{"id":"76c0edde-6b3b-486e-a65b-b2916a272bd0","arxiv_id":"2502.01463","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Strain-driven ferroelectric instabilities can be traced to a sign reversal of the electron-phonon Berry curvature induced by band inversion.","lead":"The paper derives a connection between the geometric phase (Berry curvature) of electron-phonon coupling and the force constants that govern phonons, and argues that strain-induced band inversion flips this curvature's sign, softening a phonon and driving ferroelectric transitions. It tests this idea on a BiOCl monolayer and several other materials with density functional calculations.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central soft-phonon mechanism depends on C^s|2,3 staying strain-insensitive; this is asserted after Eq. 1 but never demonstrated, so the sign flip of C^s|1 may not control the total C^s.","rationale":"The reader's CONDITIONAL verdict identifies the same weakest assumption. I agree. The strongest claim in the paper is the causal link from EPC Berry curvature polarity reversal to phonon softening. That link has two premises: the quantum term flips sign, and the classical terms do not. The first is argued analytically with a two-band model and illustrated with a TB model for BiOCl. The second is merely asserted. Since Eq. 6 only gives C^s|1, the total force constant C^s could remain positive or even harden if C^s|2,3 varies oppositely. The paper's DFT phonon calculations show the total softening, and the TB model reproduces it, but the TB model is fitted to the DFT bands and does not independently compute C^s|2,3; therefore it cannot validate the invariance premise. A direct decomposition of the DFPT force matrix is the natural check. I also note a possible sign convention issue in the passage from Eq. 2 to Eq. 5, but the argument's dependence on a sign flip is robust to that, so the classical-term invariance is the more load-bearing concern. No change to the reader's CONDITIONAL verdict is needed; the concern is exactly the one already flagged.","tokens_in":9873,"tokens_out":5778,"duration_ms":54233,"concrete_test":"Perform DFPT calculations for the BiOCl monolayer at biaxial tensile strains from 0% to 5%, and evaluate the three terms in Eq. 1 separately, or at least the projection of each term onto the Eu mode eigenvector. Plot C^s|1, C^s|2,3, and their sum versus strain. The mechanism is supported only if C^s|1 flips sign near the critical strain, approximately 2.6%, while C^s|2,3 remains essentially flat; if C^s|2,3 shifts by a comparable amount, the claim that the quantum geometric term drives the instability fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that strain-induced band inversion reverses the sign of the EPC Berry curvature contribution C^s|1 (Eqs. 6 and 9), thereby driving C^s negative and softening the phonon. The inference requires that the classical terms C^s|2,3 in Eq. 1, comprising the electron-density response and the pure-ionic response, have no comparable polarity-reversal variation under symmetry-preserving strain. The text asserts this 'basically' after Eq. 1 and repeats it in the Geometric aspect section, citing the Jahn-Teller literature [22,45,47,51], but no direct calculation or decomposition is provided. Strain that produces band inversion also changes orbital overlaps, effective charges, and the ionic sublattice response; any of these can alter C^s|2,3 by an amount comparable to the expected change in C^s|1. Without a numerical separation of the three terms in Eq. 1 across the critical strain, the mechanism is not established; a change in C^s|1 alone is insufficient to explain phonon softening.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a quantum-geometric mechanism for strain-induced ferroelectric phase transitions. Within density-functional perturbation theory, the authors rewrite the electronic part of the interatomic force matrix, C^{ab}_{ij}|1, in terms of a force-gradient factor γ^s_nm and the imaginary part of an electron-phonon-coupling (EPC) quantum geometric tensor G^s_mn = ⟨∂_s m|n⟩⟨n|∇_k m⟩, which is the central result, Eq. (6). They argue that external strain that inverts the orbital energy difference of a quasi-degenerate electronic subsystem reverses the sign of Im[G^s_mn], flipping C^s|1 from positive to negative; because the classical terms C^s|2,3 are asserted to be essentially strain-insensitive, the total force constant C^s crosses zero and the Γ-point phonon softens. The mechanism is illustrated with a two-band Bloch-sphere model, tested on the BiOCl monolayer through a three-band tight-binding model fitted to the DFT band structure, interpreted through the bonding/anti-bonding phase of the FE-inducing wavefunction, and claimed to extend to seven additional materials.","tokens_in":10062,"tokens_out":20372,"duration_ms":165825,"significance":"If the central claim survives scrutiny, the paper makes a substantial contribution: it supplies a concrete microscopic quantity—the sign of the EPC Berry curvature, Im[G^s_mn]—that controls phonon softening and thereby connects soft-phonon theory, the (pseudo) Jahn-Teller effect, and quantum geometry, with a testable prediction that band inversion reverses this sign at the critical strain. The formal rewriting of the DFPT force matrix into Eq. (6), the analytic two-band argument, and the wavefunction interpretation in Fig. 2 are genuine strengths, and the multi-material survey in the supplement, if fully documented, would support generality. The model analysis is analytic and falsifiable rather than a post-hoc fit of the final result. However, the significance is conditional: the bridge from 'C^s|1 flips sign' to 'C^s becomes negative' rests on an undemonstrated assertion about C^s|2,3, and the BiOCl validation is partly a re-statement of fitted data. These gaps are closable within the scope of the manuscript, but until they are closed the main claim is not fully established.","major_comments":[{"comment":"The load-bearing assumption that the classical terms C^s|2,3 are strain-insensitive is asserted but never demonstrated. After Eq. (1) the text states that these terms 'are basically not supposed to have polarity-reversal variation induced by regulations that conserve the mirror symmetry of crystal lattice,' and the Geometric aspect section repeats that they 'typically do not undergo sharp transitions [22,45,47,51].' The assertion is deferred to Sec. II of the supplement, but the main text presents it as the pivot of the mechanism without any quantitative support. No numerical decomposition of Eq. (1) into its three terms, or of C^s into C^s|1 and C^s|2,3, is presented anywhere in the main text; this decomposition is precisely what is needed to establish that the computed sign flip of Im[G^s_mn] actually controls the total force constant. Strain that produces band inversion also changes the ground-state density n(r) (through the altering orbital weights), the orbital overlaps, the effective charges, and the ionic sublattice response, and any of these can shift C^s|2,3 by an amount comparable to the expected C^s|1 change. The paper should report, for BiOCl, the strain dependence of C^s|1 and C^s|2,3 separately—for example by direct evaluation of the three terms of Eq. (1) in DFPT, or by comparing the full DFPT force constant with the TB-model C^s|1—so that the mechanism is verified rather than assumed. The cited works [22,45,47,51] concern vibronic coupling and do not quantify the strain dependence of the ionic force constants.","section":"After Eq. (1); 'Geometric aspect' section"},{"comment":"The linearization in Eq. (3) is the step that turns the DFPT response (Eq. (2)) into the geometric form (Eqs. (5)–(6)), but the main text does not establish its small parameter for the systems under study. The justification offered is that for 'localized valence electrons with |ζ_mn − R_j| ≫ η_mn' the ionic potential is smooth and nearly linear near ζ_mn. In the BiOCl application, however, the FE-inducing states are predominantly O 2p orbitals and the soft Eu mode displaces the same oxygen sublattice on which these orbitals are centered; for such on-site matrix elements |ζ_mn − R_j| is of order η_mn, not much larger, and the neglected term o(|(r − ζ_mn)/(ζ_mn − R_j)|^2) is uncontrolled. The derivation referenced to Sec. I of the supplement should state the precise small parameter for the expansion and verify it for the O-2p states used in the BiOCl model; absent that, Eq. (6) cannot be claimed as a general rewriting of the DFPT interatomic force matrix.","section":"Eq. (3) and following text"},{"comment":"The two essential features of the EPC Berry curvature—(i) sign reversal under h_z → −h_z and (ii) divergence at quasi-degenerate points—support the entire general mechanism, yet in the main text they are asserted rather than derived, with the derivation deferred to Sec. IV A of the supplement. This would be acceptable for a short step, but it is not: Im[G^s_mn] depends on ∂_s θ, ∂_s φ, ∇_k θ, and ∇_k φ, and the sign-reversal claim must be checked together with the strain dependence of γ^s_nm, which enters Eq. (6) multiplicatively and is not held fixed by any argument in the text. In addition, the full-occupation case—the one relevant for large-gap semiconductors, feature (iii)—is said to receive its contribution from coupling to 'other unoccupied states |n⟩,' which are never specified; the eigenstates displayed in Eq. (10) only describe the internal |ψ+⟩–|ψ−⟩ transition. The main text should display the key result of the Sec. IV A calculation for both occupation regimes, or the authors should state plainly the auxiliary assumptions under which features (i) and (ii) hold.","section":"'Geometric aspect' section, after Eq. (10)"},{"comment":"The confirmatory BiOCl analysis is partly circular. The three-band tight-binding model is fitted to the DFT band structure and orbital weights (the text states it 'offers a great description to DFT results'), the EPC matrix elements are free parameters, and this same model is then used both to obtain the sign reversal of Im[G] (Fig. 1(c)) and to reproduce the strain dependence of the soft-phonon energy (Fig. S2(a)). Agreement of the model with the DFT data it was fitted to is a consistency check, not an independent test of the proposed mechanism. To close this gap the authors should either (i) compute Im[G^s_mn] directly from first-order DFPT or from the DFT wavefunctions without a fitted model, or (ii) predict the critical strain using parameters fixed at zero strain, or (iii) apply the sign-flip criterion to a material whose bands were not used in any fitting. As written, the application demonstrates that a fitted TB model is consistent with the mechanism, which is weaker than the claimed verification.","section":"'Application to BiOX'; Fig. 1(c), Fig. S2(a)"}],"minor_comments":[{"comment":"The second line of Eq. (2) introduces f_mn without defining it; specify f_mn = f_m − f_n and state explicitly that the replacement is valid for gapped systems.","section":"Eq. (2)"},{"comment":"The meaning of the subscript ∥ in 'Eu phonon mode (denoted by index ∥)' and of the ∥/⊥ notation in Eq. (11) is given only in the Fig. 1 caption; define it in the text.","section":"'Application to BiOX'; Eq. (11)"},{"comment":"The word 'regulations' is used several times (after Eq. (1), in the Geometric aspect section, and in the conclusions) where 'external perturbations' or 'tuning parameters' is meant; also 'regulations that conserve the mirror symmetry' should read 'perturbations that preserve the mirror symmetry.'","section":"After Eq. (1)"},{"comment":"Please check the prefactor in Eq. (9): with p^s_mn = −e Q̄_s Im[G^s_mn] and E^s_nm = Q̄_s γ^s_mn/e, the right-hand side equals ½∫[dk] Σ_mn f_m γ^s_nm Im[G^s_mn] Q̄_s^2, whereas ½ C^s|1 Q̄_s^2 from Eq. (6) equals ∫[dk] Σ_mn f_m γ^s_nm Im[G^s_mn] Q̄_s^2; the two differ by a factor of two unless the summation convention differs.","section":"Eq. (9)"},{"comment":"The concluding claim that the theory 'provides a quantum geometric interpretation of the (pseudo) Jahn-Teller effect' is not substantiated in the main text; the two-state subsystem coupled to a vibration is structurally identical to the standard pseudo-JT model, and the authors should either show what is gained relative to that framework or cite the specific derivation.","section":"Conclusions"},{"comment":"Eq. (8) is presented as a 'fundamental relation' taken from the unpublished preprint [41]; since the interpretation of the divergence at quasi-degeneracy rests on the 1/ε_mn^2 structure, its derivation should be given in the supplement rather than relying on a non-peer-reviewed reference.","section":"Eq. (8)"},{"comment":"The claimed extension to 'a broad range of materials' (BiOBr, BiOI, Bi2O2Se, BiCuOSe, Bi3O4Br, PbClF, PbO) is confined to Sec. VI of the supplement; the main text should at least show the critical-strain table or a representative case, otherwise the generality claim is unverifiable from the Letter itself.","section":"'Application to BiOX'"},{"comment":"Some references are incomplete: [15] ('Arxiv-Cond-Mat.supr-Con (2018)') and [18] ('Physical Review Letter') contain errors, and [34] and [40] lack page numbers; also, the supplemental material [45] is cited without a DOI or repository link.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"I assessed the main text of the Letter; the derivations for the three under-supported steps (the Eq. (3) linearization, the Sec. IV A two-band features, and the Eq. (1) decomposition) are all deferred to the supplemental material, which was not part of the review package. If the supplement already contains a numerical decomposition of Eq. (1) into C^s|1 and C^s|2,3 across the critical strain, Major Comment 1 can be largely resolved; the editor may wish to ask the authors to move that test into the main text, as it is the pivot of the entire mechanism. Similarly, if Sec. IV A addresses the full-occupation case with explicit external states, Major Comment 3 is partly answered. The novelty is real and the paper fits the journal; my major_revision recommendation is driven by the wish to see the load-bearing assertions demonstrated, not by any concern about the direction of the work."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things up front. First, the paper's central move is new: writing one piece of the interatomic force matrix as the imaginary part of an electron-phonon quantum geometric tensor (Eq. 6) and arguing that strain-driven band inversion flips its sign, softening a phonon. That is not in the prior literature, which used EPC quantum geometry for optical responses and phonon Hall effects, not for the force matrix. Second, the paper is honest about its own scaffolding: the key linearization in Eq. 3, the claim that the classical terms C_2,3 do not reverse sign, and the two-band derivation are all deferred to the supplemental, which I could not check. That is the main soft spot, and it is load-bearing.\n\nWhat is genuinely good: the formal DFPT-to-Eq.6 path is coherent, and the physical picture—fully occupied bands contributing to phonon energy through inter-band coherence even with zero net polarization—is a useful corrective to focusing only on Born effective charges. The two-band model cleanly illustrates how orbital energy inversion flips Im[G], and the wavefunction bonding/anti-bonding argument in Fig. 2 is a nice, concrete way to see why the sign of the orbital splitting matters. The multi-material DFT survey (BiOBr, BiOI, Bi2O2Se, etc.) is a reasonable claim of generality, though I have not seen the supplemental tables.\n\nNow the soft spots, in proportion. The biggest one is exactly what your stress-test note flags: the entire mechanism requires C_2,3 to be roughly strain-insensitive, but the paper asserts this after Eq. 1 and in the Geometric section, citing Jahn-Teller literature rather than showing a numerical decomposition of C_1 versus C_2,3 across the critical strain for BiOCl. Strain that inverts bands also changes overlaps and effective charges; without a direct calculation, a sign flip in C_1 alone does not prove phonon softening. This is an addressable weakness, not a refutation. Second, the BiOCl validation uses a three-band TB model fitted to the DFT bands and orbital weights, and the phonon-energy agreement (Fig. S2a) comes from that same model; that is circular in a mild but real sense. Third, the linearization in Eq. 3 is plausible but the text acknowledges the higher-order term only in the o(...) symbol; for large phonon displacements or near degeneracies, the leading-term argument needs a check.\n\nMy bottom line: the core idea is worth taking seriously and the paper deserves a proper referee, but only if the supplemental actually contains the decomposition of C_1 versus C_2,3 under strain and the fit parameters. If that is there, this is a solid PRL-class contribution. If not, it is an interesting hypothesis with a suggestive but incomplete test. I would not desk-reject; I would send it to a referee who knows both DFPT and quantum geometry and ask them to verify the supplemental derivation. The paper is for the ferroelectrics and lattice-dynamics community, and for the quantum-geometry crowd. A serious referee should engage it; my own verdict is conditional until the supplemental is available.\n\nRecommendation: send to peer review, with an explicit request to check the C_2,3 strain-invariance argument and the TB fitting procedure.","headline":"A plausible and potentially useful reformulation of the soft-phonon mechanism in terms of EPC Berry curvature, but the central strain-insensitivity assumption for the classical force terms is asserted, not shown; the BiOCl validation is suggestive, not conclusive.","tokens_in":10645,"tokens_out":818,"would_cite":false,"duration_ms":9977,"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":"This paper claims that strain-induced ferroelectric phase transitions are driven by a sign flip of the electron-phonon Berry curvature in a nearly degenerate electronic subsystem, which reverses the quantum part of the interatomic force…","keywords":["ferroelectricity","electron-phonon coupling","Berry curvature","quantum geometry","strain-induced phase transition","soft phonon","density-functional perturbation theory","BiOCl monolayer"],"falsifier":"A direct DFPT decomposition of the force-constant matrix of BiOCl into the quantum term $C^{s}|1$ and the classical terms $C^{s}|2,3$, computed as a function of biaxial strain across the critical value of about 2.6 percent, would settle the claim: if the classical terms shift appreciably or change sign while the $E_u$ mode softens, the proposed mechanism fails, whereas if they stay flat and only $C^{s}|1$ crosses zero, the mechanism is supported.","tokens_in":9634,"feed_emoji":"⚡","tokens_out":10385,"duration_ms":81033,"temperature":0.7,"pith_summary":"This paper claims that strain-induced ferroelectric phase transitions have a quantum-geometric origin: an external strain can invert the orbital energy ordering of a nearly degenerate electronic subsystem, which reverses the sign of a Berry curvature attached to electron-phonon coupling, turning a phonon's restoring force negative and driving the mode soft. The claim is derived within density-functional perturbation theory, where the quantum part of the interatomic force matrix is expressed as an integral over the imaginary part of an electron-phonon coupling (EPC) quantum geometric tensor, $G^{s}_{mn}=\\langle\\partial_s m|n\\rangle\\langle n|\\nabla_{\\mathbf{k}}m\\rangle$. The paper then shows the mechanism operating in the BiOCl monolayer, where biaxial tensile strain above about 2.6 percent softens an $E_u$ phonon mode exactly as band inversion of oxygen $p$ states flips the curvature, and it reports the same signature in seven further materials. A sympathetic reader would care because the mechanism unifies the soft-phonon and (pseudo) Jahn-Teller pictures into a single computable, strain-tunable microscopic quantity.","feed_headline":"Why strain softens phonons: a Berry-curvature sign flip","feed_subtitle":"Quantum geometry ties strain-induced band inversion to ferroelectric transitions in BiOCl and beyond.","key_machinery":"The central object is the EPC quantum geometric tensor $G^{s}_{mn}=\\langle\\partial_s m|n\\rangle\\langle n|\\nabla_{\\mathbf{k}}m\\rangle$, defined in the hybrid Hilbert space parameterized by the displacement of phonon mode $s$ and the electronic wavevector $\\mathbf{k}$; its imaginary part, the EPC Berry curvature $\\mathrm{Im}[G^{s}_{mn}]$, is a Berry magnetic field in that space. The derivation reduces the quantum term $C^{s}|1$ of the interatomic force matrix to an integral of $\\gamma^{s}_{nm}\\ \\mathrm{Im}[G^{s}_{mn}]$ (its Eq. 6), so the sign and magnitude of the curvature directly set the sign of the phonon restoring force. The argument turns on two properties of this curvature in a nearly degenerate two-band subsystem: it diverges as the gap closes, making it the dominant contribution to $C^{s}|1$, and it flips sign when the orbital energy difference $h_z$ is inverted by strain.","core_discovery":"On the paper's own terms, the discovery is that the contribution $C^{s}|1$ of the perturbed electron density to the interatomic force matrix of a phonon mode $s$ is controlled by the EPC Berry curvature $\\mathrm{Im}[G^{s}_{mn}]$, the imaginary part of the geometric tensor $G^{s}_{mn}=\\langle\\partial_s m|n\\rangle\\langle n|\\nabla_{\\mathbf{k}}m\\rangle$ built from the derivative with respect to the phonon displacement and the $\\mathbf{k}$-space Berry connection. Because this curvature diverges near quasi-degenerate band points and its polarity is tied to the orbital energy difference $h_z$, a symmetry-preserving regulation such as strain that inverts the band ordering sends $C^{s}|1$ to $-C^{s}|1$ while the classical ionic terms $C^{s}|2,3$ stay essentially fixed, so the total force constant can cross zero and the phonon frequency becomes imaginary. The paper demonstrates the reversal in a two-orbital Pauli model, reproduces the DFT-computed $E_u$ soft-mode energy of the BiOCl monolayer as a function of strain with a three-band tight-binding model, and reports the same band-inversion-plus-curvature-reversal signature in BiOBr, BiOI, Bi2O2Se, BiCuOSe, Bi3O4Br, PbClF, and PbO.","pith_inferences":["If the polarity-reversal mechanism is as generic as the paper suggests, candidate strain-tunable ferroelectrics could be screened by computing orbital energy differences and EPC Berry curvature at high-symmetry points alone, without full phonon calculations.","A testable extension the paper leaves implicit: transiently flipping the orbital energy difference by photoexcitation should transiently soften the same phonon mode, linking the geometric mechanism to ultrafast, THz-frequency structural switching.","Because the quantum force-constant term can be asymmetric under exchange of spatial directions, materials with the same band-inversion signature but lower lattice symmetry may show a strain-induced chirality or rotational character in the softened mode, an extrapolation beyond the mirror-symmetric cases treated here."],"forward_implications":["Strain engineering of ferroelectricity reduces to tuning band inversion in specific FE-inducing electronic subsystems, so fully occupied valence bands can drive the transition without closing the global band gap.","The mechanism gives a microscopic reading of the (pseudo) Jahn-Teller effect as the singular behavior of EPC quantum geometry, since both require near-degenerate states coupled to a phonon.","The soft mode that appears at the transition breaks inversion or mirror symmetry, and the resulting polarization carries the same symmetry representation as the unstable mode, identifying the transition as ferroelectric.","Because the EPC Berry curvature is detectable through phonon-mediated optical responses, the predicted softening should appear as a strain-dependent optical signature, not only as a calculated phonon frequency.","The same band-inversion-plus-curvature-reversal condition should govern strain-induced ferroelectric transitions in any material with a quasi-degenerate subsystem, consistent with the paper's DFT results for BiOBr, BiOI, Bi2O2Se, BiCuOSe, Bi3O4Br, PbClF, and PbO."],"supporting_citations":[{"why":"Supplies the DFPT expression for the interatomic force matrix (Eq. 1) that the paper splits into quantum and classical contributions.","marker":"[44]"},{"why":"Establishes the relation between the quantum geometric tensor and the electron-phonon matrix element and velocity (its Eq. 8), grounding the claimed experimental detectability.","marker":"[41]"},{"why":"Provides the Berry-connection identity $r_{nm}=i\\langle n|\\nabla_{\\mathbf{k}}m\\rangle$ used to rewrite the force matrix in geometric form.","marker":"[25]"},{"why":"Supplies the geometric theory of the insulating state from which the $k$-space Berry connection is adopted.","marker":"[28]"},{"why":"Justifies expressing the ionic potential as a sum of spherical single-ion potentials, the step behind the derivative approximation in Eq. 3.","marker":"[48]"},{"why":"Documents the (pseudo) Jahn-Teller paradigm that the paper reinterprets and supports the assertion that classical vibronic terms do not vary sharply under symmetry-preserving perturbations.","marker":"[22]"},{"why":"Shows the spatial asymmetry of the quantum force-constant term, supporting its treatment as a distinct geometric contribution to lattice dynamics.","marker":"[47]"}],"fun_headline_variants":["Berry curvature sign flip drives ferroelectric transitions","Strain flips Berry curvature, softens phonons","Quantum geometry behind strain-induced ferroelectricity","How strain inverts Berry curvature to trigger ferroelectricity","Electron-phonon Berry curvature: the key to strain-driven ferroelectricity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument's load-bearing premise is that the classical, purely ionic contributions to the interatomic force constants stay essentially unchanged when strain is applied while the crystal's mirror symmetry is preserved, so the sign flip of the quantum-geometric term $C^{s}|1$ is not cancelled or overwhelmed by a strain-driven shift in $C^{s}|2,3$.","fun_headline_variants_meta":{"raw":{"variants":["Berry curvature sign flip drives ferroelectric transitions","Strain flips Berry curvature, softens phonons","Quantum geometry behind strain-induced ferroelectricity","How strain inverts Berry curvature to trigger ferroelectricity","Electron-phonon Berry curvature: the key to strain-driven ferroelectricity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000237,"raw_usage":{"total_tokens":1545,"prompt_tokens":1022,"completion_tokens":523,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":638,"completion_tokens_details":{"reasoning_tokens":439}},"tokens_in":638,"tokens_out":523,"duration_ms":5636,"temperature":1.0,"reasoning_tokens":439,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T15:12:54.331445+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct DFPT decomposition of the force-constant matrix of BiOCl into the quantum term $C^{s}|1$ and the classical terms $C^{s}|2,3$, computed as a function of biaxial strain across the critical value of about 2.6 percent, would settle the claim: if the classical terms shift appreciably or change sign while the $E_u$ mode softens, the proposed mechanism fails, whereas if they stay flat and only $C^{s}|1$ crosses zero, the mechanism is supported.","supporting_citations":[{"cited_title":"Baroni, S","cited_arxiv_id":null,"evidence_quote":"Supplies the DFPT expression for the interatomic force matrix (Eq. 1) that the paper splits into quantum and classical contributions."},{"cited_title":"Calandra, G","cited_arxiv_id":null,"evidence_quote":"Justifies expressing the ionic potential as a sum of spherical single-ion potentials, the step behind the derivative approximation in Eq. 3."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows the spatial asymmetry of the quantum force-constant term, supporting its treatment as a distinct geometric contribution to lattice dynamics."}],"review_version":1}