{"id":"7cb81f90-162e-45e5-9092-81df7a521c50","arxiv_id":"2412.03418","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A time-dependent quasiparticle wavefunction is defined from the Green function and shown to obey a simple equation of motion, with a model that reproduces sodium and aluminum spectra.","lead":"The paper proposes a new way to define an electron's quasiparticle wavefunction directly from the many-body Green function, without building a self-energy first. The idea leads to a simpler equation of motion and a small model that reproduces measured and computed photoelectron spectra of sodium and aluminum, suggesting cheaper calculations for correlated materials.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quasiparticle equation of motion is conditional on the unproven existence of a time-local dynamical xc potential of the xc-hole form in Eq. (22); without this, Eqs. (30)-(35) have no general foundation.","rationale":"I agree with the reader's weakest assumption: Eq. (22) is the load-bearing input for the equation-of-motion part of the paper. The norm proof in Eq. (18) also uses an unjustified unitary diagonalization of a generally non-normal G_kk'(t), and the paper should repair it, but the bound itself is likely provable from the overlap form of G_kk', so I do not treat that as the primary obstacle. The Na and Al applications do not independently validate the formalism: the parameters Z, gamma, and eta are extracted from GW and then applied to the same kind of system, so they are illustrations rather than tests. An exact-diagonalization check of the assumed form and sum rule for Vxc would settle whether the EOM is a general theorem or only true under special assumptions. The reader's CONDITIONAL verdict remains appropriate; no change.","tokens_in":11417,"tokens_out":14333,"duration_ms":138362,"concrete_test":"Take the 1D Hubbard chain with 4 sites at half filling and U/t=4, compute the exact G(t) by full diagonalization, and at each t≠0 form Vxc(t)=([i∂t - h]G(t))G(t)^{-1} with a regularized inverse, using the Hubbard U as the interaction v. Then check whether Vxc(t) can be represented as in Eq. (24) for some xc hole satisfying the sum rule (26) for all times. If no such representation exists for this generic interacting system, the existence assumption behind Eqs. (30)-(35) fails; if it exists, the concern is answered affirmatively for that system.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Eq. (22) asserts that the exact Green function obeys [i∂t - h - Vxc]G = δ with Vxc(r,r';t) the Coulomb potential of an xc hole (Eq. (24)) and sum rule (26). This is not a consequence of the definition of G in Eq. (2): the exact G satisfies Dyson's equation with a self-energy that is convolutional in time, and reducing that to a first-order, time-local equation requires Vxc(t) to encode the full history of G for every pair (r,r') and every t. That is a strong existence statement, cited to refs. 3-7 but not proved or demonstrated for a general interacting system in this paper. The subsequent derivation is a chain of rewritings of this assumed equation: Eq. (30) follows only if Eq. (22) holds in the chosen orbital basis; Eq. (31) is a projection of that; and Ξ_q in Eq. (35) only deserves the name 'effective field' if Vxc exists with the stipulated xc-hole structure. If Eq. (22) is merely a formal definition of Vxc by inversion of G, the xc-hole representation and sum rule need separate proof. Until that existence is settled, the central result of Section III is conditional, not a derivation from the general definition of the Green function.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a new definition of a quasiparticle wavefunction directly from the time-dependent Green function, ψ*_k(r',t)=Σ_{k'} G_{kk'}(t) φ*_{k'}(r'), and argues that its total probability is bounded by unity and can decay for interacting systems. Within the authors' earlier dynamical exchange-correlation (xc) potential formalism, the paper derives a formally one-particle equation of motion for this object, with an effective q-dependent potential Ξ_q(r,t). Analytic results for the Hubbard dimer and a model for the homogeneous electron gas are presented, and the model is applied to Na and Al, with spectra compared to experiment and to cumulant-expansion and GW results. The central claim is that this yields a self-energy-free route to quasiparticle wavefunctions and spectra, with an effective potential that has a simple, approximately static-plus-plasmon form.","tokens_in":11790,"tokens_out":10492,"duration_ms":94444,"significance":"If the central claims hold, the paper offers an appealing new formulation of quasiparticles that bypasses the conventional self-energy eigenvalue equation and directly yields a time-dependent quasiparticle wavefunction with satellites. The analytic Hubbard-dimer solution is a concrete strength, and the HEG model reproduces the known cumulant-expansion satellite structure with parameters extracted from GW rather than fitted to experiment. The comparison with photoemission spectra for Na and Al is suggestive. However, the generality of the result is conditional on the existence of a time-local dynamical xc potential of the specific xc-hole form, and the proof of the norm bound contains a technical gap; these issues must be addressed before the paper can be accepted.","major_comments":[{"comment":"The proof of the norm bound rests on the assertion that G(t) is unitarily diagonalizable with eigenvalues bounded by 1. This is not justified: G(t) is not Hermitian or normal in general, and a nonnormal matrix need not be unitarily diagonalizable, nor are its eigenvalues bounded by 1 in modulus by the given argument. The conclusion of Eq. (20) is nevertheless correct and can be proved directly: for t<0, the row vector (G_{kk'})_{k'} is the projection of the state c_k(t)|Ψ0⟩ onto the subspace spanned by {c_{k'}|Ψ0⟩}, whose norm is at most ||c_k|Ψ0⟩|| ≤ 1 by the Pauli principle. The authors should replace the diagonalization argument in Sec. II and in Appendix A with a valid proof or otherwise provide a correct justification.","section":"Section II, Eqs. (18)-(20) and Appendix A"},{"comment":"The quasiparticle equation of motion, Eqs. (31)-(36), is derived from the assumed existence of a time-local dynamical xc potential Vxc(r,r';t) satisfying Eq. (22) with the xc-hole representation of Eq. (24) and the sum rule of Eq. (26). This is a strong existence assumption imported from earlier work; the exact Green function generally satisfies Dyson's equation with a self-energy that is convolutional in time, and reducing it to the first-order time-local form of Eq. (22) is nontrivial. The manuscript does not prove or even state explicitly that this reduction is a postulate rather than a consequence of the definition of G. The abstract's claim that the quasiparticle wavefunction and its equation of motion are 'derived from the general definition of the Green function without reference to self-energy' is therefore too strong. The authors should either supply a derivation or precise conditions for Eq. (22), or clearly demarcate it as a foundational assumption of their dynamical xc formalism and qualify the abstract and conclusions accordingly.","section":"Section III, Eq. (22)"},{"comment":"The derivation of the quasiparticle equation of motion is obscured by the notation in Eq. (30). In Eq. (22), Vxc(r,r';t) acts as a nonlocal integral kernel on the first coordinate of G(r,r';t), but Eq. (30) writes ∆V(r,r';t) as a multiplicative factor on the product φ_k(r)ψ*_k(r',t). The projection leading to Eq. (31) and the definition of ∆V_{qk}(r,t) in Eq. (32) are consequently not transparent. The authors should present the derivation either in the orbital-basis form of Eq. (27) (which is used in the Hubbard dimer) or with the correct integral-kernel notation, because the effective potential Ξ_q in Eq. (35) is the central object of the paper and its definition must be unambiguous.","section":"Section III, Eqs. (30)-(32)"}],"minor_comments":[{"comment":"The expansion of G_q(t) into A0, A1, A2 is a truncation of the full exponential series e^{λ(e^{iω_p t}-1)} to second order in λ, not an exact result. This should be stated explicitly, along with the expected accuracy of the truncation for the parameters used.","section":"Section IV.B, Eqs. (75)-(79)"},{"comment":"The formula λ = 1 - √(2Z-1) is real only for Z ≥ 1/2. The authors should note this domain of validity, since the plotted Z values are in this range but the condition is not stated.","section":"Section IV.B, Eq. (80)"},{"comment":"The quasiparticle wavefunction ψ*_k depends on the choice of the one-particle basis {φ_k}. The paper acknowledges this, but the practical consequences — e.g., how the norm bound and the effective potential transform under a change of basis — deserve a short discussion, since the wavefunction is not a gauge-invariant object.","section":"Section II and Section VII"},{"comment":"The figure captions are too terse for a self-contained reading. For example, Fig. 4 does not state which parameters (rs, Z, γ) are used for the Na and Al curves, and Figs. 5-8 would benefit from a statement of the model parameters and the GW comparison details.","section":"Figure captions"}],"recommendation":"major_revision","confidential_remarks":"The paper relies heavily on the authors' prior dynamical xc formalism (refs. 3-7). The editor may wish to ensure that the existence of the time-local Vxc in Eq. (22) is accepted in the community; if it is contested, the present manuscript does not settle the question. The norm-bound proof is readily fixable, but the status of Eq. (22) is a more substantive point that should be addressed explicitly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The quasiparticle wavefunction defined by psi*_k = sum_k' G_kk' phi*_k' is a genuinely clean idea: it comes directly from the Green function, reduces to the orbital in the noninteracting limit, and naturally carries decay and satellite structure. The equation of motion in Section III is a coherent consequence of the dynamical xc potential formalism, but the load-bearing assumption is Eq. (22), which asserts a first-order, time-local equation for the exact Green function with Vxc of the xc-hole form. That assumption is imported from the authors' earlier work (refs. 3-7) and is not proved here. So the central result is conditional on that formalism.\n\nWhat the paper does well: the Hubbard dimer is fully analytic and shows Xi is piecewise constant, in stark contrast to the frequency-dependent self-energy. The electron-gas model is simple and reproduces the qualitative features of Na and Al spectra, including plasmon satellites at integer multiples of omega_p. The authors are explicit that the parameters come from one-shot GW, so these are demonstrations, not blind tests. The paper is honest about that.\n\nThe soft spots are two. First, the norm-bound proof in Sec. II diagonalizes G(t) with a unitary matrix, but G is not in general normal, so that step is unjustified. The bound itself may be true (any contraction matrix has row norms no larger than 1), but the proof needs repair. Second, the derivation of Eqs. (30)-(35) depends entirely on Eq. (22). If that equation is just a formal inversion defining Vxc, then the xc-hole representation and sum rule need separate proof. The paper should either prove existence in a nontrivial case or state more carefully that it is an assumption of the dynamical xc formalism.\n\nThe heavy self-citation is expected here because the formalism is the authors' own; the issue is not self-citation, it's that the key existence question is not revisited. The new definition of the quasiparticle wavefunction is independent of that question and stands on its own.\n\nThis paper deserves a serious referee. The flaws are addressable: fix the norm proof, clarify the status of Eq. (22), and add at least one non-fitted test of the model. I would recommend conditional acceptance after such revisions. If the authors can strengthen the proof of the norm bound and acknowledge the conditional nature of the EOM, it will be a useful contribution to the electronic structure literature.","headline":"A clean, useful redefinition of the quasiparticle wavefunction, but the equation of motion rests on an unproven time-local Vxc and the norm-bound proof has a fixable gap.","tokens_in":12262,"tokens_out":5852,"would_cite":true,"duration_ms":53668,"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":"Quasiparticles get a self-energy-free wavefunction that decays in time","keywords":["quasiparticle","Green function","dynamical exchange-correlation potential","quasiparticle wavefunction","plasmon satellites","Hubbard dimer","homogeneous electron gas"],"falsifier":"Compute the exact Green function of a small correlated system (e.g., a Hubbard dimer or a small molecule in full configuration interaction) and check whether there exists a Hermitian or even non-Hermitian local-in-time potential $V_{xc}(r,r';t)$ satisfying Eq. (22) exactly at all times and all positions. If the residual of $[i\\partial_t-h-V_{xc}]G-\\delta$ cannot be made zero for any choice of $V_{xc}$ of this form, then Eqs. (30)-(35) and the interpretation of $\\Xi_q$ as an effective field do not follow.","tokens_in":11186,"feed_emoji":"⚛️","tokens_out":1933,"duration_ms":20608,"temperature":0.7,"pith_summary":"This paper proposes a new definition of a quasiparticle wavefunction that comes directly from the Green function, without the usual self-energy eigenvalue equation. The wavefunction is constructed by contracting the Green function onto a chosen set of one-particle orbitals, so it carries not only the main quasiparticle peak but also satellite modes from coupling to collective excitations like plasmons. In an interacting system this object has total probability less than one, so it naturally decays in time. Within the dynamical exchange-correlation potential formalism, it obeys a single-particle equation of motion with an effective q-dependent potential, giving a route to quasiparticle spectra that avoids computing the self-energy. The authors test the idea on the Hubbard dimer and on electron-gas-like metals, where a simple model reproduces the main features of Na and Al photoemission spectra.","feed_headline":"A self-energy-free quasiparticle wavefunction that decays in time","feed_subtitle":"Green-function contraction yields one-body equations with effective potentials and reproduces Na and Al photoemission.","key_machinery":"The key object is the time-dependent quasiparticle wavefunction $\\psi^*_k(r',t)=\\sum_{k'}G_{kk'}(t)\\phi^*_{k'}(r')$, obtained by expanding the field operators in a complete one-particle basis and contracting the Green function onto that basis. The proof of decay rests on the inequality $\\sum_{k'}|G_{kk'}(t)|^2\\le1$, derived by diagonalizing the matrix $G_{kk'}(t)$ with a unitary transformation and using the Pauli-principle bound $\\langle\\Psi_0|\\hat c_k^\\dagger\\hat c_k|\\Psi_0\\rangle\\le1$ together with unitarity of time evolution. The effective potential $\\Xi_q(r,t)=[\\psi^*_q(r,t)]^{-1}\\sum_k\\Delta V_{qk}(r,t)\\psi^*_k(r,t)$ then converts the coupled dynamical-xc equations of motion into a single one-body equation per quasiparticle, where $\\Delta V_{qk}$ couples different quasiparticle channels through the dynamical xc potential.","core_discovery":"The paper's central claim is that the object defined by $\\psi^*_k(r',t)=\\sum_{k'}G_{kk'}(t)\\phi^*_{k'}(r')$ is a legitimate quasiparticle wavefunction. Because the matrix elements of the Green function in any orthonormal orbital basis satisfy $\\sum_{k'}|G_{kk'}(t)|^2\\le1$, the integrated density $\\int dr|\\psi_k(r,t)|^2\\le1$, with equality only in noninteracting systems. This object therefore describes a quasiparticle that decays in time and contains the full excitation structure of the Green function rather than a single eigenmode. When the Green function obeys the dynamical exchange-correlation equation of motion $[i\\partial_t-h-V_{xc}]G=\\delta$, the quasiparticle wavefunction satisfies a one-body equation $[i\\partial_t-\\varepsilon_q-\\Xi_q(r,t)]\\psi^*_q(r,t)=0$, where $\\Xi_q$ is an effective q-dependent, generally non-Hermitian potential built from dynamical $V_{xc}$ matrix elements. The decay follows from the imaginary part of $\\Xi_q$, which is absent in the traditional self-energy eigenvalue equation. In the electron gas this reproduces the main quasiparticle peak plus plasmon satellites whose oscillator strengths match the cumulant expansion.","pith_inferences":["If the existence of the dynamical xc potential in the closed form of Eq. (22) holds beyond the tested models, the effective-potential route could be extended to nonequilibrium and finite-temperature dynamics, since the wavefunction definition itself is carried to those regimes in Appendix A.","The inequality $\\int dr|\\psi_k|^2\\le1$ suggests a natural diagnostic for approximate Green functions: any approximation that violates this bound cannot correspond to a fermionic many-body system, so the condition could serve as a consistency check on new approximations.","The model's separation of $\\Xi_q$ into a static renormalizing part and a dynamical plasmon part hints that the same structure might extend to systems with phonons or magnons, where the collective frequency would replace the plasmon frequency in the satellite pattern.","A testable extension is to compute $V_{xc}$ for real materials from first principles within the dynamical xc formalism and compare the resulting $\\Xi_q$, spectra, and lifetimes against the present parametrized model, which would separate model errors from errors of the underlying formalism."],"forward_implications":["Quasiparticle wavefunctions and spectra can in principle be computed without constructing the self-energy, from the dynamical xc potential $V_{xc}$ alone.","The time-decay and satellite structure of a quasiparticle are built into the wavefunction itself, so satellite weights and quasiparticle renormalization are obtained at the wavefunction level rather than from poles of a self-energy expression.","The equation of motion is local in space and time for each quasiparticle channel, which suggests practical implementations can treat a subspace of relevant quantum numbers (e.g., a limited set of band indices) without solving the full problem.","If the simple form of $\\Xi_q$ found in the Hubbard dimer and the electron gas holds generally, quasiparticle spectra of metals become expressible via a constant energy shift plus terms oscillating at collective-excitation frequencies.","A local-density approximation based on the model potential, with density-dependent band narrowing and lifetime parameters, is suggested as an implementable next step for realistic materials.","The proposed wavefunction is observable in principle: its time decay directly implies a finite quasiparticle lifetime, and its satellite modes correspond to the incoherent features seen in angle-resolved photoemission spectra."],"supporting_citations":[{"why":"Supplies the standard Green function definition and the framework of Landau quasiparticles that the new definition is contrasted with.","marker":"1"},{"why":"Provides the traditional quasiparticle eigenvalue equation with self-energy that this paper proposes to replace.","marker":"2"},{"why":"Gives the dynamical exchange-correlation potential equation of motion that the quasiparticle equation of motion is derived from in this paper.","marker":"3"},{"why":"Supplies the dynamical xc potential result for the interacting Hubbard dimer, used as one of the two central examples.","marker":"4"},{"why":"Gives the dynamical xc potential formalism for the homogeneous electron gas, used to derive the model for metals.","marker":"5"},{"why":"Introduces the earlier definition of the quasiparticle wavefunction from the Green function that this paper develops further.","marker":"6"},{"why":"Reports the analogous simple form of the effective potential in the one-dimensional Heisenberg spin model, used as supporting evidence for the generic form.","marker":"7"},{"why":"Provides the cumulant expansion results for the homogeneous electron gas that the model spectral functions of Na and Al are compared with.","marker":"17"},{"why":"Gives the one-shot GW method used to extract the parameters $\\gamma$ and $Z$ for the model.","marker":"20"},{"why":"Provides the measured Na photoemission spectrum that the model's total spectral function is compared with.","marker":"21"}],"fun_headline_variants":["Quasiparticle wavefunction that decays in time, no self-energy","Green function yields decaying quasiparticle wavefunction","Decaying quasiparticle wavefunction without self-energy","From Green function to decaying quasiparticle states"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire derivation of the quasiparticle equation of motion presupposes that an interacting many-electron system has a dynamical exchange-correlation potential $V_{xc}$ such that the exact Green function obeys $[i\\partial_t-h-V_{xc}]G=\\delta$ with $V_{xc}$ written as the Coulomb potential of an xc hole.","fun_headline_variants_meta":{"raw":{"variants":["Quasiparticle wavefunction that decays in time, no self-energy","Green function yields decaying quasiparticle wavefunction","Decaying quasiparticle wavefunction without self-energy","From Green function to decaying quasiparticle states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001096,"raw_usage":{"total_tokens":4583,"prompt_tokens":962,"completion_tokens":3621,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":3556}},"tokens_in":578,"tokens_out":3621,"duration_ms":25087,"temperature":1.0,"reasoning_tokens":3556,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:25:49.493462+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the exact Green function of a small correlated system (e.g., a Hubbard dimer or a small molecule in full configuration interaction) and check whether there exists a Hermitian or even non-Hermitian local-in-time potential $V_{xc}(r,r';t)$ satisfying Eq. (22) exactly at all times and all positions. If the residual of $[i\\partial_t-h-V_{xc}]G-\\delta$ cannot be made zero for any choice of $V_{xc}$ of this form, then Eqs. (30)-(35) and the interpretation of $\\Xi_q$ as an effective field do not follow.","supporting_citations":[{"cited_title":"L Fetter and J","cited_arxiv_id":null,"evidence_quote":"Supplies the standard Green function definition and the framework of Landau quasiparticles that the new definition is contrasted with."},{"cited_title":"Hedin and S","cited_arxiv_id":null,"evidence_quote":"Provides the traditional quasiparticle eigenvalue equation with self-energy that this paper proposes to replace."},{"cited_title":"Aryasetiawan, Phys","cited_arxiv_id":null,"evidence_quote":"Gives the dynamical exchange-correlation potential equation of motion that the quasiparticle equation of motion is derived from in this paper."},{"cited_title":"Aryasetiawan and T","cited_arxiv_id":null,"evidence_quote":"Supplies the dynamical xc potential result for the interacting Hubbard dimer, used as one of the two central examples."},{"cited_title":"Karlsson and F","cited_arxiv_id":null,"evidence_quote":"Gives the dynamical xc potential formalism for the homogeneous electron gas, used to derive the model for metals."},{"cited_title":"Aryasetiawan, Phys","cited_arxiv_id":null,"evidence_quote":"Introduces the earlier definition of the quasiparticle wavefunction from the Green function that this paper develops further."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the analogous simple form of the effective potential in the one-dimensional Heisenberg spin model, used as supporting evidence for the generic form."},{"cited_title":"Almbladh and L","cited_arxiv_id":null,"evidence_quote":"Provides the cumulant expansion results for the homogeneous electron gas that the model spectral functions of Na and Al are compared with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the one-shot GW method used to extract the parameters $\\gamma$ and $Z$ for the model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the measured Na photoemission spectrum that the model's total spectral function is compared with."}],"review_version":1}