Pith. sign in

REVIEW 2 major objections 5 minor 14 references

Scalar RDF convergence can miss slow chemical ordering in CoNiV; shell-resolved bond statistics and their PCA modes track the real SRO.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-14 09:20 UTC pith:LK2E3WHO

load-bearing objection Solid methods note: lag-JSD and PCA cleanly show RDF can miss slow SRO, with a coherent V-sublattice mode in CoNiV, though second-shell sampling is incomplete and the MTP is imported unvalidated. the 2 major comments →

arxiv 2607.10775 v1 pith:LK2E3WHO submitted 2026-07-12 physics.comp-ph

Understanding Chemical Short-Range Order in CoNiV via Mode Analysis

Jilang Miao , Miaomiao Jin This is my paper
classification physics.comp-ph
keywords chemical short-range orderWarren-Cowley parametersCoNiVprincipal component analysishybrid MD/MCJensen-Shannon divergenceL12-like orderingmachine-learned potential
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

In equiatomic face-centered-cubic CoNiV, atoms are not randomly mixed: vanadium strongly avoids other vanadium atoms as nearest neighbors and piles up at second-neighbor positions, forming a short-range L12-like motif. The paper shows that ordinary radial distribution functions and total energy can look fully relaxed long before this chemical pattern has finished developing. By grouping neighbors into fixed coordination shells, computing the full set of pair probabilities, and running principal component analysis on those vectors, the authors extract three collective modes—the dominant one being coherent V-sublattice ordering on shells 2 and 4. Lagged Jensen–Shannon distances confirm that the bond statistics keep evolving after the RDF has plateaued. The practical payoff is a compact, shell-resolved diagnostic that tells simulators when chemical short-range order has truly converged and which pair correlations move together.

Core claim

In hybrid MD/MC snapshots of equiatomic CoNiV at 300 K, first-shell V–V avoidance and second-shell V–V enrichment develop after the radial structure has already stabilized; principal-component analysis of the shell-resolved pair-probability vectors isolates this process as a coherent V-sublattice ordering mode on shells 2 and 4 (with shell 3 remaining weak), together with separate Ni–Co redistribution and Co–V exchange modes, proving that scalar RDF convergence alone is an insufficient equilibration metric for chemical short-range order.

What carries the argument

Shell-resolved pair-probability vectors (24-dimensional for three elements and four FCC shells) fed to per-replica-centered PCA, which extracts collective modes whose loadings identify which shells and element pairs vary together; the first mode is the V-sublattice ordering amplitude.

Load-bearing premise

The machine-learned interatomic potential used for every energy evaluation correctly ranks the relative pair preferences that drive the observed ordering.

What would settle it

An independent long hybrid MD/MC trajectory of the same alloy driven by a different, validated potential (or by direct DFT-based Monte Carlo) that shows either no second-shell V–V enrichment or first-shell ordering that continues to evolve after the RDF has plateaued would overturn the claimed mode structure and the RDF-versus-SRO timescale separation.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The manuscript analyzes chemical short-range order (SRO) in equiatomic fcc CoNiV using hybrid MD/MC snapshots generated with a level-20 moment tensor potential. Shell boundaries are fixed from the total RDF; shell-resolved Warren–Cowley parameters and pair fractions are then tracked. The dominant pairwise signal is first-shell V–V avoidance and second-shell V–V enrichment, interpreted as a short-range L12-like motif, while shell 3 remains weak. Lagged Jensen–Shannon distances show that RDF statistics plateau while the pair-fraction vectors continue to evolve. PCA of per-replica-centered pair fractions in a post-saturation window (steps ≥ 21 000) yields three modes: a V-sublattice ordering amplitude (PC1, shells 2 and 4), a Ni–Co redistribution mode (PC2), and a mixed Co–V mode (PC3). The authors conclude that scalar RDF convergence can miss slow chemical relaxation and that shell-resolved bond statistics plus PCA provide a compact diagnostic for SRO development.

Significance. If the sampling and potential are adequate, the work supplies a practical, unsupervised protocol (lag-JSD on pair fractions plus PCA of shell-resolved bond statistics) that cleanly separates structural from chemical equilibration in multicomponent alloys. The explicit demonstration that first-shell WC parameters flatten while second-shell V–V order continues, and that this slow direction is recovered as the leading PC, is useful for the community that relies on hybrid MD/MC for SRO. Six independent replicas, between-replica error bars, and the frank statement that second-shell magnitudes are lower bounds are methodological strengths. The main limitation is that the L12-like mode claim rests on an incompletely equilibrated second shell and on an MTP whose ranking of pair energies is not validated inside this study.

major comments (2)
  1. [Results, Slow Chemical Ordering; Fig. 4; Collective Modes] Results, Slow Chemical Ordering and Fig. 4: the paper itself states that α_VV^(2) continues to decrease through the full 170 000-step trajectory and that reported second-shell magnitudes are only lower bounds. PCA is performed on the post-saturation window chosen because first-shell α had flattened; PC1 (29 % variance) is then interpreted as a coherent V-sublattice / L12-like mode because its loadings couple shells 2 and 4 with the expected V-pair signs. If the order parameter that defines the mode is still drifting, the loadings, the variance attribution, and the claim of a “robust short-range L12-like motif” are conditioned on an incompletely equilibrated ensemble. Either longer sampling until second-shell stationarity, or a quantitative demonstration that the PC1 direction is stable under further evolution, is needed before the mode can be treated as established.
  2. [Methods, Hybrid MD/MC Simulations] Methods, Hybrid MD/MC Simulations: all energetics and therefore all SRO are generated by a level-20 MTP imported from a concurrent arXiv preprint (ref. 9) without any independent validation against DFT or experiment inside this manuscript. Because the central physical claim is a specific L12-like pair preference (first-shell V–V avoidance, second-shell enrichment), the ranking of relative pair energies by the potential is load-bearing. At minimum, a short comparison of key WC parameters or formation energies against DFT (or against published experimental CSRO motifs for CoNiV) should be supplied so that readers can judge whether the observed modes are physical or potential-specific.
minor comments (5)
  1. [Results, Slow Chemical Ordering; Fig. 3] Fig. 3 caption and body text refer to “Fig. ??” / “as shown in Fig. ??”; the cross-reference is broken and should be fixed.
  2. [Introduction; Methods] Equation numbering: the WC definition is labeled (1) in the Introduction and then the JSD is also labeled (2); the text later says “using Eq. (2)” for WC. Renumber consistently.
  3. [Title; Abstract] Title and abstract use “CoNiV” / “NiCoV” interchangeably; pick one ordering and keep it throughout.
  4. [Results, Collective Modes] The shell-weight definition used to localize PCs (“sum of the squared PC-vector entries over the six pair features”) is clear but should be written as an explicit formula so that the 0.78 / 0.94 numbers are reproducible.
  5. [Methods] A brief statement of how many MC swaps are accepted per 100 attempts (or the acceptance rate) would help readers judge sampling efficiency.

Circularity Check

0 steps flagged

No significant circularity; WC, lag-JSD and PCA are applied unsupervised to simulation pair fractions, with L12-like interpretation from observed signs rather than by construction.

full rationale

The paper's chain is: hybrid MD/MC with an external MTP generates snapshots; RDF minima define shells; pair fractions yield WC parameters and the 24-D feature vectors; lag-JSD compares block distributions of RDF vs Ppair; PCA is run on per-replica-centered Ppair in a post-21k-step window chosen after first-shell WC flattened. None of these steps is definitional of the claimed modes or of the RDF-vs-SRO timescale separation. PC loadings and variance fractions are data-driven eigenvectors, not fitted targets renamed as predictions. The L12-like reading follows from the observed sign pattern (first-shell V-V avoidance, second-shell enrichment, shells 2/4 coupling) matching the known ideal L12 reference, which is an external comparison rather than a self-referential definition. The concurrent arXiv potential citation supplies an input Hamiltonian, not a uniqueness theorem or load-bearing prior result that forces the present conclusions. Window selection after inspecting first-shell traces is ordinary analysis practice and does not make the subsequent PCA or lag-JSD results true by construction. The derivation is therefore self-contained against its own simulation outputs; incomplete second-shell equilibration is a convergence/validity concern, not circularity.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 1 invented entities

The central claim rests on standard statistical definitions plus the assumption that the imported machine-learned potential correctly ranks chemical pair energies; free parameters are analysis-window choices and simulation hyperparameters rather than fitted physical constants.

free parameters (3)
  • post-saturation analysis window start = 21000 MC/MD steps
    Chosen by hand as steps >= 21 000 after visual inspection that first-shell WC traces had flattened; all PCA and lag-JSD results depend on this cut.
  • shell boundary radii rs,min = minima of replica-averaged g(r)
    Taken from minima of the averaged total g(r); small shifts would reassign pairs among shells and alter WC and PC loadings.
  • MC swap attempt schedule = 100 attempts / 100 MD steps
    100 attempts per pair type every 100 MD steps; controls the effective chemical sampling rate and therefore the observed relaxation times.
axioms (4)
  • domain assumption The level-20 moment-tensor potential accurately describes the relative formation energies of the local chemical environments that drive SRO in CoNiV.
    All forces and energies, hence all accepted swaps, come from this potential (Methods).
  • domain assumption Warren-Cowley parameters and pair fractions computed on fixed FCC coordination shells fully capture the relevant SRO state for the subsequent PCA.
    The entire mode analysis is performed on the 24-dimensional Ppair vectors defined by those shells.
  • domain assumption Atom-swap Monte Carlo in the NVT ensemble with the stated attempt rates samples the equilibrium chemical distribution on the timescales of interest.
    Standard hybrid MD/MC premise; incomplete second-shell saturation shows the assumption is only partially realized.
  • standard math Jensen-Shannon distance between block-averaged discrete distributions is a valid stationarity metric for both RDF histograms and bond-probability vectors.
    Used without modification from information theory (Eq. 2).
invented entities (1)
  • V-sublattice ordering amplitude (PC1) no independent evidence
    purpose: Interprets the leading principal component as a coherent L12-like order parameter spanning shells 2 and 4.
    The entity is an interpretive label placed on a data-driven PC; no independent experimental observable is predicted beyond the already-known second-shell V enrichment.

pith-pipeline@v1.1.0-grok45 · 11243 in / 2871 out tokens · 50369 ms · 2026-07-14T09:20:22.340996+00:00 · methodology

0 comments
read the original abstract

We analyze chemical short-range order in equiatomic fcc NiCoV using molecular-dynamics snapshots generated with a machine-learned interatomic potential. Radial distribution functions identify stable coordination shells, while shell-resolved Warren-Cowley parameters and bond probabilities reveal continued chemical ordering after the radial structure has largely converged. The dominant signal is V-V avoidance in the first shell and V-V enrichment in the second shell, consistent with an L1$_2$-like local ordering tendency, while the third-shell response remains weak. Lagged Jensen-Shannon diagnostics show that bond statistics relax more slowly than the RDF. Principal component analysis of per-replica-centered bond probabilities resolves three collective modes: a V-sublattice ordering amplitude, a Ni-Co redistribution mode, and a Co-V exchange-like mode. These results show that scalar RDF convergence can miss slow chemical relaxation, and that shell-resolved bond statistics provide a compact route for tracking SRO development in multicomponent alloys.

Figures

Figures reproduced from arXiv: 2607.10775 by Jilang Miao, Miaomiao Jin.

Figure 1
Figure 1. Figure 1: Replica- and time-averaged total g(r) of equiatomic CoNiV at 300 K. Dashed lines mark the fixed shell boundaries rs,min used for all pair statistics. METHODS Hybrid MD/MC Simulations Hybrid molecular dynamics/Monte Carlo (MD/MC) sim￾ulations were performed in LAMMPS [5] for an equiatomic CoNiV face-centered cubic (FCC) solid solution containing 2048 atoms (8 × 8 × 8 conventional cells; L = 28.72 Å; a0 ≈ 3.… view at source ↗
Figure 2
Figure 2. Figure 2: reports the WC parameters of the first two shells, averaged over the post-saturation window of each replica and over the six replicas; the uncertainty is the between￾replica standard deviation. The first shell shows strong V–V avoidance, together with preferential Ni–V and Co–V nearest￾neighbor pairing. In the second shell, the V-containing entries change sign, indicate the opposite preference. Thus, V ato… view at source ↗
Figure 3
Figure 3. Figure 3: Mean JS distance between blocks separated by in [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Six-replica mean of selected WC parameters over the [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: PC scores obtained by projecting the full trajectories [PITH_FULL_IMAGE:figures/full_fig_p003_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

14 extracted references · 1 linked inside Pith

  1. [1]

    Short-range order and its impact on the CrCoNi medium-entropy alloy,

    R. ZHANG et al., “Short-range order and its impact on the CrCoNi medium-entropy alloy,”Nature,581, 283–287 (2020)

    2020

  2. [2]

    Direct observation of chemical short-range order in a medium-entropy alloy,

    X. CHEN et al., “Direct observation of chemical short-range order in a medium-entropy alloy,”Nature,592, 712–716 (2021)

    2021

  3. [3]

    Dual-scale chemical ordering for cryogenic properties in CoNiV-based alloys,

    T. LU et al., “Dual-scale chemical ordering for cryogenic properties in CoNiV-based alloys,”Nature,645, 385–391 (2025)

    2025

  4. [4]

    An Approximate Theory of Order in Alloys,

    J. M. COWLEY, “An Approximate Theory of Order in Alloys,”Physical Review,77, 669–675 (1950)

    1950

  5. [5]

    LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales,

    A. P. THOMPSON et al., “LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales,”Comp. Phys. Comm.,271, 108171 (2022)

    2022

  6. [6]

    Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials,

    A. V . SHAPEEV, “Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials,”Multiscale Modeling&Simulation, 14, 1153–1173 (2016)

    2016

  7. [7]

    The MLIP package: moment tensor potentials with MPI and active learning,

    I. S. NOVIKOV et al., “The MLIP package: moment tensor potentials with MPI and active learning,”Machine Learning: Science and Technol- ogy,2, 025002 (2021)

    2021

  8. [8]

    MLIP-3: Active learning on atomic environ- ments with moment tensor potentials,

    E. PODRY ABINKIN et al., “MLIP-3: Active learning on atomic environ- ments with moment tensor potentials,”The Journal of Chemical Physics, 159, 084112 (2023)

    2023

  9. [9]

    Chemical Short-Range Order Regulates Hydrogen Ener- getics and Hydrogen-Dislocation Interactions in CoNiV ,

    B. CHEN et al., “Chemical Short-Range Order Regulates Hydrogen Ener- getics and Hydrogen-Dislocation Interactions in CoNiV ,”arXiv preprint arXiv:2604.05352(2026)

    Pith/arXiv arXiv 2026

  10. [10]

    Scalable parallel Monte Carlo algorithm for atomistic simulations of precipitation in alloys,

    B. SADIGH et al., “Scalable parallel Monte Carlo algorithm for atomistic simulations of precipitation in alloys,”Physical Review B,85, 184203 (2012)

    2012

  11. [11]

    Divergence measures based on the Shannon entropy,

    J. LIN, “Divergence measures based on the Shannon entropy,”IEEE Transactions on Information Theory,37, 145–151 (1991)

    1991

  12. [12]

    A new metric for probability distributions,

    D. M. ENDRES and J. E. SCHINDELIN, “A new metric for probability distributions,”IEEE Transactions on Information Theory,49, 1858–1860 (2003)

    2003

  13. [13]

    Structure motif of chemical short-range order in a medium-entropy alloy,

    X. CHEN et al., “Structure motif of chemical short-range order in a medium-entropy alloy,”Materials Research Letters,10, 149–155 (2022)

    2022

  14. [14]

    Short-range order in face-centered cubic VCoNi alloys,

    T. KOSTIUCHENKO et al., “Short-range order in face-centered cubic VCoNi alloys,”Physical Review Materials,4, 113802 (2020)

    2020