angular-momentum IEI and the proof that internuclear magnetic forces do no net work."}],"review_version":1}