Pith. sign in

REVIEW 4 major objections 5 minor 54 references

Local lattice dynamics of hcp zinc from EXAFS and machine-learning interatomic potentials

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read By fitting temperature-dependent EXAFS with reverse Monte Carlo simulations, this paper extracts shell-resolved mean-square relative displacements for eight coordination shells of hcp zinc and shows that fine-tuning the CHGNet…

desk verdict A useful new shell-resolved MSRD dataset and an independent CHGNet fine-tuning check, but the central anisotropy ratio hinges on an unvalidated Gaussian decomposition and missing error bars. read the letter →

arxiv 2608.07081 v1 pith:QRBJABZN submitted 2026-08-07 cond-mat.mtrl-sci

Vitalijs Dimitrijevs , Pjotrs Žguns , Inga Pudza , Aleksandr Kalinko , Alexei Kuzmin This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords hcpzincEXAFSreverseMonteCarlomean-squarerelativedisplacementlatticedynamicsmachine-learninginteratomicpotentialCHGNetanisotropy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that temperature-dependent EXAFS, analysed with reverse Monte Carlo simulations, can resolve how thermal vibrations differ between the in-plane and out-of-plane directions in hexagonal close-packed zinc, and that the resulting shell-by-shell measures provide a stringent test for machine-learning interatomic potentials. The authors extract mean-square relative displacements for eight coordination shells, fit them with a correlated Einstein model, and obtain effective force constants that are roughly two to three times larger for in-plane (A-type) than for out-of-plane (B-type) neighbours. They also show that the ratio of the second-shell to first-shell MSRDs tracks the out-of-plane/in-plane displacement ratio measured by single-crystal diffraction, confirming the anisotropy with a local probe. Finally, molecular dynamics with the CHGNet universal machine-learning potential overestimates thermal disorder, while a version fine-tuned on a small set of DFT relaxation trajectories brings the simulated EXAFS spectrum and radial distribution function close to experiment. The payoff is a workflow in which EXAFS, which sees around specific atoms, can validate and refine machine-learning potentials for anisotropic metals.

What carries the argument

The central machinery is a reverse Monte Carlo loop over a periodically repeated supercell: an evolutionary algorithm moves zinc atoms while a real-space multiple-scattering code computes the configuration-averaged EXAFS spectrum, and the fit is scored in both wavenumber and distance space against measured spectra at each temperature. The final atomic configurations give zinc–zinc radial distribution functions, which are decomposed into Gaussian peaks whose variances are the shell-resolved mean-square relative displacements; fitting their temperature dependence to the correlated Einstein model yields effective force constants. For the potential validation, the same multiple-scattering calculation is applied to snapshots from constant-temperature molecular dynamics driven by the CHGNet universal machine-learning interatomic potential, so the experimental EXAFS spectrum itself becomes the arbiter of whether the potential produces the right thermal disorder.

What would settle it

A decisive check would be to rerun the 300 K reverse Monte Carlo fit with a non-Gaussian or two-shell-correlated decomposition of the first two overlapping peaks and see whether the $\mathrm{MSRD}_2/\mathrm{MSRD}_1$ anisotropy moves outside the quoted errors; a complementary check is to compare the shell-resolved MSRDs with values from inelastic neutron scattering or first-principles phonon calculations.

Watch

Extended reading notes

Core claim

At 10–300 K, the Zn K-edge EXAFS of hcp zinc was modelled by reverse Monte Carlo in a 288-atom supercell, with configuration-averaged spectra computed by real-space multiple scattering over all absorbing atoms. From the final configurations, the Zn–Zn radial distribution function was decomposed into Gaussian peaks, and the variance of each peak gives the mean-square relative displacement (MSRD) for that coordination shell. Fitting each $\mathrm{MSRD}(T)$ to the correlated Einstein model gives effective force constants: $34\pm1$ N/m for the six in-plane nearest neighbours at 2.66 Å, versus $12\pm1$ N/m for the six second-shell neighbours at 2.91 Å, with the ratio $\mathrm{MSRD}_2/\mathrm{MSRD}_1$ rising with temperature in line with the diffraction-derived $U_{33}/U_{11}$. The paper further shows that the original CHGNet universal machine-learning potential, used in 300 K NVT molecular dynamics, softens the lattice and broadens the RDF, damping the computed EXAFS; fine-tuning CHGNet on roughly 400 DFT frames of strained and displaced zinc supercells substantially corrects this, bringing both EXAFS and RDF into agreement with the RMC reference and experiment.

Load-bearing premise

The load-bearing premise is that decomposing the total zinc–zinc radial distribution function into independent Gaussian peaks, one per coordination shell, does not bias the extracted mean-square relative displacements, even when the first shell (2.66 Å) and the second shell (2.91 Å) overlap at high temperature.

Editorial extensions

If this is right

  • The eight-shell MSRD dataset gives a quantitative, shell-resolved picture of thermal disorder in a strongly anisotropic metal, not just the usual first-shell picture.
  • Effective force constants from the correlated Einstein model show which crystallographic directions are stiff: in-plane A-type neighbours ($\kappa \approx 34$ N/m at 2.66 Å) resist thermal motion much more than out-of-plane B-type neighbours ($\kappa \approx 12$ N/m at 2.91 Å).
  • The $\mathrm{MSRD}_2/\mathrm{MSRD}_1$ anisotropy ratio reproduces the temperature trend of the diffraction-measured $U_{33}/U_{11}$, strengthening the case that local EXAFS and global diffraction are measuring the same anisotropic displacements.
  • The original CHGNet potential's softening is visible as over-broad RDF peaks and a damped EXAFS spectrum; fine-tuning on a few hundred DFT frames repairs most of this, so EXAFS can act as an experimental check on machine-learning potentials.
  • Using the RMC reference RDF, the paper establishes a workflow where EXAFS data, not just energies and forces, decide whether a machine-learning potential is physically trustworthy.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not test cadmium, but the same RMC/EXAFS pipeline could be applied to Cd, whose even larger $c/a \approx 1.89$ should produce a stronger MSRD anisotropy; whether current machine-learning potentials capture that would be a direct test of the method's resolving power.
  • A natural next step is to use the RMC-derived radial distribution function at 300 K as a training target for fine-tuning other universal machine-learning potentials, since it encodes thermal disorder directly rather than through energies and forces.
  • The reported force-constant pattern implies a minimal two-parameter anisotropic Einstein model for zinc's local dynamics; one consequence the authors do not pursue is that this model should reproduce diffraction Debye–Waller factors at all temperatures, not just their ratio.
  • The unresolved $c$-axis C-type shell could likely be recovered with a longer simulation or symmetry-restrained RMC sampling, completing the full anisotropy map of the first coordination shell.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper studies the local lattice dynamics of hcp zinc by combining temperature-dependent (10–300 K) Zn K-edge EXAFS with reverse Monte Carlo/evolutionary-algorithm (RMC/EA) simulations and molecular dynamics with a machine-learning interatomic potential (CHGNet). The authors extract mean-square relative displacements (MSRDs) for eight coordination shells, fit their temperature dependence with a correlated Einstein model to obtain effective force constants, and report a pronounced anisotropy between in-plane (A-type) and out-of-plane (B-type) interactions. They also compare the MSRD2/MSRD1 ratio with the diffraction-based U33/U11 ratio. In the second part, they fine-tune the CHGNet universal machine-learning potential on DFT relaxation trajectories and compare simulated EXAFS and RDFs with experiment at 300 K, finding that fine-tuning substantially improves agreement relative to the vanilla potential.

Significance. If the central methodological concern is resolved, the paper offers a valuable shell-resolved view of anisotropic thermal motion in a strongly anisotropic hcp metal, together with a concrete demonstration that fine-tuning a universal machine-learning potential on DFT data improves its fidelity to EXAFS measurements. The work includes several commendable internal checks: the RMC/EA analysis explicitly treats multiple-scattering paths up to fourth order and is repeated with k3 weighting at 10 and 300 K, and the MD-EXAFS comparison is a genuine independent test because the potential was trained on DFT energies/forces/stresses rather than on the experimental EXAFS. The paper also reports quantitative training metrics (MAEs of 1 meV/atom, 22 meV/Å, 0.07 GPa) for the fine-tuned potential, which strengthens the credibility of the MLIP part. However, the absence of error bars on all extracted MSRDs and the reliance on a Gaussian decomposition of overlapping coordination shells leave the statistical significance of the central anisotropy claim unverified.

major comments (4)
  1. [Section 2.2 and Section 3, Figs. 4 and 5(a)] The MSRDs for the first and second coordination shells are obtained by decomposing the total RDF into independent Gaussian peaks, but at 300 K the first (2.66 Å) and second (2.91 Å) shells form a single heavily overlapping peak (Fig. 4). A two-Gaussian fit to such a composite peak has a known trade-off between the widths and areas of the two components, so MSRD1 and MSRD2 are not independently determined by the RDF alone. The k3-weighting cross-check described in Section 2.2 changes only the EXAFS weighting, not the Gaussian decomposition model, so it cannot detect this bias. Please provide a direct test: compute per-shell MSRDs from the RMC configurations by assigning atoms to coordination shells from the initial ideal hcp structure (or an equivalent neighbor-based assignment) and compare those values with the Gaussian-decomposition results. If the direct per-shell MSRDs confirm the anisotropy, report them; otherwise, the central anisotropy claim requires qualification.
  2. [Section 2.2 and Figs. 5–6] No error bars are shown for the MSRDs in Fig. 5(a), for the MSRD-versus-distance curves in Fig. 6(a), or for the MSRD2/MSRD1 ratio in Fig. 6(b), although the RMC/EA simulations were repeated ten times at each temperature. Without a measure of run-to-run or fit uncertainty, the statistical significance of the reported anisotropy (approximately 1.2–1.6 times) and of the Einstein-model force constants in Table 1 cannot be assessed. Please add error bars derived from the ten independent RMC runs (or another defensible uncertainty estimate) to all reported MSRDs and ratios, and state explicitly how the uncertainties of the force constants in Table 1 were obtained.
  3. [Section 3, Fig. 6(b)] The claim that the MSRD2/MSRD1 ratio 'closely matches' the diffraction-based U33/U11 ratio is based on a qualitative visual comparison of two curves without uncertainty bands. Since MSRD includes correlation between atoms while MSD is a single-atom displacement, the two ratios are not expected to be equal in general; the text itself later states that the ratios indicate displacements only 'approximately 1.2–1.6 times larger,' which is a weaker statement than 'closely matches.' Please provide a quantitative comparison (e.g., a confidence interval for the difference, or a reduced chi-square value) and clarify the expected relationship between MSRD2/MSRD1 and U33/U11, or soften the wording accordingly.
  4. [Section 2.3 and Fig. 7] The improvement of the fine-tuned CHGNet over the vanilla potential is demonstrated only through visual comparison of the EXAFS spectra and RDFs in Fig. 7. Since the abstract and conclusions state that fine-tuning 'substantially improves' agreement with experiment, please include a quantitative metric for both the EXAFS (e.g., R-factor or RMSD in k-space) and the RDF (e.g., integrated absolute difference over the plotted range) so that the degree of improvement is numerically grounded.
minor comments (5)
  1. [Abstract] The phrase 'reverse Monte Carlo method enable' should be 'reverse Monte Carlo method enables'.
  2. [Section 2.2] The sentence 'demonstrating the robustness of the results with respect to the choice of EXAFS weighting and and showing that they are not biased by the reduced contribution of the high-k region' contains a duplicated 'and'.
  3. [Section 3] The sentence beginning 'The anisotropy of thermal vibrations is clearly evidenced by the difference between the out-off-plane MSD' contains a grammatical error and the typo 'out-off-plane'; the intended phrase is 'out-of-plane'.
  4. [Section 2.2 and Section 3] The text refers to the Supplementary Material for details of the Gaussian decomposition of the RDFs, but no supplementary material is included with the arXiv submission, so the decomposition procedure (initial parameters, fitting ranges, constraints) cannot be checked. Please provide the supplement or include the relevant details in an appendix.
  5. [Section 2.3] The abbreviation 'uMLIPs' appears with inconsistent capitalization (e.g., 'uMLIPS' in the conclusions); please use one consistent form.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: MSRDs and force constants are fitted outputs, and the CHGNet comparison is an independent DFT-trained test against experimental EXAFS.

full rationale

The derivation chain is not circular. The central MSRDs are obtained from RMC/EA fits directly to experimental Zn K-edge EXAFS (Section 2.2, Fig. 3), and the MSRD temperature dependences are then fitted to the correlated Einstein model (Eq. 1) to report effective force constants; these force constants are explicitly fitted outputs rather than predictions derived from the model itself (Table 1). The anisotropy claim is benchmarked against externally measured single-crystal diffraction U33/U11 from [3], providing an independent cross-check (Fig. 6(b)). The CHGNet comparison is also an independent test: the potential is fine-tuned only on DFT energies, forces, and stresses from relaxation trajectories (Section 2.3), not on the experimental EXAFS or on the RMC-derived RDF. The subsequent MD-EXAFS and MD-RDF comparisons at 300 K (Fig. 7) therefore constitute a genuine transfer test of the tuned MLIP against experiment. The Gaussian decomposition of the RDF is a post-processing model assumption; any possible bias from overlapping first/second shells is a statistical accuracy concern, not a circularity, and the paper includes a k3-weighting robustness check (Section 2.2, Figs. S2-S5). Self-citations to the group's EvAX RMC code [23,24], wavelet tools [31], MD-EXAFS approach [44,45], and prior CHGNet benchmarking [53] are methodological references and are not used to justify the central numerical results. No equation is defined in terms of a target result, and no fitted parameter is presented as an independent prediction. The few self-citations are not load-bearing, so the circularity score is minimal.

Assumptions & free parameters 8 free parameters · 5 assumptions · 0 invented entities

The manuscript introduces no new physical entities. Its quantitative claims rest on model-based extraction of MSRDs (RMC/EA plus Gaussian decomposition plus Einstein model) and on a small DFT fine-tuning set for CHGNet. The eight per-shell Einstein force constants are fitted parameters rather than ab initio predictions.

free parameters (8)
  • Effective force constant kappa of shell 1 (A-type, r=2.66 Å) = 34 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model; central to the anisotropy claim.
  • Effective force constant kappa of shell 2 (B-type, r=2.91 Å) = 12 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model; central to the anisotropy claim.
  • Effective force constant kappa of shell 3 (B-type, r=3.94 Å) = 16 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
  • Effective force constant kappa of shell 4 (A-type, r=4.61 Å) = 24 ± 2 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
  • Effective force constant kappa of shell 5 (B-type, r=4.75 Å) = 16 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
  • Effective force constant kappa of shell 7 (A-type, r=5.32 Å) = 27 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
  • Effective force constant kappa of shell 8 (B-type, r=5.61 Å) = 12 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
  • Effective force constant kappa of shell 9 (B-type, r=6.06 Å) = 21 ± 1 N/m
    Fitted to the extracted MSRD(T) with the correlated Einstein model.
assumptions (5)
  • domain assumption The correlated Einstein model (Eq. 1) with a single Einstein frequency per shell adequately represents the temperature dependence of each shell's MSRD.
    Used to convert MSRD(T) into effective force constants kappa; if multiple frequencies or strong anharmonicity matter, the kappa values are only effective averages.
  • domain assumption RMC/EA with FEFF8.5L multiple scattering up to fourth order and Morlet wavelet transform fitting yields atomic configurations whose RDFs faithfully represent the true thermal disorder.
    The MSRDs are extracted from RMC configurations; the inversion's reliability is central to the claim of accurate MSRDs.
  • domain assumption Gaussian decomposition of the RDF into independent Gaussian peaks, one per coordination shell, accurately separates overlapping shells, particularly the first (2.66 Å) and second (2.91 Å) shells.
    MSRD values are the variances of these Gaussians; peak overlap and anharmonic peak shapes could bias them.
  • domain assumption The experimental lattice parameters from Nuss et al. (ref 3), used to construct the RMC supercells, are accurate across 10-300 K.
    The simulation box dimensions are fixed to these values; errors would propagate into MSRDs and force constants.
  • domain assumption PBE-DFT relaxation trajectories are a sufficient reference for fine-tuning CHGNet to correct its softening in zinc at 300 K.
    The fine-tuned potential inherits any PBE inaccuracies, and only 400 DFT frames are used for training.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Local lattice dynamics of hcp zinc from EXAFS and machine-learning interatomic potentials." pith.science (2026). https://pith.science/paper/QRBJABZN

@misc{pith2026260807081,
  author       = {Pith},
  title        = {Pith review of: Local lattice dynamics of hcp zinc from EXAFS and machine-learning interatomic potentials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QRBJABZN}},
  note         = {Machine review of arXiv:2608.07081}
}
read the original abstract

The lattice dynamics of hexagonal close-packed (hcp) zinc, a prototypical anisotropic metal, is studied using temperature-dependent Zn K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy combined with atomistic simulations. The reverse Monte Carlo method enable the extraction of mean-square relative displacements (MSRDs) for eight coordination shells, providing a shell-resolved description of thermal motion. The MSRD temperature dependence, analysed using the correlated Einstein model, yields effective interatomic force constants and reveals pronounced anisotropy between in-plane and out-of-plane interactions. This anisotropy is further quantified by the ratio of MSRDs for the first and second coordination shells, which closely matches the anisotropic displacement parameters from diffraction experiments. Molecular dynamics simulations using the CHGNet universal machine-learning interatomic potential show that the original model overestimates thermal disorder, while a fine-tuned version substantially improves agreement with experimental EXAFS spectrum and radial distribution function. Overall, EXAFS-informed analysis is effective for validating and refining machine-learning interatomic potentials.

Figures

Figures reproduced from arXiv: 2608.07081 by the authors.

Figure 1
Figure 1. c a B-type C-type A-type [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. 0 2 4 6 8 10 12 14 16 18 −2 −1 0 1 2 3 10 K 50 K 100 K 150 K 200 K 250 K 300 K EXAFS (k)k 2 (Å − 2 ) Wavenumber k (Å−1 ) hcp Zn Zn K-edge (a) 0 2 4 6 8 10 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 FT (k)k 2 (Å − 3 ) Distance R (Å) hcp Zn Zn K-edge 10 K 50 K 100 K 150 K 200 K 250 K 300 K (b) [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Experimental (solid circles) and RMC/EA-simulated (solid lines) Zn K-edge EXAFS spectra 𝜒(𝑘)𝑘 2 and their Fourier transforms (FTs) for hcp zinc at four selected temperatures (10, 100, 200, and 300 K). Both the moduli and imaginary parts of the FTs are shown [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: 23456 0 20 40 60 80 RDF gZn-Zn(r) (atoms/Å) Distance r (Å) hcp Zn Zn K-edge 10 K 50 K 100 K 150 K 200 K 250 K 300 K A A B C B B B B A [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 6
Figure 6. Figure 6: 0 50 100 150 200 250 300 1.0 1.5 2.0 2.5 3.0 3.5 Anisotropy Temperature (K) hcp Zn U33/U11 MSRD2/MSRD1 (b) 23456 0.00 0.01 0.02 0.03 0.04 0.05 10 K 50 K 100 K 150 K 200 K 250 K 300 K MSRD  2 (Å 2 ) Distance r (Å) (a) hcp Zn A B B AB A B B [PITH_FULL_IMAGE:figures/ful…
Figure 7
Figure 7. Figure 7: 23456 0 20 40 60 RDF gZn-Zn(r) (atoms/Å) Distance r (Å) hcp Zn Zn K-edge 300 K (b) A B B A C A B B B RMC/EA CHGNet (tuned), Vexp CHGNet (tuned), VPBE CHGNet (vanilla), Vexp [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages

  1. [1]

    E. R. Jette, F. Foote, Precision determination of lattice constants, J. Chem. Phys. 3 (1935) 605–616

    work page 1935

  2. [2]

    R.W.G.Wyckoff,CrystalStructures,volume1,IntersciencePublish- ers, New York, 1963

    work page 1963

  3. [3]

    J. Nuss, U. Wedig, A. Kirfel, M. Jansen, The structural anomaly of zinc:Evolutionoflatticeconstantsandparametersofthermalmotion in the temperature range of 40 to 500 K, Z. fur Anorg. Allg. Chem. 636 (2010) 309–313

    work page 2010

  4. [4]

    T. H. K. Barron, R. W. Munn, Mean-square atomic displacements in zinc, Acta Crystallogr. 22 (1967) 170–173. V. Dimitrijevs et al.:Preprint submitted to ElsevierPage 7 of 9 Local lattice dynamics of hcp zinc from EXAFS and machine-learning interatomic potentials

    work page 1967

  5. [5]

    E. F. Skelton, J. L. Katz, Examination of the thermal variation of the mean square atomic displacements in zinc and evaluation of the associated Debye temperature, Phys. Rev. 171 (1968) 801–808

    work page 1968

  6. [6]

    Albanese, A

    G. Albanese, A. Deriu, C. Ghezzi, Anharmonic contributions to the Debye–Wallerfactorforzinc,ActaCrystallogr.A32(1976)904–909

    work page 1976

  7. [7]

    Tomaschitz, Thermodynamics of lattice vibrations in non-cubic crystals: the zinc structure revisited, Acta Crystallogr

    R. Tomaschitz, Thermodynamics of lattice vibrations in non-cubic crystals: the zinc structure revisited, Acta Crystallogr. A 77 (2021) 420–432

    work page 2021

  8. [8]

    M.Merisalo,F.K.Larsen, Anharmonicityoflatticevibrationsinzinc byelasticneutronscattering, ActaCrystallogr.A33(1977)351–354

    work page 1977

Show all 54 references
  1. [9]

    E.Rossmanith, X-raymeasurementoftheroot-mean-squaredisplace- mentoftheatomsincadmiumsinglecrystals, ActaCrystallogr.A34 (1978) 497–500

    1978

  2. [10]

    Merisalo, F

    M. Merisalo, F. K. Larsen, On antisymmetric atomic vibrations in zinc, Acta Crystallogr. A 35 (1979) 325–327

    1979

  3. [11]

    P. D. Pathak, R. J. Desai, Thermal properties of some H.C.P. metals. II. Debye-Waller factors of zinc, cadmium and magnesium, Phys. Status Solidi (a) 64 (1981) 741–745

    1981

  4. [12]

    L.-M. Peng, G. Ren, S. L. Dudarev, M. J. Whelan, Debye-Waller factors and absorptive scattering factors of elemental crystals, Acta Crystallogr. A 52 (1996) 456–470

    1996

  5. [13]

    N. V. Hung, T. S. Tien, L. H. Hung, R. R. Frahm, Anharmonic effective potential, local force constant and EXAFS of hcp crystals: theoryandcomparisontoexperiment, Int.J.Mod.Phys.B22(2008) 5155–5166

    2008

  6. [14]

    A. S. Masadeh, M. T. M. Shatnawi, G. Adawi, Y. Ren, Total- scatteringpair-distributionfunctionanalysisofzincfromhigh-energy synchrotron data, Mod. Phys. Lett. B 33 (2019) 1950410

    2019

  7. [15]

    A. S. Masadeh, M. T. M. Shatnawi, Z. Y. Abu Waar, G. M. Alzoubi, Y.Ren, Realspacestudyoflocalbondingforzincstructurebasedon temperature-dependent x-ray pair distribution function analysis, AIP Adv. 12 (2022) 095313

    2022

  8. [16]

    K. N. Trueblood, H.-B. Bürgi, H. Burzlaff, J. D. Dunitz, C. M. Gramaccioli, H. H. Schulz, U. Shmueli, S. C. Abrahams, Atomic dispacement parameter nomenclature. Report of a subcommittee on atomic displacement parameter nomenclature, Acta Crystallogr. A 52 (1996) 770–781

    1996

  9. [17]

    Bunker, Application of the ratio method of EXAFS analysis to disordered systems, Nucl

    G. Bunker, Application of the ratio method of EXAFS analysis to disordered systems, Nucl. Instrum. Methods 207 (1983) 437–444

    1983

  10. [18]

    Fornasini, F

    P. Fornasini, F. Monti, A. Sanson, On the cumulant analysis of EXAFS in crystalline solids, J. Synchrotron Rad. 8 (2001) 1214– 1220

    2001

  11. [19]

    Mustre de Leon, S

    J. Mustre de Leon, S. D. Conradson, I. Batistić, A. R. Bishop, I. D. Raistrick,M.C.Aronson,F.H.Garzon,Axialoxygen-centeredlattice instabilitiesinYB 2Cu3O7:Anapplicationoftheanalysisofextended x-ray-absorption fine structure in anharmonic systems, Phys. Rev. B 45 (1992) 2447–2457

    1992

  12. [20]

    Welter, R

    E. Welter, R. Chernikov, M. Herrmann, R. Nemausat, A beamline for bulk sample x-ray absorption spectroscopy at the high brilliance storage ring PETRA III, AIP Conf. Proc. 2054 (2019) 040002

    2019

  13. [21]

    Kalinko, XAESA v0.07, 2023

    A. Kalinko, XAESA v0.07, 2023. URL:https://gitlab.desy.de/ aleksandr.kalinko/xaesa

    2023

  14. [22]

    A.Kuzmin,J.Chaboy, EXAFSandXANESanalysisofoxidesatthe nanoscale, IUCrJ 1 (2014) 571–589

    2014

  15. [23]

    J.Timoshenko,A.Kuzmin,J.Purans, ReverseMonteCarlomodeling of thermal disorder in crystalline materials from EXAFS spectra, Comput. Phys. Commun. 183 (2012) 1237–1245

    2012

  16. [24]

    Timoshenko, A

    J. Timoshenko, A. Kuzmin, J. Purans, EXAFS study of hydrogen intercalation into ReO3 using the evolutionary algorithm, J. Phys.: Condens. Matter 26 (2014) 055401

    2014

  17. [25]

    J. J. Rehr, R. C. Albers, Theoretical approaches to X-ray absorption fine structure, Rev. Mod. Phys. 72 (2000) 621–654

    2000

  18. [26]

    A. L. Ankudinov, B. Ravel, J. J. Rehr, S. D. Conradson, Real-space multiple-scattering calculation and interpretation of x-ray-absorption near-edge structure, Phys. Rev. B 58 (1998) 7565–7576

    1998

  19. [27]

    Hedin, B

    L. Hedin, B. I. Lundqvist, Explicit local exchange-correlation poten- tials, J. Phys. C: Solid State Phys. 4 (1971) 2064

    1971

  20. [28]

    S. I. Zabinsky, J. J. Rehr, A. Ankudinov, R. C. Albers, M. J. Eller, Multiple-scattering calculations of X-ray-absorption spectra, Phys. Rev. B 52 (1995) 2995–3009

    1995

  21. [29]

    Ravel, A practical introduction to multiple scattering theory, J

    B. Ravel, A practical introduction to multiple scattering theory, J. Alloys Compd. 401 (2005) 118–126

    2005

  22. [30]

    J. J. Rehr, J. J. Kas, M. P. Prange, A. P. Sorini, Y. Takimoto, F. Vila, Ab initio theory and calculations of X-ray spectra, C. R. Phys. 10 (2009) 548–559

    2009

  23. [31]

    J.Timoshenko,A.Kuzmin, WaveletdataanalysisofEXAFSspectra, Comput. Phys. Commun. 180 (2009) 920–925

    2009

  24. [32]

    Sevillano, H

    E. Sevillano, H. Meuth, J. J. Rehr, Extended X-ray absorption fine structure Debye-Waller factors. I. Monatomic crystals, Phys. Rev. B 20 (1979) 4908–4911

    1979

  25. [33]

    A. H. Larsen, J. J. Mortensen, J. Blomqvist, I. E. Castelli, R. Chris- tensen, M. Dułak, J. Friis, M. N. Groves, B. Hammer, C. Hargus, E. D. Hermes, P. C. Jennings, P. B. Jensen, J. Kermode, J. R. Kitchin, E. L. Kolsbjerg, J. Kubal, K. Kaasbjerg, S. Lysgaard, J. B. Maronsson, ...

    2017

  26. [34]

    B.Deng,P.Zhong,K.Jun,J.Riebesell,K.Han,C.J.Bartel,G.Ceder, Chgnet as a pretrained universal neural network potential for charge- informedatomisticmodelling,Nat.Mach.Intell.5(2023)1031–1041

    2023

  27. [35]

    Tolborg, A

    K. Tolborg, A. Walsh, Low-cost vibrational free energies in solid solutions with machine learning force fields, J. Phys. Chem. Lett. 14 (2023) 11618–11624

    2023

  28. [36]

    J. P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approxi- mation made simple, Phys. Rev. Lett. 77 (1996) 3865–3868

    1996

  29. [37]

    J. P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approxi- mationmadesimple[phys.rev.lett.77,3865(1996)], Phys.Rev.Lett. 78 (1997) 1396–1396

    1996

  30. [38]

    G.Kresse,J.Hafner, Abinitiomoleculardynamicsforliquidmetals, Phys. Rev. B 47 (1993) 558–561

    1993

  31. [39]

    Kresse, J

    G. Kresse, J. Hafner, Ab initio molecular-dynamics simulation of theliquid-metal–amorphous-semiconductortransitioningermanium, Phys. Rev. B 49 (1994) 14251–14269

    1994

  32. [40]

    Kresse, J

    G. Kresse, J. Furthmüller, Efficiency of ab-initio total energy calcu- lations for metals and semiconductors using a plane-wave basis set, Comp. Mater. Sci. 6 (1996) 15–50

    1996

  33. [41]

    Kresse, J

    G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54 (1996) 11169–11186

    1996

  34. [42]

    G.Kresse,D.Joubert,Fromultrasoftpseudopotentialstotheprojector augmented-wave method, Phys. Rev. B 59 (1999) 1758–1775

    1999

  35. [43]

    Methfessel, A

    M. Methfessel, A. T. Paxton, High-precision sampling for brillouin- zone integration in metals, Phys. Rev. B 40 (1989) 3616–3621

    1989

  36. [44]

    Matter 21 (2009) 055401

    A.Kuzmin,R.A.Evarestov, Quantummechanics-moleculardynam- icsapproachtotheinterpretationofX-rayabsorptionspectra,J.Phys.: Condens. Matter 21 (2009) 055401

    2009

  37. [45]

    Kuzmin, J

    A. Kuzmin, J. Timoshenko, A. Kalinko, I. Jonane, A. Anspoks, Treatment of disorder effects in X-ray absorption spectra beyond the conventional approach, Rad. Phys. Chem. 175 (2020) 108112

    2020

  38. [46]

    E. A. Stern, Number of relevant independent points in X-ray- absorptionfine-structurespectra, Phys.Rev.B48(1993)9825–9827

    1993

  39. [47]

    Jonane, A

    I. Jonane, A. Anspoks, A. Kuzmin, Advanced approach to the local structure reconstruction and theory validation on the example of the W L3-edge extended X-ray absorption fine structure of tungsten, Model. Simul. Mater. Sci. Eng. 26 (2018) 025004

    2018

  40. [48]

    G.Bakradze,E.Welter,A.Kuzmin,Peculiaritiesofthelocalstructure innewmedium-andhigh-entropy,low-symmetrytungstates, J.Phys. Chem. Solids 172 (2023) 111052

    2023

  41. [49]

    Drescher, A

    S. Drescher, A. Kuzmin, E. Welter, J. Freudenberger, A. Smekhova, Component-dependentlatticedistortionsandatomicscaleinsightsin multi-componentAu-Cu-Ni-Pd-Ptbasedalloys, NanoRes.18(2025) 94907122. V. Dimitrijevs et al.:Preprint submitted to ElsevierPage 8 of 9 Local lattice dy...

    2025

  42. [50]

    C.H.Booth,F.Bridges,E.D.Bauer,G.G.Li,J.B.Boyce,T.Claeson, C. W. Chu, Q. Xiong, XAFS measurements of negatively corre- lated atomic displacements HgBa2CuO4+𝛿, Phys. Rev. B 52 (1995) R15745–R15748

    1995

  43. [51]

    Jeong, R

    I.-K. Jeong, R. H. Heffner, M. J. Graf, S. J. L. Billinge, Lattice dy- namicsandcorrelatedatomicmotionfromtheatomicpairdistribution function, Phys. Rev. B 67 (2003) 104301

    2003

  44. [52]

    Batatia, P

    I. Batatia, P. Benner, Y. Chiang, A. M. Elena, D. P. Kovács, J. Riebe- sell, X. R. Advincula, M. Asta, W. J. Baldwin, N. Bernstein, et al., A foundation model for atomistic materials chemistry, J. Chem. Phys. 163 (2025) 184110

    2025

  45. [53]

    Žguns, I

    P. Žguns, I. Pudza, A. Kuzmin, Benchmarking CHGNet Universal Machine Learning Interatomic Potential against DFT and EXAFS: The Case of Layered WS2 and MoS2, J. Chem. Theory Comput. 21 (2025) 8142–8150

    2025

  46. [54]

    B. Deng, Y. Choi, P. Zhong, J. Riebesell, S. Anand, Z. Li, K. Jun, K. A. Persson, G. Ceder, Systematic softening in universal machine learning interatomic potentials, npj Comput. Mater. 11 (2025) 9. V. Dimitrijevs et al.:Preprint submitted to ElsevierPage 9 of 9

    2025

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.