{"id":"16ebf334-62b7-4052-9c2a-9090935bf395","arxiv_id":"1908.08758","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Born-Huang, Lemaitre-Maloney, and collective-mode derivations of nonaffine elastic constants coincide for unstressed equilibrium lattices, and a reduced-Hessian shortcut reproduces the same values.","lead":"This paper compares three standard mathematical recipes for computing the elastic stiffness of a crystal and shows they give the same answers for relaxed, stress-free crystals. It also offers a shortcut, the reduced-fields method, that avoids diagonalizing the full vibration matrix.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The equivalence in the abstract holds only for zero-stress reference states; Eq. (34) drops the bond-tension term using force balance, so the unqualified \"in equilibrium\" claim overstates the proven domain.","rationale":"After line-by-line check, I find no error in the 1D chains or in the algebraic identity between Eq. (39) and Eq. (36) under zero stress, zero temperature, and pairwise central forces. The reader's weakest assumption correctly identifies the boundary: Eq. (34) uses force balance to discard the bond-tension term, and the conclusion admits the methods coincide only for unstressed systems. The abstract is therefore overbroad; this is a presentation and caveat problem rather than a mathematical collapse. The α-quartz validation is thin, with four numbers from a self-cited companion simulation and no code or error bars, but it is not needed to support the formal equivalence. I also note that the zero-eigenvector basis described near Eq. (33) is not a valid set of Hessian eigenvectors as written; with the correct translation eigenvectors the cancellation still goes through, so this is a derivation-presentation issue rather than the main load-bearing gap. Since the reader already issued CONDITIONAL on essentially these grounds, the verdict should stand unchanged.","tokens_in":17594,"tokens_out":14326,"duration_ms":154925,"concrete_test":"Recompute the two-mass chain of §3.1 under a small reference tension F (nonzero bond tensions t_IJ satisfying Σ_J t_IJ n_IJ = F). Obtain C from BH minimization of the exact energy Eq. (40) and from LM Eq. (36) using the full affine force field of Eq. (34) without dropping the (1/2)t_IJ(δ_{κμ}n^ν + δ_{κν}n^μ) term. If the two results no longer match, the central equivalence depends on the zero-stress assumption and the abstract statement is false in its unqualified form.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central equivalence is established only for a reference state in mechanical equilibrium with zero internal stress. The step that carries this weight is Eq. (34), where the affine force field is written as Ξ^κ_{I,μν} = Σ_J (R_IJ s_IJ − t_IJ)n^κ n^μ n^ν; the term (1/2)t_IJ(δ_{κμ}n^ν + δ_{κν}n^μ) is dropped using Σ_J t_IJ n^μ_IJ = 0. That identity is the statement that no net bond-tension force acts on any atom, which holds at zero stress but fails under applied pressure or any internal prestress. When the reference state is prestressed, the dropped term contributes to the affine force field, and the LM modal-sum expression Eq. (36) is no longer the same object as the BH reduced-fields expression Eq. (39); finite-strain (Birch/Wallace) corrections also enter. The conclusion explicitly concedes this, stating that values obtained using the BH and LM methods only coincide if the system is unstressed, but the abstract's unqualified \"in equilibrium, elastic constants must be the same in all these methods\" omits the condition. The numerical α-quartz example is run under a zero-stress barostat, so it cannot validate the broader claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reviews three approaches to computing nonaffine elastic constants of solids: the Born-Huang (BH) Helmholtz-energy method, the Lemaitre-Maloney (LM) Gibbs-force method, and an acoustic collective-modes route. The authors argue that, in the linear elastic limit and in equilibrium, all three methods must give the same elastic constants, and they support this with explicit 1D chain calculations and with a 3D numerical example for α-quartz. They also introduce a 'reduced fields' formulation, Eq. (39), in which the nonaffine correction is written using an invertible reduced Hessian, and they claim this is equivalent to the LM modal-sum expression, Eq. (36).","tokens_in":17793,"tokens_out":5488,"duration_ms":59516,"significance":"If the equivalence is established under the stated conditions, the paper provides a useful unifying view of apparently different nonaffine-elasticity formalisms and offers a practical reduced-Hessian algorithm that avoids handling zero modes. The 1D two-mass chain derivations are clean and check out: BH, LM, collective modes, and the reduced-field method all yield C = ak/2. The α-quartz check shows internal consistency between the LM and reduced-field routes, though it is based on the same simulation data as the companion paper Cui et al. (2019b). The central limitation is that the equivalence is derived and validated only for an unstressed reference state; the abstract currently states the claim without that qualification.","major_comments":[{"comment":"The abstract states that 'in equilibrium' the elastic constants must be the same in all these methods, but the derivation in Section 2.3, Eq. (34), and the Conclusion both restrict the statement to unstressed systems; under prestress, finite-strain (Birch/Wallace) corrections enter and the methods need not coincide. Please qualify the abstract and the Introduction by adding 'at zero internal stress' to the statement of the central claim.","section":"Abstract and Conclusion"},{"comment":"The paper asserts that the LM modal-sum expression, Eq. (36), and the reduced-field expression, Eq. (39), 'will produce the same result', but it does not provide a general proof for an arbitrary 3D crystal. The 1D two-mass example verifies the identity, but the general claim requires showing that after fixing the d translational degrees of freedom, H̃^{-1} equals Σ_{i>d} e_i e_i^T / λ_i on the nonzero subspace, and that there are no additional zero modes beyond translations. Please add this derivation or explicitly state the extra assumption about the zero-mode structure.","section":"Section 3, Eqs. (36) and (39)"},{"comment":"The numerical agreement in Table 1 is obtained using the same simulation data and the same computational procedure as the companion paper Cui et al. (2019b). This confirms mutual consistency between the LM and reduced-field methods, but it does not independently validate the elastic constants against experiment or against an independent implementation. The text should make this distinction explicit.","section":"Section 3.2, Table 1"}],"minor_comments":[{"comment":"There are several typographical and stylistic issues: 'Lemaitre-Maloney's theories', 'Strasbourrg', 'Brich', 'or on', and inconsistent capitalization of 'Nonaffine' in the title. A careful proofreading pass is needed.","section":"Throughout"},{"comment":"The inserted paragraph on finite-pressure affine moduli and the Birch correction is important, but it reads as a standalone block. Consider integrating it with the qualification in the Conclusion so that the scope of the equivalence is stated in one place.","section":"Section 2.3"},{"comment":"For the collective-mode calculation the two masses are explicitly set equal, but Section 3.1 does not clearly state that this is an assumption of the example; please state it in the text and in the caption of Fig. 1(b).","section":"Section 3.1"},{"comment":"The notation n_{μ,max} = αL_μ for the reciprocal-space cutoff would benefit from a brief definition of L_μ as the cell length in direction μ.","section":"Section 3.2, Eq. (51)"}],"recommendation":"major_revision","confidential_remarks":"The only 3D numerical check in the paper is taken from the authors' own companion work (Cui et al., 2019b), so the comparison is an internal consistency test rather than an independent benchmark. Please ensure the published version states this clearly. The paper is within scope for cond-mat.soft and the 1D analytical examples are valuable, but the abstract currently overstates the domain of validity of the main equivalence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You can safely read this as a methods-comparison paper with a small practical pay-off. The main formulas check out, but the abstract overstates the domain: the equivalence between Born-Huang, Lemaitre-Maloney, and collective-mode paths holds for zero-stress reference states, and the authors themselves say so in the conclusion. The stress-test note is fair. Eq. (34) drops the bond-tension term using Σ_J t_IJ n_IJ = 0, which is fine at zero pressure and not fine under prestress. So the unqualified \"in equilibrium\" in the abstract is misleading, not a load-bearing mathematical error.\n\nWhat's genuinely useful: the reduced-fields formula, Eq. (39), gives the nonaffine correction as a simple inverse of the reduced Hessian. The 1D chains are worked cleanly and all three methods give C=ak/2, which builds confidence. The identity between the modal sum and the reduced-Hessian expression is valid when the only zero modes are translations. The paper also flags that some prior sums over eigenmodes counted zero modes; that point is likely correct in spirit, but it is asserted without showing the target equations, so a referee should ask for a precise citation and demonstration.\n\nThe soft spots are mostly presentational. The α-quartz validation is a four-number table from the authors' companion simulation with no error bars, data, or code; it demonstrates internal consistency, not an external benchmark. That's acceptable for a review, but weak as evidence. The text also contains a few apparently pasted-in paragraphs about finite-pressure affine moduli that interrupt Section 2.3; they concede the limitation, which is good, but they read as an editorial artifact. The correction to Zaccone and Scossa-Romano (2011) and Cui et al. (2019a) should be substantiated with their equations, not just stated.\n\nBottom line: this paper is for practitioners who want a compact, reliable path to nonaffine elastic constants and a clear statement of when the formalisms coincide. It deserves a serious referee; the abstract needs tightening, and the numerical section needs a little more transparency, but I don't see a fatal flaw. I would lean conditional accept.","headline":"A mostly clean methods-comparison: the reduced-fields formula is useful and the math holds at zero stress, but the abstract overstates the domain.","tokens_in":18404,"tokens_out":3175,"would_cite":true,"duration_ms":35155,"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 shows that the Born–Huang, Lemaitre–Maloney, and collective-mode methods return identical nonaffine elastic constants for unstressed equilibrium crystals.","keywords":["elasticity","nonaffinity","lattice dynamics","Born-Huang theory","Lemaitre-Maloney formalism","reduced Hessian","elastic constants","quartz simulation"],"falsifier":"Take a single non-centrosymmetric crystal at zero stress (for example the α-quartz model used in the paper), compute elastic constants by Eq. (36) and by Eq. (39); any difference beyond numerical precision shows the identity is wrong. Then repeat the same pair of computations under increasing hydrostatic pressure: the paper predicts a growing divergence between the methods, so observing no divergence would falsify the zero-stress premise.","tokens_in":17295,"feed_emoji":"💎","tokens_out":10211,"duration_ms":96817,"temperature":0.7,"pith_summary":"This paper sets out to show that three standard routes to the linear elastic constants of a lattice—minimizing Helmholtz energy (Born–Huang), balancing local forces in a Gibbs-type ensemble (Lemaitre–Maloney), and taking the long-wavelength limit of lattice vibrations—are three faces of one calculation. The proof rests on identifying a common object, the affine force field, and showing that the nonaffine correction takes the same quadratic form in every method: the normal-mode sum of Lemaitre–Maloney equals the reduced-fields formula of Eq. (39). If the equivalence holds, an equilibrium stress-free crystal has method-independent elastic constants, so simulations and theories can choose whichever formulation is computationally cheapest. The paper verifies the claim on one- and two-mass linear chains and on a numerical model of α-quartz, a non-centrosymmetric crystal where nonaffine relaxation is strong.","feed_headline":"Elastic constants: Helmholtz, Gibbs, and phonons agree","feed_subtitle":"For unstressed crystals, the choice of formalism no longer changes the result—energy minimization, force balance, and wave speeds match.","key_machinery":"The load-bearing object is the affine force field $\\Xi^\\kappa_{I,\\mu\\nu} = -\\partial^2 U/\\partial R^\\kappa_I \\partial \\eta_{\\mu\\nu}$, the force felt by atom $I$ when the cell is strained affinely by a unit strain. Together with the Hessian $H$, it enters the nonaffine correction through the sum over non-zero eigenmodes, $-\\frac{1}{V}\\sum_{i>d}(\\mathbf{e}_i\\cdot\\boldsymbol{\\Xi}_{\\mu\\nu})(\\mathbf{e}_i\\cdot\\boldsymbol{\\Xi}_{\\xi\\iota})/\\lambda_i$. The paper's bridge is the reduced Hessian $\\tilde{H}$, formed by deleting rows and columns corresponding to the $d$ translation zero modes, which is invertible and makes the BH minimization condition of Eq. (38) identical to the LM force-balance condition of Eq. (27). This reduced-fields form carries the equivalence: the same quadratic expression emerges from energy minimization, force balance, and the long-wave expansion of the dynamical matrix.","core_discovery":"On the paper's own terms, the central discovery is that the affine force field $\\boldsymbol{\\Xi}$ and the Hessian $H$ of the potential are the shared objects behind all three formalisms, and the nonaffine correction to elastic constants is always the same negative-definite quadratic form. In the LM language it is $C_{\\mathrm{NA}} = -(1/V)\\sum_{i>d} (\\mathbf{e}_i\\cdot\\boldsymbol{\\Xi}_{\\mu\\nu})(\\mathbf{e}_i\\cdot\\boldsymbol{\\Xi}_{\\xi\\iota})/\\lambda_i$, a sum over all non-translational eigenmodes, while in the BH/reduced-fields language it is $C_{\\mathrm{NA}} = -(1/V)\\tilde{\\boldsymbol{\\Xi}}^T \\tilde{H}^{-1}\\tilde{\\boldsymbol{\\Xi}}$, obtained after deleting the $d$ frozen translation modes. The paper shows these expressions are mathematically identical, and that the long-wavelength acoustic branches of the dynamical matrix reproduce the same tensor through Eq. (25). Consequently, for a lattice in mechanical equilibrium with no internal stress, the linear elastic constants no longer depend on which of the three methods is used.","pith_inferences":["The same reduced-Hessian identity should carry over to disordered athermal solids, where inversion symmetry is absent everywhere and nonaffine relaxation is the norm; if so, the expensive normal-mode sum in existing glass elasticity calculations could be replaced by a single inverse of the reduced Hessian.","A sharp experimental consequence of the zero-stress condition: under uniaxial tension, elastic constants inferred from phonon dispersion and from the static stress–strain slope should drift apart by an amount set by the Birch correction, providing a direct test of where the equivalence breaks.","One untested extension is finite temperature: replacing $H$ by the ensemble-averaged curvature and $\\boldsymbol{\\Xi}$ by a fluctuation correlator may preserve a similar identity in the stress-fluctuation formalism, but whether the quadratic form survives thermal averaging is not addressed by the paper."],"forward_implications":["For any unstressed equilibrium crystal, elastic constants computed from energy minimization, force balance, and phonon dispersion will agree, so the choice among the three methods is a matter of numerical convenience.","The nonaffine correction can be computed by inverting the reduced Hessian once and contracting with affine force fields, avoiding eigen-decomposition of the full Hessian and the need to discard zero modes.","Computational studies that sum over all $Nd$ modes, including the $d$ translation modes, overcount the nonaffine correction; the correct sum runs over the $(N-1)d$ non-zero modes.","If the reference state is prestressed, the small-strain elastic constants from BH and LM no longer coincide; finite-strain corrections of the Birch–Wallace type must be added, so the equivalence is a zero-stress theorem."],"supporting_citations":[{"why":"Supplies the Helmholtz-energy method of energy minimization and the long-wave collective-mode route that the paper compares.","marker":"Born and Huang, 1954"},{"why":"Introduces the Gibbs/local-force formalism and the eigenmode sum rule for nonaffine corrections that the paper reproduces as Eq. (36).","marker":"Lemaitre and Maloney, 2006"},{"why":"Extends the LM nonaffine response formalism and provides the analytic form used for the comparison and the correction of the mode count.","marker":"Zaccone and Scossa-Romano, 2011"},{"why":"Connects local inversion-symmetry breaking to nonaffine relaxation, motivating the quartz test case.","marker":"Milkus and Zaccone, 2016"},{"why":"Supplies the numerical α-quartz lattice-dynamics simulation whose elastic constants the paper reproduces with Eq. (39).","marker":"Cui et al., 2019b"},{"why":"Provides the Buckingham-plus-Coulomb force field used for the quartz simulation.","marker":"van Beest et al., 1990"},{"why":"Provides the fitted short-range potential parameters and cut-offs used to define the quartz model.","marker":"Carre et al., 2008"},{"why":"Supplies the Ewald summation method used to handle the long-range Coulomb interactions in the quartz simulation.","marker":"Ewald, 1921"},{"why":"Gives the finite-strain pressure corrections invoked to explain why prestressed systems break the equivalence.","marker":"Birch, 1938"},{"why":"Documents how different strain definitions alter the affine moduli under stress, delimiting the zero-stress claim.","marker":"Wallace, 1970"}],"fun_headline_variants":["Three formalisms, one set of elastic constants","Nonaffine elasticity: Helmholtz, Gibbs, and phonons agree","Crystal elastic constants: no method divide at equilibrium","Same elastic constants from energy, force, and wave speeds","Nonaffine corrections: identical across theories"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim collapses if the reference lattice is not stress-free: the derivation drops a term proportional to the bond tensions, relying on the equilibrium condition that those tensions cancel, and the paper itself states that Born–Huang and Lemaitre–Maloney values coincide only in the unstressed case.","fun_headline_variants_meta":{"raw":{"variants":["Three formalisms, one set of elastic constants","Nonaffine elasticity: Helmholtz, Gibbs, and phonons agree","Crystal elastic constants: no method divide at equilibrium","Same elastic constants from energy, force, and wave speeds","Nonaffine corrections: identical across theories"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000227,"raw_usage":{"total_tokens":1440,"prompt_tokens":880,"completion_tokens":560,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":496,"completion_tokens_details":{"reasoning_tokens":482}},"tokens_in":496,"tokens_out":560,"duration_ms":5228,"temperature":1.0,"reasoning_tokens":482,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:30:57.200403+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a single non-centrosymmetric crystal at zero stress (for example the α-quartz model used in the paper), compute elastic constants by Eq. (36) and by Eq. (39); any difference beyond numerical precision shows the identity is wrong. Then repeat the same pair of computations under increasing hydrostatic pressure: the paper predicts a growing divergence between the methods, so observing no divergence would falsify the zero-stress premise.","supporting_citations":[{"cited_title":"and Huang, K","cited_arxiv_id":null,"evidence_quote":"Supplies the Helmholtz-energy method of energy minimization and the long-wave collective-mode route that the paper compares."},{"cited_title":"and Maloney, C","cited_arxiv_id":null,"evidence_quote":"Introduces the Gibbs/local-force formalism and the eigenmode sum rule for nonaffine corrections that the paper reproduces as Eq. (36)."},{"cited_title":"and Scossa-Romano, E","cited_arxiv_id":null,"evidence_quote":"Extends the LM nonaffine response formalism and provides the analytic form used for the comparison and the correction of the mode count."},{"cited_title":"and Zaccone, A","cited_arxiv_id":null,"evidence_quote":"Connects local inversion-symmetry breaking to nonaffine relaxation, motivating the quartz test case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Buckingham-plus-Coulomb force field used for the quartz simulation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Ewald summation method used to handle the long-range Coulomb interactions in the quartz simulation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the finite-strain pressure corrections invoked to explain why prestressed systems break the equivalence."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents how different strain definitions alter the affine moduli under stress, delimiting the zero-stress claim."}],"review_version":1}