{"id":"e4ee809b-b63c-4368-89bb-0452759d391d","arxiv_id":"2608.10041","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A closed-form analytic expression for the shape-dependent boundary term in Madelung lattice sums is derived for arbitrary triclinic crystal geometries, enabling accurate direct-sum evaluation of Madelung constants.","lead":"This paper derives a closed-form mathematical formula for the shape-dependent boundary term that appears when computing electrostatic (Madelung) sums in crystals, valid for any crystal lattice geometry. The formula enables accurate direct-sum evaluation of Madelung constants in low-symmetry triclinic lattices, a problem previously left in integral form.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The numerical validation leans on an unproven (2p+1)^{-2} finite-size correction; without it, p=60 agreement is only ~10^{-5}, so the claimed 10^{-10} accuracy may reflect the fitted ansatz rather than the analytic boundary term.","rationale":"The paper's principal analytical contribution, the closed-form parallelogram-potential boundary term in Eq. (8), is supported by a plausible derivation, a correct rectangular-prism limit, and consistency between regular and irregular shapes. The GitHub code and use of known exact Madelung constants are independent verification assets. However, the numerical validation of the direct-sum method depends on a finite-size correction whose scaling law is asserted without proof, and the reported precision is entirely contingent on that law. The reader identified exactly this assumption as the weakest point; I agree. Because the concern is about the validation protocol rather than a demonstrated error in the central formula, it does not warrant rejection, but it does justify the CONDITIONAL verdict already given. A focused numerical test of the Q_p = K^2 residual would settle whether the correction is a true asymptotic law or a fitting artifact. No additional concern about the analytic formula itself rose to the same level of load-bearing importance.","tokens_in":15417,"tokens_out":11000,"duration_ms":115984,"concrete_test":"Compute Q_p = (2p+1)^2[(ν - ν_b) - M_exact] for NaCl and ZnS using the released code at p = 10, 20, 60, and, if feasible, p = 100. If Q_p is constant to below 10^{-10}, the K^{-2} law is confirmed; if Q_p drifts by more than 10^{-9}, sub-leading terms contaminate the correction. Then recompute the corrected p=60 values using C estimated from (p,p-1) versus (p,p-2); if the two corrected values differ by more than 10^{-9}, the reported 10^{-10} errors are an artifact of the fitting scheme.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central analytic boundary-term formula, Eq. (8), is not the weakest point: the derivation from the dipole-continuum surface integral is internally consistent, and the rectangular-prism limit in Eq. (37) matches earlier results. The load-bearing weakness is the finite-size correction used to validate it. Section V assumes the residual decays exactly as C(2p+1)^{-2} and estimates C from the difference between consecutive p values, then subtracts C/(2p+1)^2. Without this correction, the p=60 data agree with exact Madelung constants only to about 10^{-5}; the tables' 10^{-10} agreement is generated entirely by this unproven two-point extrapolation. The abstract states that the (2p+1)^{-2} decay is 'demonstrated', but the paper provides no derivation or independent check, and the wollastonite application reports six-digit energies with no error analysis. If sub-leading terms of order (2p+1)^{-3} or shape-dependent corrections are non-negligible at K=121, the fitted C would absorb them and bias the corrected values. The regular/irregular shape comparison provides some evidence that the boundary term itself is correct, but it does not test the correction law independently.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper derives a closed-form analytic expression for the shape-dependent boundary term ν_b(r|s) in finite Coulomb lattice sums, Eq. (8), for triclinic Bravais lattices with parallelogram-faced exact crystal shapes. The boundary term is expressed as a sum over facets of the gradient of the electrostatic potential of a uniformly charged parallelogram, with explicit formulas for φ and g in Eqs. (20)–(32). The manuscript further asserts that the residual finite-size correction decays as (2p+1)^{-2}, and uses a fitted coefficient to correct direct sums. Numerical validation is reported for NaCl, ZnS, and CaF2 in regular and irregular shapes, and the method is applied to triclinic wollastonite.","tokens_in":15675,"tokens_out":7552,"duration_ms":71071,"significance":"If the analytic formula is correct, it resolves a long-standing gap in lattice-sum theory: shape-dependent boundary terms are usually expressed only as surface integrals. The derivation is explicit, self-contained given the prior decomposition of Ref. 15, and consistent with known limits: Eq. (37) recovers the rectangular-prism case and Sec. IV reproduces Rayleigh’s geometric factors 2π and 2π/3. The availability of code and data on GitHub is a strength. The central weakness is that the numerical validation’s claimed high accuracy is produced by an unproven, fitted finite-size correction, so the verification of the analytic formula is weaker than the tables suggest. Revision is required to make the validation load-bearing.","major_comments":[{"comment":"The claimed (2p+1)^{-2} decay of the finite-size correction is asserted without derivation. The text says the difference ν(r,p|s)−ν(r,p−1|s) scales as C[(2p+1)^{-2}−(2p−1)^{-2}], from which C is estimated and subtracted as C/(2p+1)^2, but no proof or independent check of this leading-order law is given. Without the correction, the p=60 NaCl value in Table II is ν−ν_b = 1.747580 versus the exact 1.74756459463318, an error of about 1.5×10^{-5}; the reported 1.1×10^{-10} error is entirely produced by the fitted two-point extrapolation. Because the abstract states that this decay is “demonstrated,” the gap is load-bearing. Please supply a derivation of the asymptotic correction or an independent verification (e.g., higher-order Richardson fits or a comparison with Ewald values at several p), and report uncorrected residuals alongside corrected ones.","section":"Section V, Eq. (42), Tables II–III"},{"comment":"The wollastonite electrostatic energies are quoted to six digits with no error estimate, yet they are used to conclude that the cubic-to-triclinic transition is energetically favored. A convergence study over p (e.g., p=10, 20, 40) and over regular/irregular shapes, or explicit error bars, is needed to support the claimed six-digit accuracy.","section":"Section V, Table IV"}],"minor_comments":[{"comment":"The phrase “arbitrary crystal geometries” is broader than what is actually treated: the derivation applies to the family of centrosymmetric, parallelogram-faced exact shapes defined in Section II (regular shapes plus their composite unions). Please state the scope precisely.","section":"Section III, abstract"},{"comment":"The appendix’s text reads “where the logarithmic and arctangent contributions are given, respectively, by and φ_atan”; the explicit expression for φ_log appears to be missing. The formulas are present in the main text, but the appendix should be self-contained.","section":"Appendix B, after Eq. (B18)"},{"comment":"The symbol c is used both for the facet-center vector c_j in Eq. (6) and for the intermediate vector c = r ± a in Eq. (B14), which is confusing. Rename one of the two.","section":"Section II, Eq. (6), and Appendix B, Eq. (B14)"},{"comment":"The tables would be clearer if the absolute error after correction were reported directly; the current use of ϵ and the √3/2 factor for ZnS is opaque and makes the validation harder to check.","section":"Section V, Tables II–III"},{"comment":"Reference [46] is an online database entry with only an access date; the journal’s reference style should include the specific structure determination and, if possible, a DOI.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The core analytic result appears sound, but the numerical validation is currently tied to an unproven fitted correction. This is fixable with a derivation or additional convergence tests, so I recommend major revision rather than rejection. The novelty relative to the authors’ prior Refs. 14 and 15 should also be highlighted in the revised introduction."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: the central result—a closed-form boundary term ν_b for arbitrary triclinic lattices—is new and, as far as I can tell, correct. Earlier analytic boundary terms covered rectangular prisms in orthogonal lattices; here the boundary term is mapped to the potential gradient of uniformly charged parallelograms, and the formula reduces cleanly to the known rectangular limit. The derivation in Sections II–III and Appendices A–B is self-contained, uses standard Euler-substitution techniques, and gives explicit expressions for both φ and g. The authors also ship Fortran code and data, and the benchmark values are checked against external OEIS constants. That is real, reproducible work. The test on both regular and irregular shapes for NaCl, ZnS, and CaF₂ is a genuine check: the corrected values agree between shapes, which would not happen if the boundary term were simply wrong.\n\nNow the soft spot. The claimed 10^-10 agreement depends on a two-point correction ν_corr = C/(2p+1)^2, with C fitted from the difference between consecutive p values. The paper never derives that decay law; Section V simply states that the difference scales that way. Looking at the tables, p=60 without the correction gives roughly 1.5×10^-5 error for NaCl. The dramatic agreement is generated entirely by the fitted ansatz. That does not invalidate the boundary-term formula—the analytic derivation stands on its own—but it does mean the abstract's phrase \"we demonstrate\" is too strong. This is an empirical fit, not a proof. A referee should ask for either a derivation of the leading correction or an independent check against Ewald sums or higher-order extrapolations at several p values.\n\nThe wollastonite application reports six-digit energies with no error analysis. That is a minor issue relative to the main result, but it should be addressed. The Rayleigh connection in Section IV is a nice historical note, though it does not carry the main argument. The citation pattern is mostly reasonable; there is a cluster of self-citations to the authors' prior exact-shape work, but those are prior published results rather than circular references.\n\nBottom line: the core analytic contribution looks solid and solves a real long-standing problem. The finite-size correction is a genuine weakness, but it sits in the validation layer, not in the boundary-term derivation itself. I would send this to peer review and ask for a major revision focused on the correction law and the wollastonite error estimate. I would cite it if I worked on lattice-sum or direct-summation methods.","headline":"New closed-form boundary term for triclinic Madelung sums is real and likely correct; the main soft spot is an unproven two-point finite-size correction that generates the claimed 10^-10 validation accuracy.","tokens_in":16170,"tokens_out":1974,"would_cite":true,"duration_ms":19792,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Madelung boundary term made analytic for any triclinic crystal.","keywords":["Madelung constant","lattice sum","conditional convergence","boundary term","triclinic lattice","direct summation","electrostatic potential","finite-size correction"],"falsifier":"For a fixed crystal shape, compute $D(p) = \\nu(p|s) - \\nu(p-1|s)$ after subtracting the analytic boundary term, and check whether $D(p)/[(2p+1)^{-2} - (2p-1)^{-2}]$ is constant as $p$ grows; if the ratio drifts or fails to converge, the assumed $(2p+1)^{-2}$ decay law is incomplete. Alternatively, compare the corrected direct sum against a high-accuracy Ewald result for a triclinic cell at $p=100$ and $p=200$; discrepancies above the claimed precision would falsify the correction scheme.","tokens_in":15223,"feed_emoji":"⚛️","tokens_out":5748,"duration_ms":49424,"temperature":0.7,"pith_summary":"Conditionally convergent Madelung lattice sums contain a shape-dependent boundary term whose analytic form has resisted solution for decades. This paper claims to close that problem: for arbitrary triclinic crystal geometries, the boundary term equals a sum over the crystal's parallelogram facets, where each facet contributes a term built from the gradient of the electrostatic potential of a uniformly charged parallelogram, given explicitly in closed form. The paper further claims that after subtracting this boundary term, the residual finite-size correction for a crystal of characteristic size $p$ decays as $(2p+1)^{-2}$. If these claims are right, direct summation of the lattice sum, without Ewald or Fourier transforms, becomes a practical and accurate route to Madelung constants even for low-symmetry triclinic crystals.","feed_headline":"Madelung boundary term made analytic for any triclinic crystal","feed_subtitle":"Direct summation with closed-form facet potentials recovers exact Madelung constants for low-symmetry lattices.","key_machinery":"The carrying object is the parallelogram potential $\\phi(a,b,r)$ and its gradient $g(a,b,r)$. Here $\\phi$ is the electrostatic potential at field point $r$ of a uniformly charged parallelogram centered at the origin and spanned by vectors $2a$ and $2b$, written as a double integral over $t,\\tau\\in[-1,1]$. The paper evaluates this integral in closed form via Euler substitution, yielding $\\phi=\\phi_{\\log}+\\phi_{\\text{atan}}$, where $\\phi_{\\log}$ sums logarithmic terms over the four vertices and $\\phi_{\\text{atan}}$ is a weighted sum of arctangents. The gradient $g$ decomposes into three algebraic vector terms $G_1,G_2,G_3$, a logarithmic term $L$, and an arctangent term $A$. The boundary term is then $\\nu_b(r|s) = \\frac{1}{2V}\\sum_j \\frac{r\\cdot(a_j\\times b_j)}{|a_j\\times b_j|}\\, r\\cdot g(a_j,b_j,c_j)$, with $c_j$ pointing from the crystal center to each facet center. The key identity enabling the derivation is $\\nabla_x(1/|x|) = \\nabla_{c_j}(1/|t a_j + \\tau b_j + c_j|)$, which turns the facet surface integral into a single closed-form potential calculation that is then differentiated.","core_discovery":"The central discovery is that the shape-dependent boundary term $\\nu_b(r|s)$ in the decomposition $\\nu = \\nu_{\\text{pbc}} + \\nu_b + \\nu_{\\text{corr}}$ is not an obstacle but a computable geometric object. After applying the divergence theorem, the boundary term is expressed as a sum over the $N_s$ parallelogram facets of the finite crystal, and each facet contribution is built from the gradient $g(a,b,r)=\\nabla \\phi(a,b,r)$ of the potential of a uniformly charged parallelogram. The parallelogram potential $\\phi$ is evaluated in closed form as $\\phi_{\\log}+\\phi_{\\text{atan}}$ with explicit logarithmic and arctangent terms, and its gradient decomposes into the algebraic, logarithmic, and arctangent parts given in Eqs. (27)-(32). The paper also establishes that the remaining finite-size correction scales as $(2p+1)^{-2}$, so two consecutive crystal sizes suffice to extrapolate to the bulk limit. Numerical validation on NaCl, ZnS, and CaF$_2$ in an FCC Bravais lattice recovers the exact Madelung constants to about $10^{-10}$ at $p=60$, and the method is demonstrated on the triclinic wollastonite lattice.","pith_inferences":["Editorial inference: because the boundary term is expressed through facet potentials, the same $\\phi$ and $g$ formulas could be reused as geometric building blocks in force calculations, potentially lowering the cost of direct summations for finite nanocrystals beyond what the paper demonstrates.","Editorial inference: the paper's recovery of Rayleigh's $2\\pi$ and $2\\pi/3$ geometric factors suggests the parallelogram-potential machinery may transfer to effective-medium problems, such as conductivity, heat conduction, and refractive index, where the same conditional lattice sums appear.","Editorial inference: the $(2p+1)^{-2}$ correction law is verified numerically but not proven; if proven, the method's error estimate would become rigorous, and the same scaling could plausibly apply to other inverse-distance lattice sums beyond electrostatics."],"forward_implications":["For any finite crystal of exact shape and size, the shape-dependent part of the Madelung sum is computed exactly from facet geometry, so bulk Madelung constants can be extracted from direct sums without Ewald or Fourier methods.","The boundary-term formula works for both regular and irregular or composite crystal shapes; regular and irregular shapes yield consistent bulk energies after correction, confirming the separation of boundary and finite-size effects.","The $(2p+1)^{-2}$ decay law for the residual correction means that direct summation with two or three crystal sizes gives a systematic extrapolation to the thermodynamic limit.","The triclinic wollastonite application shows the method is practical for low-symmetry lattices: the computed per-ion electrostatic energies favor the triclinic phase over the cubic perovskite structure at the same density."],"supporting_citations":[{"why":"Established that Madelung sums require a shape-dependent boundary term and set the problem the paper solves.","marker":"[10, 11]"},{"why":"Defined exact shape and size for finite crystals and separated boundary from finite-size effects, the foundation this paper generalizes.","marker":"[14, 15]"},{"why":"Provided the divergence-theorem surface-integral approach for orthogonal lattices that the new parameterized integral method extends to triclinic cells.","marker":"[17]"},{"why":"Rayleigh's classic treatment of periodic obstacles supplies the $2\\pi$ and $2\\pi/3$ geometric factors recovered by the boundary-term formula.","marker":"[18]"},{"why":"OEIS exact Madelung constants for NaCl and ZnS used as benchmarks for the direct-sum validation.","marker":"[29, 30]"},{"why":"Wollastonite lattice parameters and atomic coordinates used in the triclinic application.","marker":"[45, 46]"}],"fun_headline_variants":["Analytic boundary term kills Madelung sum conditional convergence","Closed-form boundary term for Madelung constants in triclinic lattices","Madelung boundary term solved: exact sums for any crystal shape","Madelung sums made exact with analytic facet potentials","Direct-sum Madelung: boundary term reduced to charged parallelograms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The extrapolation to the bulk relies on the finite-size correction after subtracting the boundary term being exactly $C/(2p+1)^2$, so that fitting $C$ from two consecutive crystal sizes removes all remaining error; if other correction terms of comparable or larger order exist, the reported $10^{-10}$ agreement would be biased.","fun_headline_variants_meta":{"raw":{"variants":["Analytic boundary term kills Madelung sum conditional convergence","Closed-form boundary term for Madelung constants in triclinic lattices","Madelung boundary term solved: exact sums for any crystal shape","Madelung sums made exact with analytic facet potentials","Direct-sum Madelung: boundary term reduced to charged parallelograms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00078,"raw_usage":{"total_tokens":3448,"prompt_tokens":951,"completion_tokens":2497,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":2407}},"tokens_in":567,"tokens_out":2497,"duration_ms":15988,"temperature":1.0,"reasoning_tokens":2407,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:13:53.845655+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a fixed crystal shape, compute $D(p) = \\nu(p|s) - \\nu(p-1|s)$ after subtracting the analytic boundary term, and check whether $D(p)/[(2p+1)^{-2} - (2p-1)^{-2}]$ is constant as $p$ grows; if the ratio drifts or fails to converge, the assumed $(2p+1)^{-2}$ decay law is incomplete. Alternatively, compare the corrected direct sum against a high-accuracy Ewald result for a triclinic cell at $p=100$ and $p=200$; discrepancies above the claimed precision would falsify the correction scheme.","supporting_citations":[{"cited_title":"Infinite boundary terms of ewald sums and pair- wise interactions for electrostatics in bulk and at interfaces,","cited_arxiv_id":null,"evidence_quote":"Provided the divergence-theorem surface-integral approach for orthogonal lattices that the new parameterized integral method extends to triclinic cells."},{"cited_title":"Infinite boundary terms and pairwise interactions: A unified framework for periodic coulomb systems,","cited_arxiv_id":null,"evidence_quote":"Rayleigh's classic treatment of periodic obstacles supplies the $2\\pi$ and $2\\pi/3$ geometric factors recovered by the boundary-term formula."}],"review_version":1}