{"id":"193bb484-5373-48e8-879c-754b95038a89","arxiv_id":"2607.10211","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Self-consistent spectral dressing of the three-phonon bubble broadens HgTe phonons enough to suppress lattice thermal conductivity fivefold to experiment without explicit higher-order scattering.","lead":"A self-consistent way of treating phonons as broad frequency distributions, not sharp quasiparticles, cuts the calculated heat conductivity of HgTe by about five times to match experiment. It does so using only three-phonon processes and is checked against molecular-dynamics spectra.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"Transport formula may not fully capture conductivity once modes are strongly non-quasiparticle.","rationale":"The reader correctly isolates the modeling assumption that residual higher-order diagrams and off-diagonal transport remain negligible once the cubic bubble is self-consistently dressed. That assumption is load-bearing for the strongest claim (factor-of-five reduction to experiment without explicit higher-order interactions). The spectral-function and MD comparisons are solid for the self-energy itself, and the elimination of external smearing is a genuine methodological advance; the soft spot is only the transport step that converts those spectra into κ_l. Because the paper already flags the <5 % coherence estimate and the concurrent literature, the appropriate stance remains CONDITIONAL rather than REJECT. A single targeted recalculation of the full conductivity (or an explicit four-phonon estimate on the broadened spectra) would settle whether the concern lands.","tokens_in":11491,"tokens_out":558,"duration_ms":5159,"concrete_test":"Recompute κ_l at 300 K with the full Wigner/Green’s-function conductivity (including off-diagonal terms) using the converged SCSF self-energies, and/or evaluate the four-phonon scattering rates on the same broadened spectral densities; if either contribution exceeds ~15–20 % of the reported diagonal SCSF value, the claim that cubic self-consistency alone accounts for the experimental κ_l is weakened.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that self-consistent dressing of the cubic bubble alone reduces κ_l by ~5× to experiment rests on evaluating the Peierls-Boltzmann expression (Eq. 5) with the effective lifetime of Eq. 6 extracted from the diagonal spectral function A_λ(ω). After convergence, many modes (especially TO near Γ and LA/TA near zone boundary) become strongly non-Lorentzian with large γ/ω (Figs. 1e–f, 3). In that regime the diagonal spectral-function lifetime is only an approximation to the full many-body heat current; the paper asserts off-diagonal coherence is <5 % but does not recompute it with the converged, broadened spectra, nor does it compare against a Green–Kubo or Wigner evaluation that uses the same self-energy. If residual four-phonon or coherence channels remain sizable once the spectral functions are broad, the attribution of the entire experimental match to cubic self-consistency alone is overstated.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript introduces a self-consistent spectral function (SCSF) framework for anharmonic phonons: each mode is represented by its full spectral density A_λ(ω), which is used to dress the internal lines of the cubic bubble self-energy (Eqs. 1–4). Real and imaginary parts of Σ_λ(ω) are updated on equal footing via Kramers–Kronig, eliminating external energy-conserving smearing. Applied to zincblende HgTe, five iterations substantially broaden acoustic and optical spectra, open scattering channels absent in one-shot quasiparticle calculations, and reduce κ_l by roughly a factor of five to the experimental scale with a steeper T dependence (≈T^−1.68). Mode-resolved SCSF spectra at the L point and along Γ–L are compared to power spectra from MD with a machine-learning potential. The authors position SCSF as a bridge between quasiparticle perturbation theory and fully dynamical simulations, generalizable to higher-order processes.","tokens_in":11739,"tokens_out":1693,"duration_ms":20272,"significance":"If the central results hold, the work offers a practical, first-principles route to go beyond the quasiparticle approximation for phonon linewidths and thermal transport without introducing free smearing parameters or explicitly computing four-phonon diagrams. Strengths include: (i) a clean many-body construction that dresses the cubic bubble with spectral densities; (ii) demonstrated collapse of κ_l to a unique value independent of the initial Gaussian width (Fig. 2a); (iii) mode-resolved validation of spectral functions against MD power spectra (Fig. 3), which is stronger evidence than transport coefficients alone; and (iv) an explicit note on concurrent related work. The HgTe case is well chosen because conventional three-phonon BTE calculations overestimate κ_l and are known to be smearing-sensitive. The framework is of clear interest to the anharmonic lattice-dynamics and thermal-transport communities.","major_comments":[{"comment":"The headline claim that cubic SCSF alone reduces κ_l by ~5× to experiment rests on evaluating the Peierls–Boltzmann form (Eq. 5) with the diagonal effective lifetime τ_eff of Eq. 6. After convergence, many modes (TO near Γ; LA/TA near zone boundary) are strongly non-Lorentzian with large γ/ω (Figs. 1e–f, 3). In that regime the diagonal spectral-function lifetime is only an approximation to the full many-body heat current. The manuscript asserts off-diagonal coherence contributions are <5% but does not recompute them with the converged, broadened spectra, nor does it compare against a Green–Kubo or Wigner evaluation that uses the same self-energy. Please either recompute coherence (and, if feasible, a Wigner/GK estimate) with the converged SCSF self-energies, or substantially qualify the attribution of the full experimental match to cubic self-consistency alone.","section":"Results, Eqs. (5)–(6); Figs. 1e–f, 2, 3"},{"comment":"Relatedly, the abstract and conclusion state that SCSF reproduces experimental κ_l and its temperature dependence “without explicitly invoking higher-order interactions,” and that faster-than-T^−1 decay “emerges here from self-consistent broadening of the cubic bubble.” Prior work on HgTe (including Ref. 56) has attributed ultralow κ_l in part to resonant four-phonon scattering. The manuscript does not quantify residual four-phonon (or higher) rates once the spectral functions are broad, nor does it show that those channels become negligible under SCSF. A controlled comparison—e.g., one-shot 3ph vs SCSF-3ph vs 3ph+4ph on the same force constants—or a clear statement that the experimental agreement is consistent with, but not uniquely diagnostic of, cubic self-consistency is needed so the claim is not overstated.","section":"Abstract; Results and discussion (Fig. 2b); Conclusion"},{"comment":"Eq. (6) defines τ_eff from ∫ A_λ²(ω) n(ω)[n(ω)+1] dω. The paper should state more carefully the assumptions under which this formula remains valid when A_λ is multi-peaked or strongly non-Lorentzian (as for TO near Γ in Fig. 1f). In particular, clarify whether the Bose factors and the prefactor n_λ(n_λ+1) are evaluated at the bare/renormalized quasiparticle frequency or integrated consistently with the spectral density, and whether sum-rule or positivity constraints on A_λ are enforced during the self-consistent loop. This is load-bearing for interpreting the mode-resolved scattering rates in Fig. 2c and the cumulative κ_l in Fig. 2d.","section":"Theory and methods, Eqs. (3)–(6); Fig. 1f"}],"minor_comments":[{"comment":"Fig. 1(c–d) would benefit from a quantitative convergence metric (e.g., integrated |γ^(n)−γ^(n−1)| or peak position/width vs iteration) rather than only visual inspection of five iterations.","section":"Fig. 1(c–d)"},{"comment":"The normalization convention g_λ(ω)=ω A_λ(ω)/ω_λ is stated briefly; a short sentence on sum rules and how numerical frequency grids preserve them would help reproducibility.","section":"Theory and methods, after Eq. (3)"},{"comment":"In Fig. 2(b), the power-law exponents (−1.03, −0.88, −1.68) should specify the fitted temperature window; experimental data scatter can affect the apparent exponent.","section":"Fig. 2(b)"},{"comment":"The MD comparison (Fig. 3) mentions an “appropriate temperature normalization factor” applied to the power spectra; state the factor explicitly and whether the same normalization is used in Fig. S2.","section":"Fig. 3; Supplementary Materials"},{"comment":"Preliminary CuCl results are cited as supporting generality (Fig. S3). A one-sentence quantitative summary in the main text (factor of reduction and comparison to experiment) would strengthen the generality claim without expanding the letter.","section":"Results and discussion (paragraph on generality)"},{"comment":"Typographical/notation: “the of group velocity” in the text after Eq. (5) should read “the group velocity”; ensure consistent use of λ vs qj for mode indices between Eqs. (5) and (6).","section":"Eqs. (5)–(6)"},{"comment":"The Note added helpfully distinguishes the present work from Refs. 72–73; consider one additional sentence on whether those formulations would yield the same dressed bubble or a different resummation, to orient readers.","section":"Note added"}],"recommendation":"major_revision","confidential_remarks":"The spectral-function construction and MD validation are solid and publishable; the main risk is over-claiming that cubic SCSF alone fully explains experimental κ_l in HgTe without a stronger transport or higher-order check. If the authors recompute coherence with converged spectra and temper the “without higher-order interactions” language, this could become a strong letter. Concurrent arXiv work (72, 73) is acknowledged; novelty of the explicit SCSF iteration plus MD spectral comparison still looks sufficient for this journal if claims are tightened."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real advance here is dressing the internal lines of the cubic bubble with full spectral densities and iterating the complex self-energy until both real and imaginary parts converge. That is not just another one-shot linewidth or frequency-only SCPH run. On HgTe it removes the smearing dependence, opens phase space that the quasiparticle calculation misses, and brings κ_l down by roughly a factor of five to the experimental scale with a steeper T dependence, all without putting four-phonon vertices in by hand.\n\nWhat the paper does well is concrete. Convergence is shown in a few iterations and is independent of the initial Gaussian seed. Mode-resolved spectral functions at L (and along Γ–L in the SI) line up with power spectra from an independently trained MTP MD trajectory. That is stronger evidence than a transport number alone. The derivation from the bubble through the spectral replacement and Kramers–Kronig closure is standard many-body theory written cleanly. Concurrent related preprints are disclosed rather than ignored.\n\nThe soft spot is real but limited. After self-consistency many modes (TO near Γ, LA/TA near zone boundary) are strongly non-Lorentzian with large γ/ω. The paper still evaluates κ_l with the diagonal Peierls formula and the effective lifetime extracted from A_λ(ω), asserting off-diagonal coherence stays <5 %. It does not recompute the coherence term or a Green–Kubo/Wigner current with the converged, broadened spectra. So the clean attribution of the entire experimental match to cubic self-consistency alone is a bit stronger than the transport evidence strictly supports. Residual higher-order diagrams could still matter; the paper does not close that door. Code and full computational details live in the SI, which is a practical friction for reuse.\n\nThis is for people who actually compute phonon lifetimes and κ_l in anharmonic crystals, especially systems with restricted three-phonon phase space or large intrinsic widths. The method is useful even if the transport formula needs tightening later. I would send it to referees; it is solid enough to deserve that time. Worth reading and, for me, worth citing when I next touch beyond-quasiparticle transport.","headline":"Clean cubic self-consistency that broadens spectra and drops HgTe κ_l by ~5× to experiment, with real MD mode-level checks; transport formula is the soft spot once modes go non-Lorentzian.","tokens_in":12328,"tokens_out":603,"would_cite":true,"duration_ms":6916,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Self-consistent spectral dressing of three-phonon processes broadens HgTe phonons enough to cut lattice thermal conductivity fivefold to experiment without higher-order interactions.","keywords":["phonon spectral functions","anharmonic lattice dynamics","lattice thermal conductivity","quasiparticle approximation","three-phonon scattering","self-consistent self-energy","HgTe","beyond quasiparticle"],"falsifier":"If explicit four-phonon calculations on the same force constants still produce a large further drop in HgTe thermal conductivity after SCSF convergence, or if high-resolution inelastic neutron or X-ray spectra at the L-point disagree with the converged SCSF lineshapes, the claim that cubic self-consistency alone accounts for the measured transport fails.","tokens_in":12378,"feed_emoji":"🌡️","tokens_out":961,"duration_ms":18617,"temperature":0.7,"pith_summary":"Standard first-principles phonon calculations treat modes as sharp quasiparticles with fixed frequencies, then evaluate three-phonon scattering once. This paper shows that the correct picture is a feedback loop: each phonon is a full frequency distribution that dresses the internal lines of the cubic bubble self-energy, which in turn updates the distributions until real and imaginary parts converge together. In zincblende HgTe the loop opens scattering channels that a one-shot calculation forbids, strongly broadens acoustic and optical spectra, and suppresses lattice thermal conductivity by roughly a factor of five, matching both the experimental magnitude and its temperature slope. Mode-resolved spectral functions also agree with power spectra from molecular dynamics. The result matters because many compounds whose heat transport has been blamed on four-phonon or higher processes may simply need this self-consistent treatment of the cubic interaction.","feed_headline":"Self-consistent phonons cut HgTe heat flow fivefold","feed_subtitle":"Dressing three-phonon bubbles with full spectral densities matches experiment without four-phonon terms","key_machinery":"Self-consistent spectral function (SCSF) iteration: normalized spectral densities g_λ(ω) replace fixed frequencies inside the cubic bubble; the resulting frequency-dependent self-energy (real and imaginary parts linked by Kramers–Kronig) updates the densities until both converge, eliminating external energy-conserving smearing.","core_discovery":"By representing every phonon with its full spectral density and iteratively dressing the internal lines of the three-phonon bubble, self-consistency substantially broadens acoustic and optical spectra in HgTe, activates scattering channels inaccessible under the quasiparticle approximation, and reduces lattice thermal conductivity by about a factor of five to the experimental scale while recovering its temperature dependence, without explicitly adding higher-order multi-phonon vertices.","pith_inferences":["Compounds whose three-phonon phase space is artificially closed by sharp frequencies (large gaps, flat bands) may systematically require this treatment before four-phonon diagrams are invoked.","The identical self-consistency loop could be transferred to electron–phonon or magnon–phonon spectral functions wherever the quasiparticle assumption fails.","If later dressing of explicit four-phonon vertices still converges quickly, the practical hierarchy of which anharmonic order dominates transport may need reordering.","Rapid five-iteration convergence for HgTe suggests the cubic feedback is a low-cost correction that existing Boltzmann-transport packages could absorb."],"forward_implications":["Many materials whose three-phonon Boltzmann calculations overestimate κ_l can be brought into experimental agreement by cubic spectral self-consistency alone.","Temperature scalings steeper than T^{-1}, often attributed to four-phonon scattering, can arise from self-consistently broadened three-phonon processes.","Mode-resolved spectral functions from SCSF match molecular-dynamics power spectra, giving a first-principles route to anharmonic lineshapes without higher-order vertices.","External numerical broadening parameters for energy conservation become unnecessary once spectral densities are used.","The same loop is stated to apply immediately to compounds with large acoustic–optical gaps or strong anharmonicity (CuCl and related halides)."],"fun_headline_variants":["Self-consistent spectral phonons slash HgTe κ fivefold","Full phonon spectra cut HgTe heat transport by factor of five","Dressing three-phonon bubbles drops HgTe conductivity to experiment","Self-consistent phonons activate new channels, match HgTe κ(T)","Beyond quasiparticles: spectral functions fix HgTe thermal conductivity"],"cache_read_input_tokens":128,"weakest_assumption_plain":"Once the cubic bubble is dressed with full spectral densities, leftover higher-order phonon processes stay small enough that the diagonal Peierls–Boltzmann conductivity already matches experiment.","fun_headline_variants_meta":{"raw":{"variants":["Self-consistent spectral phonons slash HgTe κ fivefold","Full phonon spectra cut HgTe heat transport by factor of five","Dressing three-phonon bubbles drops HgTe conductivity to experiment","Self-consistent phonons activate new channels, match HgTe κ(T)","Beyond quasiparticles: spectral functions fix HgTe thermal conductivity"]},"model":"grok-4.5","effort":"low","cost_usd":0.004312,"raw_usage":{"total_tokens":1232,"prompt_tokens":770,"num_sources_used":0,"completion_tokens":92,"cost_in_usd_ticks":43120000,"prompt_tokens_details":{"text_tokens":770,"audio_tokens":0,"image_tokens":0,"cached_tokens":0},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":370,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":770,"tokens_out":92,"duration_ms":4123,"temperature":1.0,"reasoning_tokens":370,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T13:27:27.709176+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"If explicit four-phonon calculations on the same force constants still produce a large further drop in HgTe thermal conductivity after SCSF convergence, or if high-resolution inelastic neutron or X-ray spectra at the L-point disagree with the converged SCSF lineshapes, the claim that cubic self-consistency alone accounts for the measured transport fails.","supporting_citations":[],"review_version":1}