{"id":"ac872f4a-f3a0-4484-8ad7-a2029f0c9195","arxiv_id":"2608.05051","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"A single Fourier transform of a real-space pair displacement correlation map computes thermal diffuse scattering to all phonon orders, matching silicon data with R2 below 5%.","lead":"X-ray scattering from vibrating crystals includes a weak fuzzy signal, thermal diffuse scattering, that is normally hard to compute. This paper shows it can be obtained in one Fourier transform from the way atom pairs move together, matching silicon measurements to about 5% after fitting only a scale and background.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Folding of acoustic displacement correlations on the finite grid is the load-bearing numerical assumption; the silicon R^2 validation masks the affected near-Bragg region, so a direct convergence check is needed before the single-FFT implementation is called exact.","rationale":"Section 2.2–2.5 derive the all-order harmonic TDS formula correctly: the Gaussian average of the displacement phase factor, the Bragg/diffuse split, and the ΔPDF representation (Eqs. 8–21) are internally consistent, and the single-FFT structure is a real advance over truncated phonon expansions. The silicon comparison is honestly described: only scale and background are refined, and the residuals near Bragg peaks are acknowledged. The weakest point is not the harmonic derivation but the finite-grid implementation of the covariance. The inverse DFT on the same N-grid periodizes C_κκ'(R), and because C enters Eq. (15) through an exponential, the periodization is not equivalent to evaluating the exact intensity at the sampled h points. This error is concentrated near acoustic/Bragg positions, and the R^2 metric is computed after punch-and-fill removes those positions, so the headline 4.8% does not by itself certify the single-FFT exactness. The check proposed above—a finer or direct-summation covariance while keeping the same output grid—would settle the magnitude. I do not see a flaw in the mathematical core; the paper is strong and the concern is a numerical convergence assumption that the authors themselves flag. The appropriate verdict is therefore conditional acceptance pending that quantification, rather than outright rejection or an unconditional accept.","tokens_in":11674,"tokens_out":13260,"duration_ms":155211,"concrete_test":"Recompute the silicon benchmark with C_κκ'(R) obtained by inverse DFT on a 120^3 q-grid (or direct Fourier summation) so that the 30^3 real-space cell used for the ΔPDF is alias-free, and compare the resulting diffuse map with the published 30^3-grid result at the same h points, including near-Bragg points such as (0.03,0,0), (0.1,0,0), and (1.03,0,0), after re-fitting the same scale and background. If the local intensity differences stay below ~1% and R^2 changes by <0.5%, the folding error is negligible; if not, the exactness claim must be qualified and the R^2 validation should be repeated without excluding the regions where the folded correlations are largest.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section 2.6 explicitly acknowledges that the inverse DFT of the phonon covariance on the same N×N×N grid used for reciprocal-space sampling folds slow acoustic tails back into the real-space box: the correlation of pair (N+2,0,0) is aliased onto (2,0,0). This matters because Eq. (15) is exact only with the true infinite-lattice C_κκ'(R); the implemented C is the periodically wrapped version, and the wrapping enters through exp(4π^2 h^T C h), which is nonlinear in C, so a folded C is not equivalent to summing the folded intensity. For the Δq=1/30 grid the aliased terms come from separations of the order of 15–30 unit cells; acoustic correlations in silicon decay slowly and are not a priori negligible at those distances. The error is largest near the acoustic Bragg positions, and Section 4 removes exactly that region by punch-and-fill (r_p=3) before reporting R^2≈4.8%, so the validation does not independently constrain the folding error. The paper's claim that the folded contributions are the weakest and most distant is reasonable but is not quantified against a direct summation or a finer grid. This is the one place where the implemented single-FFT pipeline is approximate, and the central abstract claim that the method is exact rests, in practice, on this approximation being negligible.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a method for computing X-ray thermal diffuse scattering (TDS) from real-space displacement correlations using the three-dimensional difference pair distribution function (3D-ΔPDF). Within the harmonic approximation, the derivation expresses the full all-phonon-order diffuse intensity as a Fourier transform of a real-space ΔPDF volume built from atom-pair covariances. The implementation obtains displacement covariances from first-principles phonons via an inverse FFT and then assembles the ΔPDF from Gaussian atom-pair contributions; a single forward FFT yields the diffuse map. The method is validated against room-temperature silicon data measured at ID28 (R2≈4.8%, with only a scale and smooth background refined) and against a four-temperature silicon series from SNBL (R2≈9%). The paper further argues that the shared pair-correlation language with the Yell program allows joint treatment of thermal and static disorder.","tokens_in":12007,"tokens_out":8252,"duration_ms":83431,"significance":"If the claims hold, this is a genuinely useful advance: it gives an all-phonon-order TDS calculation at the cost of one FFT, avoids the truncation and sampling-noise problems of Laval-Born-James and special-displacement approaches, and is differentiable, opening the door to gradient-based refinement. The derivation is clearly presented, the code is available, and the validation uses an external experimental benchmark with no fitted physical parameters—only a scale and a background. The main reservation is that the numerical implementation's finite-grid folding approximation is acknowledged but not quantitatively assessed, and the reported residual is computed after masking the region where that approximation is expected to be largest.","major_comments":[{"comment":"The main text reports R²≈4.8% but defers the definition of R² and the punch-and-fill procedure to the Supplementary Information. Since this residual is the central quantitative validation, the main text should state exactly which voxels are included, how the Bragg region is filled, and how the R² is computed, so the reader can assess whether the near-Bragg masking also removes the region where the model error is largest. This is particularly important because the folding-error concern of Section 2.6 is dominated by that same region.","section":"4, Figure 3"},{"comment":"The abstract states that 'within the harmonic approximation the method is exact', which is true for the analytic equations derived in Section 2.3 but not for the implemented finite-grid pipeline, whose folding error is acknowledged but not shown to be negligible. I recommend either adding the quantitative convergence evidence requested above or qualifying the 'exact' claim to the infinite-lattice expressions, with the finite-grid implementation presented as a controlled approximation.","section":"Abstract and Section 2.6"}],"minor_comments":[{"comment":"The q=0 acoustic modes make Eq. (16) singular; the text states they are regularised to zero but does not spell out the limiting argument in the main body. A parenthetical explaining the rigid-translation cancellation, or a reference to the Supplementary Information at that point, would help.","section":"Eq. (16) and Section 3.1"},{"comment":"'6013 grid' should read '601³ grid'.","section":"Section 3.2"},{"comment":"The bottom-right residual panel uses a diverging blue/white/red map, but the caption does not explain the colour mapping beyond the range; a one-sentence explanation would improve readability.","section":"Figure 3 caption"},{"comment":"The tool name 'anaddb' should be typeset as 'Anaddb' or 'ABINIT's anaddb' for consistency with standard nomenclature.","section":"Section 3.2"},{"comment":"The statement that the ETH repository data will be available 'after short review' is vague; a persistent identifier or a release date would be more useful to readers and reviewers.","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid contribution with a sound analytic derivation and a strong validation concept. The central unresolved point is the quantitative impact of the finite-grid folding of acoustic displacement correlations, which the authors acknowledge but do not measure. I have asked for a direct convergence test because it is load-bearing for the exactness claim and because the current validation masks the affected region. This is a fixable issue and does not warrant rejection. The paper is well within the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a genuinely useful methods paper. It takes the 3D-ΔPDF construction from Yell and feeds it first-principles phonon covariances, so one FFT produces the full diffuse map with every multi-phonon order, no truncation and no sampling noise. The derivation is clean, each step states its approximation, and the silicon validation is honest: only a scale and a smooth background are refined, and the ID28 measurement lands at R²≈4.8%, with a four-temperature SNBL series reproducing the thermal growth on a single scale factor. For the crystallographic diffuse-scattering community this is a practical advance, and the stated goal of refining thermal and static disorder in the same pair-correlation language is sensible.\n\nThe main soft spot is the finite-grid folding in Section 2.6. The paper acknowledges that the inverse DFT of C(R) on the same N×N×N grid folds slow acoustic tails back into the real-space box, and argues the misplaced contributions are weakest and most distant. That argument is plausible but not quantified. The stress-test point deserves attention: the diffuse intensity enters through exp(4π²hᵀCh), which is nonlinear in C, so a folded C is not equivalent to summing folded intensities. And the reported R² is computed after punch-and-fill of the near-Bragg region, exactly where folding error would be largest, so the validation doesn't independently constrain it. I don't think this sinks the method; a finer-grid or direct-summation check would settle it, and it should be requested in review.\n\nOther issues are minor. The benchmark is a perfect crystal with no static disorder, so the joint-refinement payoff isn't demonstrated with data. The Supplementary Information is referenced repeatedly but absent from the arXiv posting, and the code/data are archived but not yet public. For a methods paper, those should be in place before acceptance. None of this undermines the central claim: within the harmonic approximation, the formula is exact, and the implementation's accuracy question is a numeric-convergence issue, not a conceptual one.\n\nWho is this for: anyone doing single-crystal diffuse scattering, especially lattice-dynamics and disorder-refinement groups. It deserves serious refereeing. My recommendation: send it out, and ask for the folding-convergence check, the SI, and the code/data before sign-off.","headline":"Useful all-order harmonic TDS method with a clean silicon validation; the finite-grid folding caveat is real but likely fixable, so send it to review.","tokens_in":12558,"tokens_out":3370,"would_cite":true,"duration_ms":34502,"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":"One Fourier transform computes all-order thermal diffuse scattering within the harmonic approximation, and silicon data match it to 4.8 percent.","keywords":["thermal diffuse scattering","phonons","three-dimensional difference pair distribution function","lattice dynamics","silicon","harmonic approximation","multi-phonon scattering","single-crystal X-ray diffraction"],"falsifier":"Compute the same silicon diffuse map by directly summing the real-space covariance contributions over lattice vectors out to a large cutoff and compare with the single-FFT map on, say, a 30×30×30 grid; if the two disagree near the Bragg peaks by an amount comparable to the 4.8% residual, the folding assumption fails.","tokens_in":11505,"feed_emoji":"🔬","tokens_out":16679,"duration_ms":143264,"temperature":0.7,"pith_summary":"The paper aims to establish that X-ray thermal diffuse scattering (TDS) can be computed exactly within the harmonic approximation, to all phonon orders, from real-space displacement correlations rather than by truncating a phonon expansion. It does this by building a three-dimensional difference pair distribution function (3D-ΔPDF) from phonon-derived displacement covariances; one Fourier transform of that map then delivers the diffuse intensity over the whole sampled reciprocal-space volume. The significance is practical and conceptual: standard calculations stop at one- and two-phonon terms because each further order costs another Brillouin-zone integral, while this method includes every order automatically with no sampling noise. The authors demonstrate it on a silicon single crystal, refining only a scale and a background, and reproduce the measured diffuse scattering to $R^2\\approx4.8\\%$ at room temperature, with a 200–350 K series following the same model to about $9\\%$ using a single temperature-independent scale. Because the same pair-correlation language is already used to model static correlated disorder, the method would let thermal and static disorder be handled together.","feed_headline":"One FFT computes all-order thermal diffuse scattering","feed_subtitle":"Phonon covariances become 3D pair correlations; silicon data match to under 5 percent with no structural refinement.","key_machinery":"The central object is the three-dimensional difference pair distribution function (3D-ΔPDF), assembled from the phonon-derived reciprocal-space kernel $\\gamma_{\\kappa\\alpha,\\kappa'\\beta}(\\mathbf{q})=(M_\\kappa M_{\\kappa'})^{-1/2}\\sum_\\nu A_\\nu(\\mathbf{q})e^*_{\\kappa\\alpha,\\nu}(\\mathbf{q})e_{\\kappa'\\beta,\\nu}(\\mathbf{q})$ with thermal amplitude $A_\\nu=(2\\langle n_\\nu\\rangle+1)/\\omega_\\nu$. An inverse DFT on the same $N\\times N\\times N$ q-grid as the intensity sampling converts this kernel to the real-space covariance $C_{\\kappa\\kappa'}(R)$, and for every atom pair the assembly lays down two centred Gaussians on the interatomic vector $R+x_\\kappa-x_{\\kappa'}$: one with the correlated relative-displacement covariance $\\Sigma^R_{\\kappa\\kappa'}=U_\\kappa+U_{\\kappa'}-2C_{\\kappa\\kappa'}(R)$, one with $U_\\kappa+U_{\\kappa'}$. Their difference is that pair's ΔPDF contribution, and a single forward FFT of the summed volume, with the form-factor product applied on transform, yields the all-order diffuse map. The reason the method is exact in the harmonic approximation is that the exponential of the covariance is kept whole, so the assembly never chooses a phonon order; the FFT pair shares one grid, so no interpolation is needed.","core_discovery":"The central claim is that all multiphonon contributions to X-ray thermal diffuse scattering are encoded in the real-space displacement covariance $C_{\\kappa\\kappa'}(R)=\\langle u_{0\\kappa}u_{R\\kappa'}^T\\rangle$, so the diffuse intensity is exactly (inside the harmonic-Gaussian description) the Fourier transform of a 3D-ΔPDF: $I_{\\mathrm{diff}}(\\mathbf{h})=N\\mathcal{F}[\\Delta\\mathrm{PDF}(\\mathbf{r})](\\mathbf{h})$. Each atom pair contributes two Gaussians on its interatomic vector, one with the correlated covariance $\\Sigma^R_{\\kappa\\kappa'}=U_\\kappa+U_{\\kappa'}-2C_{\\kappa\\kappa'}(R)$ and one with the independent-motion covariance $U_\\kappa+U_{\\kappa'}$; their difference is the diffuse signal. The intensity formula keeps the exponential $\\exp(4\\pi^2\\mathbf{h}^TC_{\\kappa\\kappa'}(R)\\mathbf{h})$ unexpanded, so no phonon order is dropped and no supercell averaging noise enters. Tests against measured silicon data support the claim: with first-principles phonons used unadjusted and only scale and background refined, the room-temperature diffuse intensity is reproduced to $R^2\\approx4.8\\%$, and a four-temperature series reproduces the thermal growth with one scale to $R^2\\approx9\\%$.","pith_inferences":["If the periodic-folding error is as small as argued, the pipeline should transfer to anisotropic and lower-symmetry crystals with cost set mainly by the output grid; testing such a crystal would confirm the approach is not silicon-specific.","The four-temperature series with one scale factor suggests that measured TDS at several temperatures could serve as a sensitive benchmark for phonon frequencies and anharmonic shifts, since the model alone must reproduce the thermal growth; the paper does not explore this use.","Because the calculation is differentiable and noise-free, it could serve as a forward model in reverse Monte Carlo or machine-learning fits of disorder, where expensive noisy supercell averages are currently the bottleneck.","A useful benchmark would be comparing this all-order result against one- and two-phonon approximations to map where truncation errors appear in scattering vector and temperature; the paper does not include such a map."],"forward_implications":["The standard one-phonon and two-phonon truncation of TDS is avoided altogether, since every multiphonon order is carried by the unexpanded covariance exponential.","A full diffuse map over large reciprocal-space volumes comes from one fast Fourier transform, so the cost is governed by the FFT grid; the silicon benchmark finishes in under ten minutes.","Because thermal and static correlated disorder are expressed in the same pair-correlation representation, crystals that show both can be modeled and ultimately refined together.","Every step in the construction is differentiable, which opens the way to gradient-based refinement of phonon or disorder parameters against measured diffuse volumes.","The output is a real-space 3D-ΔPDF, so the calculation yields interpretable per-pair displacement correlations, not just reciprocal-space intensities."],"supporting_citations":[{"why":"Establishes the harmonic-Gaussian average of the displacement phase factor, the identity that turns the intensity into an exponential of the covariance.","marker":"Born, 1942"},{"why":"Supplies the phonon-eigenvector expression for the displacement covariance tensor and the one/two-phonon method the new approach replaces.","marker":"Xu, 2010"},{"why":"Introduces the difference pair distribution function space in which the new all-order TDS construction is carried out.","marker":"Weber & Simonov, 2012"},{"why":"Provides the per-pair Gaussian assembly and single-Fourier-transform evaluation used to build the 3D-ΔPDF and its intensity.","marker":"Simonov et al., 2014"},{"why":"The closest prior all-order method, special-displacement supercells; the paper compares its sampling noise and cost against the new approach.","marker":"Zacharias et al., 2021b"},{"why":"Provides the first-principles phonon frequencies and eigenvectors (density-functional perturbation theory) that feed the silicon covariance map.","marker":"Gonze et al., 2020"}],"fun_headline_variants":["Single FFT gives exact all-order multiphonon diffuse scattering","All phonon orders from one Fourier transform of displacement correlations","Covariance to intensity: one FFT for exact multiphonon scattering","Silicon diffuse scattering matched to 5% with zero structural fits","Thermal and static disorder unified by pair-correlation language"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that periodically wrapping the slowly decaying acoustic displacement correlations into the finite real-space box, instead of summing them directly, leaves an error below the reported residual; the paper argues this but does not quantify it against a direct sum.","fun_headline_variants_meta":{"raw":{"variants":["Single FFT gives exact all-order multiphonon diffuse scattering","All phonon orders from one Fourier transform of displacement correlations","Covariance to intensity: one FFT for exact multiphonon scattering","Silicon diffuse scattering matched to 5% with zero structural fits","Thermal and static disorder unified by pair-correlation language"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000729,"raw_usage":{"total_tokens":3257,"prompt_tokens":933,"completion_tokens":2324,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":549,"completion_tokens_details":{"reasoning_tokens":2236}},"tokens_in":549,"tokens_out":2324,"duration_ms":15031,"temperature":1.0,"reasoning_tokens":2236,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T10:29:14.273362+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the same silicon diffuse map by directly summing the real-space covariance contributions over lattice vectors out to a large cutoff and compare with the single-FFT map on, say, a 30×30×30 grid; if the two disagree near the Bragg peaks by an amount comparable to the 4.8% residual, the folding assumption fails.","supporting_citations":[{"cited_title":"(1942).Reports on Progress in Physics,9(1), 294–333","cited_arxiv_id":null,"evidence_quote":"Establishes the harmonic-Gaussian average of the displacement phase factor, the identity that turns the intensity into an exponential of the covariance."}],"review_version":1}