{"id":"d347cb51-7326-4c8c-892c-c095df0a57d8","arxiv_id":"2607.18183","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"In monolayer WSe₂, excitonic correlations make photoexcited carriers relax toward the Q valleys and into exciton-wavefunction-shaped distributions, overturning the single-particle Fermi–Dirac thermalization picture and matching time-resolved ARPES.","lead":"Carrier dynamics in photoexcited WSe₂ are computed with a two-particle framework that tracks electron–hole pairs (excitons) alongside phonon scattering, instead of treating electrons and holes independently. The model predicts carriers pile up in the six Q valleys rather than the K valleys, matching time-resolved ARPES — and that the relaxed state is shaped by exciton wavefunctions, not Fermi–Dirac statistics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"XBE quasi-equilibrium claim lacks convergence/detailed-balance check; the 1.3–1.75 ps distributions may be transient, so the exciton-wavefunction-shaped 'thermalized' state is not yet established.","rationale":"I focused on convergence to the BE fixed point because the strongest claim is explicitly about the 'thermalized' carrier momentum distribution. The paper supplies a mechanism (bound-exciton channels) and a plausible fixed point (BE occupations of e-h eigenstates), but the numerical simulation is only propagated to 1.3/1.75 ps. The authors themselves describe the occupations as slowly evolving at 1.3 ps, admitting that the asymptotic state is not reached. Without a check against the exact fixed point or detailed-balance verification, the displayed distribution could be a nonuniversal transient. This is a correctness risk, not a stylistic one. I agree with the reader's weakest_assumption: the same convergence/fixed-point issue was identified. I do not see a more fundamental flaw: the XBE→SEPE reduction under the noninteracting-continuum approximation is a useful consistency check, and the qualitative Q-over-K prediction is benchmarked against a published experiment. The ARPES extraction concern is real but secondary; even if the ratio were 0.3, the theoretical prediction could still be a transient. Therefore the appropriate verdict remains CONDITIONAL (reader's verdict unchanged).","tokens_in":20658,"tokens_out":6773,"duration_ms":118354,"concrete_test":"Propagate the sudden-excitation XBE to at least 10 ps (or until dN^{λQ}/dt < 1% of its initial value for all populated states) and overlay the resulting f_c(E) at t=10 ps with the exact fixed point f_fix(E) computed from Eq. (3) using N^{λQ}_BE = [exp((E^{λQ}-μ)/k_BT)-1]^{-1}, with μ fixed by the conserved pair density. If the simulated distribution converges to f_fix, the central claim stands; if it does not, the 'correlated quasi-equilibrium' is a transient artifact and the verdict should be conditional with the convergence check as a prerequisite.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that thermalized carrier distributions are shaped by exciton wavefunctions rather than Fermi–Dirac statistics—requires that the XBE occupations N^{λQ}(t) have reached (or tightly approached) the Bose–Einstein fixed point at T=70 K. The paper asserts this after Eq. (3) ('the occupations N^{λQ} thermalize according to a Bose–Einstein distribution...') and again in the sudden-excitation section, but the only numerical evidence is a quasi-stationary snapshot at 1.3 ps (sudden) or 1.75 ps (pump), and the text itself says the occupations are 'evolving slowly toward their asymptotic values' at 1.3 ps. No convergence study, no comparison with the exact BE-projected fixed point f_fix(E) = ν^{-1}(E) Σ_{λQ,cv} [exp((E^{λQ}-μ)/k_BT)-1]^{-1} |A^{λQ}_{cvk}|^2, and no check that the Markovian rates in Eq. (5) satisfy detailed balance on the numerical k-grid. If the integration is still far from the fixed point, the exciton-wavefunction-like shape and the K-to-Q ratio ~0.3 could be transient relaxation features; then the strong claim that the long-time state 'cannot be captured by any non-interacting thermal fermion description' is unsupported. The ARPES comparison is also based on an undocumented post-processing of Ref. 52, so it does not resolve the convergence question.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces an excitonic Bloch equations (XBE) framework, based on the authors' prior work (Ref. 51), to describe phonon-driven carrier dynamics in photoexcited WSe2 monolayers. The central claims are: (i) XBE reduce to standard single-particle electron–phonon equations (SEPE) when bound exciton states are neglected, providing a consistency check; (ii) for non-resonant excitation at low density, excitonic correlations cause rapid intervalley scattering that populates Q valleys over K valleys, in agreement with time-resolved ARPES; and (iii) the long-time carrier distributions are shaped by exciton wavefunctions rather than Fermi–Dirac statistics, signaling a correlated quasi-equilibrium. The paper compares sudden-excitation and finite-pulse protocols, reporting a K-to-Q ratio of about 0.3 and nearly identical final distributions in both cases.","tokens_in":20946,"tokens_out":5211,"duration_ms":55996,"significance":"If the central claims hold, the paper challenges the standard single-particle relaxation picture for excitonic semiconductors, identifying exciton formation as a dominant relaxation channel even at low density and explaining the experimentally observed Q-valley population. The analytical reduction of XBE to SEPE when bound states are discarded is a genuine consistency check, and the qualitative mechanism — near-degenerate K/Q excitons combined with sixfold Q multiplicity — is clearly argued and plausible. The use of two excitation protocols (sudden and pump) and the explicit treatment of coherent and incoherent populations are additional strengths. However, the quantitative claims of thermalization and ARPES agreement require stronger numerical and methodological support. The paper is potentially significant but not yet fully convincing.","major_comments":[{"comment":"The central asymptotic claim is asserted but not demonstrated. After 'approximately 1 ps' the text says occupations are only 'evolving slowly toward their asymptotic values', yet Fig. 3f at 1.3 ps is presented as the steady state. No convergence study is shown, no detailed-balance check of the Markovian rates in Eq. (5) is provided, and no comparison is made with the Bose–Einstein fixed point at energies E^{λQ}. Until the integration is shown to have reached (or tightly approached) that fixed point, the exciton-wavefunction shape and the K-to-Q ratio ~0.3 could be transient relaxation features. I request a time-convergence analysis and a detailed-balance or fixed-point validation.","section":"§3 (sudden excitation); Eqs. (3)–(5), Fig. 3e–f"},{"comment":"The stated quantitative agreement with time-resolved ARPES rests on an undocumented post-processing of Ref. 52. The paper reports a K-to-Q ratio of ~0.3 without specifying how populations were extracted from the experimental spectra — e.g., energy/momentum integration windows, background subtraction, valley assignment, spin/degeneracy factors, or experimental error bars. Without this information the 'quantitative agreement' claim cannot be evaluated. The authors should either provide the extraction procedure in detail or soften the claim to qualitative consistency.","section":"§4 (pump excitation); ARPES comparison"},{"comment":"The statement that 'the occupations N^{λQ} thermalize according to a Bose–Einstein distribution evaluated at the e–h energies E^{λQ}' is inherited from Ref. 51, but the conditions under which Eq. (5) has this fixed point — e.g., detailed balance of the Γ rates and conservation of total pair number — are not stated or verified. Since this fixed-point property is the basis for the claim that f^c_k and f^v_k take exciton-wavefunction shapes, it should be made explicit and checked numerically for the actual rates used.","section":"§2, after Eq. (3)"}],"minor_comments":[{"comment":"The momentum argument in f^c_k uses A^{λQ}_{cvk-Q} while f^v_k uses A^{λQ}_{cvk}. Please clarify the convention, since this notation is potentially confusing.","section":"Eq. (3)"},{"comment":"Typo: 'intead' should be 'instead'.","section":"§4, pump excitation"},{"comment":"The citation appears as 'Ref.,52'; should be 'Ref. [52]'.","section":"§4, first paragraph"},{"comment":"The notation |Ψ^A_{e/h}(E)|² is used but not defined. Define it in the caption or main text.","section":"Fig. 3e caption"},{"comment":"The XBE-to-SEPE reduction is a central result and is only cited to a Supporting Note. Please ensure this derivation is fully available and cross-referenced, as it underlies the interpretation of XBE–SEPE differences as bound-state effects.","section":"Supporting Note 3"},{"comment":"The sentence 'the lowest-energy excitons are all dark' should be reconciled with the discussion of bright states in Eq. (4); a brief explanation of why dark excitons dominate the thermalized population would help the reader.","section":"Conclusions"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents an appealing and potentially important framework, with a useful consistency check in the XBE→SEPE limit. The main risk is that the 'thermalized' state is claimed before demonstrating approach to the fixed point; adding convergence and detailed-balance checks would substantially strengthen the paper. The ARPES post-processing also needs to be transparent. These issues are fixable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper is a serious candidate for the claim that excitonic correlations, not single-particle electron-phonon scattering, control the long-time carrier distribution in photoexcited WSe2. The mechanism is clear: once carriers reach the band edges, bound excitons dominate, and because K and Q valley excitons are nearly degenerate while Q has sixfold multiplicity, XBE predict a Q-over-K population ratio around 3, matching the tr-ARPES measurement of Madéo et al. That is a nontrivial result, and the limiting reduction of XBE to SEPE when bound states are omitted gives a genuine internal consistency check. The early transient agreement between XBE and SEPE, followed by divergence as bound electrons form, is a sensible narrative and the paper reports it fairly.\n\nWhere the paper is soft is not in the mechanism but in the evidence for the steady state. The claim that occupations thermalize to a 70 K Bose-Einstein distribution is inherited from an earlier paper by the same group (Ref 51), not re-derived or tested here. The text itself says the distributions are \"evolving slowly toward their asymptotic values\" at 1.3 ps, which is not the same as being at the fixed point. A snapshot at 1.3 or 1.75 ps can look exciton-wavefunction-like and still be a transient. The authors need a convergence study, a check that the Markovian rates satisfy detailed balance on their k-grid, or a longer-time calculation; otherwise the headline that the long-time state \"cannot be captured by non-interacting thermal fermions\" is not fully supported.\n\nThe quantitative ARPES comparison is also less solid than claimed. The K-to-Q ratio of ~0.3 is obtained after a post-processing of Ref 52 that is not described in the main text or in the available supporting material, and no error analysis is given. That makes the match suggestive rather than quantitative.\n\nOne practical issue: the Supporting Notes 1-4, which contain the rate derivations, the XBE-to-SEPE proof, and the simulation parameters, are referenced but not included in the version I saw. A referee cannot verify key computational details. That may be an artifact of the submission package, but it has to be fixed.\n\nNo sign of fabrication or invented entities; the free parameters are standard. This paper is for theorists and experimentalists working on ultrafast dynamics in TMDs and excitonic semiconductors. It deserves a serious referee, but the referee should insist on the convergence evidence and the ARPES extraction description. I would not cite the quantitative claim until those are in place, but I would bring the paper to a reading group.","headline":"The mechanism is plausible and the SEPE reduction is a real internal check, but the thermalized-exciton claim rests on a 1.3 ps snapshot without convergence evidence; the paper deserves review but needs supporting details and a cleaner ARPES comparison.","tokens_in":21543,"tokens_out":3267,"would_cite":false,"duration_ms":30897,"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":"Excitonic correlations, not free-carrier physics, determine the final carrier distribution in photoexcited WSe2.","keywords":["excitonic Bloch equations","WSe2 monolayer","carrier relaxation","intervalley scattering","exciton-phonon coupling","time-resolved ARPES","thermalization","Bethe-Salpeter equation"],"falsifier":"Time-resolved ARPES at delays beyond a few picoseconds showing the K/Q valley ratio returning toward the single-particle prediction (K over Q), or a numerical test demonstrating that the XBE occupations depend on the initial excitation conditions instead of converging to the same Bose–Einstein distribution, would falsify the claim of an excitonic quasi-equilibrium steady state.","tokens_in":20426,"feed_emoji":"⚛️","tokens_out":4762,"duration_ms":43454,"temperature":0.7,"pith_summary":"This paper argues that when electrons and holes are created in a semiconductor, treating them as independent quasiparticles misses the dominant relaxation channel. Using a two-particle framework built on excitonic Bloch equations, the authors compute the coupled evolution of electron-hole pair states in monolayer WSe2, including phonon scattering and exciton formation. They find that carriers quickly convert into bound excitons, which scatter strongly between valleys and settle into a correlated quasi-equilibrium whose momentum distribution is shaped by exciton wavefunctions rather than by Fermi–Dirac statistics. This explains why experimentally the Q valleys host more carriers than the K valleys, whereas single-particle Boltzmann-type simulations predict the opposite. The result implies that standard single-particle relaxation frameworks are inadequate for excitonic materials even at low excitation densities.","feed_headline":"Exciton wavefunctions, not Fermi-Dirac, shape WSe2 carrier distributions","feed_subtitle":"A two-particle theory reproduces time-resolved ARPES where single-particle Boltzmann models fail—relaxation is excitonic from the start.","key_machinery":"The excitonic Bloch equations (XBE): a Markovian set of equations for the occupations of all electron-hole eigenstates of the finite-momentum Bethe–Salpeter equation, including both bound excitons and unbound pairs. The key ingredients are T-matrix vertex corrections to the Fan–Migdal electron-phonon self-energy, a decomposition of occupations into coherent (polarization) and incoherent parts, and auxiliary irreducible electron-hole occupations that prevent overscreening. The bridge to observable carrier distributions is the projection formula f_ck = sum_{λQv} N^{λQ} |A^{λQ}_{cvk}|^2, which maps the bosonic exciton occupations onto fermionic single-particle distributions.","core_discovery":"The central claim is that the long-time state of a photoexcited excitonic semiconductor is not a thermalized gas of independent electrons and holes. Within the excitonic Bloch equations, the occupation numbers of electron-hole eigenstates of the finite-momentum Bethe–Salpeter equation thermalize to a Bose–Einstein distribution at the lattice temperature; projecting these occupations onto single-particle states yields distributions that inherit the momentum-space structure of the lowest-energy exciton wavefunction. The paper further shows that when bound exciton states are neglected, the equations reduce exactly to conventional semiconductor electron-phonon (Boltzmann) equations, establishing","pith_inferences":["If the thermalized state is indeed a Bose–Einstein distribution of dark excitons, then valley and momentum-resolved photoemission at late delays is effectively imaging the exciton wavefunction, suggesting a general spectroscopy of exciton structure in momentum space.","The paper's finding that the Q vs K imbalance is driven by phase-space multiplicity (six Q valleys vs two K) rather than band energies implies that valleytronic devices based on TMDs must account for excitonic scattering channels, not only single-particle phonon scattering.","The claimed independence of the final state from the excitation protocol (sudden vs pump) is a testable prediction: experiments varying pump photon energy and duration should still converge to the same non-thermal steady state within a few picoseconds."],"forward_implications":["Single-particle Boltzmann and semiconductor Bloch simulations of 2D semiconductors can qualitatively mispredict valley populations and the direction of intervalley transfer.","Long-time carrier distributions in excitonic materials cannot be fit by a Fermi–Dirac function at any temperature, so analyses assigning 'effective carrier temperatures' to such data are misleading.","Exciton formation begins reshaping the dynamics within a few hundred femtoseconds, i.e., during a typical pump pulse, not only at late times.","The XBE framework provides a route to directly compute momentum-resolved carrier populations that can be compared to time-resolved ARPES without ad-hoc thermal models."],"fun_headline_variants":["Excitonic correlations rewrite photocarrier relaxation rules","WSe2 carriers: exciton shape trumps Fermi-Dirac","Excitons dictate hot-carrier cooling in WSe2","Two-particle theory beats single-particle for WSe2 dynamics","Excitonic wave functions control carrier spread in WSe2"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The Markovian scattering rates in the excitonic Bloch equations drive the exciton occupations to a Bose–Einstein distribution at the fixed lattice temperature within the simulated 1–1.75 ps window, so the computed distributions are the true asymptotic state rather than a slowly evolving transient.","fun_headline_variants_meta":{"raw":{"variants":["Excitonic correlations rewrite photocarrier relaxation rules","WSe2 carriers: exciton shape trumps Fermi-Dirac","Excitons dictate hot-carrier cooling in WSe2","Two-particle theory beats single-particle for WSe2 dynamics","Excitonic wave functions control carrier spread in WSe2"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000209,"raw_usage":{"total_tokens":1226,"prompt_tokens":709,"completion_tokens":517,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":453,"completion_tokens_details":{"reasoning_tokens":446}},"tokens_in":453,"tokens_out":517,"duration_ms":29062,"temperature":1.0,"reasoning_tokens":446,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T15:46:24.047187+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Time-resolved ARPES at delays beyond a few picoseconds showing the K/Q valley ratio returning toward the single-particle prediction (K over Q), or a numerical test demonstrating that the XBE occupations depend on the initial excitation conditions instead of converging to the same Bose–Einstein distribution, would falsify the claim of an excitonic quasi-equilibrium steady state.","supporting_citations":[],"review_version":1}