pith. sign in

arxiv: 2605.28395 · v1 · pith:EH5PN47Znew · submitted 2026-05-27 · ❄️ cond-mat.mtrl-sci · cond-mat.dis-nn· physics.comp-ph

Can MACE Potentials Accurately Describe Magnetism and Phase Stability in Fe-Ni Alloys? A Systematic Benchmark

Kushal Ramakrishna , Mani Lokamani , Attila Cangi This is my paper

Pith reviewed 2026-06-29 11:05 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci cond-mat.dis-nnphysics.comp-ph
keywords MACE potentialsFe-Ni alloysphase stabilitymagnetismspecial quasirandom structuresdensity functional theorymachine learning potentialsmagnetic collapse
0
0 comments X

The pith

Targeted SQS training improves MACE accuracy for Fe-Ni structural properties but not for magnetic phase stability.

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

The paper benchmarks MACE machine learning potentials for iron-nickel alloys using reference data from spin-polarized PBE DFT on special quasirandom structures. A custom MACE-sqs model trained on this dataset achieves low validation errors and matches DFT and experiment better than foundation models for equations of state, equilibrium volumes, elastic constants, and thermal expansion trends in bcc and fcc phases. All tested models, however, incorrectly predict that the bcc-to-hcp transition pressure rises with nickel content instead of falling. This failure traces to incomplete capture of high-pressure magnetic collapse and composition-dependent magnetoelastic coupling.

Core claim

The MACE-sqs potential trained on SQS data for Fe-Ni alloys achieves validation errors of 2.0 meV/atom for energies and 24.3 meV/Å for forces and gives the most consistent agreement with DFT for structural, elastic, and thermal properties across compositions and bcc/fcc structures, yet all MACE models predict an incorrect increase of the bcc-to-hcp transition pressure with Ni content because high-pressure magnetic collapse and magnetoelastic effects remain uncaptured.

What carries the argument

The MACE-sqs model, a message-passing atomic cluster expansion potential trained on spin-polarized PBE DFT data for special quasirandom structures spanning Fe-Ni compositions, bcc and fcc lattices, and volumetric/shear deformations.

If this is right

  • MACE-sqs outperforms foundation models for equilibrium volumes, elastic constants, and thermal expansion trends in both bcc and fcc Fe-Ni alloys.
  • For pure iron the MACE-sqs model predicts a bcc-to-hcp transition pressure closer to experiment than the foundation models.
  • All MACE models, including the targeted one, share the same incorrect trend of rising transition pressure with added nickel.
  • Targeted SQS-based training improves accuracy for many properties while phase stability under magnetic collapse stays a shared limitation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same SQS-training strategy could be tested on other binary magnetic alloys such as Fe-Co to check transferability of the accuracy gains.
  • Explicit magnetic moment tracking or spin-dependent descriptors may be required in future MACE variants to resolve the magnetic collapse regime.
  • Discrepancies in predicted phase boundaries suggest direct validation against high-pressure diamond-anvil cell experiments on composition-series Fe-Ni samples.

Load-bearing premise

Spin-polarized PBE DFT calculations provide a sufficiently accurate and transferable reference for both structural and magnetic properties of Fe-Ni alloys across all compositions and under the high-pressure regime of magnetic collapse.

What would settle it

Experimental measurement of how the bcc-to-hcp transition pressure changes with nickel concentration in Fe-Ni alloys, or comparison of the models against higher-accuracy methods under magnetic collapse conditions.

Figures

Figures reproduced from arXiv: 2605.28395 by Attila Cangi, Kushal Ramakrishna, Mani Lokamani.

Figure 1
Figure 1. Figure 1: FIG. 1. Warren–Cowley short–range order (SRO) parameters for Fe–Ni alloys as a function of nickel concentration for bcc [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Overview of the training dataset and MACE model performance. Top left: Representative SQS supercells of Fe–Ni [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Equation of state (EOS) for Fe–Ni compositions, comparing MACE models with DFT and experiments [4, 8, 12, 68–71]. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Volume per atom of Fe–Ni alloys for bcc (left) and [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Transition pressure from the bcc to hcp structure as [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Elastic properties of Fe–Ni alloys for bcc and fcc structures as a function of nickel concentration, comparing MACE [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Lattice thermal expansion of Fe–Ni compositions, comparing MACE models with experiments [73, 76, 92–96]. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
read the original abstract

We present a systematic benchmark of MACE potentials for iron-nickel alloys, focusing on structural, elastic, magnetic, and finite-temperature properties relevant to phase stability. The reference dataset comprises spin-polarized PBE density functional theory (DFT) calculations for chemically disordered special quasirandom structures (SQS), spanning compositions, bcc and fcc crystal structures, and volumetric and shear deformations. A system-specific MACE-sqs model trained on this dataset achieves validation errors of 2.0 meV/atom for energies and 24.3 meV/Angstrom for forces. Compared with several MACE foundation models, including models trained with Hubbard U corrections, MACE-sqs gives the most consistent agreement with DFT and experiment for equations of state, equilibrium volumes, elastic constants, and thermal expansion trends in bcc and fcc Fe-Ni alloys. For the bcc-to-hcp transition, MACE-sqs predicts a pure-Fe transition pressure closer to experiment than the tested foundation models, but all models predict an incorrect increase of transition pressure with Ni content. This failure indicates that high-pressure magnetic collapse and composition-dependent magnetoelastic effects are not yet fully captured. Overall, targeted SQS-based training substantially improves the accuracy of MACE potentials for Fe-Ni alloys, while phase stability under magnetic collapse remains a key limitation for future model development.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper benchmarks several MACE machine-learning potentials, including foundation models and a custom MACE-sqs model trained on spin-polarized PBE DFT data for chemically disordered SQS structures in bcc and fcc Fe-Ni alloys. It reports that MACE-sqs achieves low validation errors (2.0 meV/atom energies, 24.3 meV/Å forces) and outperforms the foundation models on equations of state, equilibrium volumes, elastic constants, and thermal expansion trends. For the bcc-to-hcp transition, MACE-sqs improves the pure-Fe transition pressure relative to other models but, like all tested models, predicts an increase in transition pressure with Ni content, contrary to experiment; the authors attribute this to inadequate capture of high-pressure magnetic collapse and composition-dependent magnetoelastic coupling. The central conclusion is that targeted SQS training substantially improves accuracy for Fe-Ni while phase stability under magnetic collapse remains a key limitation.

Significance. If the results hold, the work supplies a concrete, reproducible benchmark demonstrating that system-specific training on SQS data measurably improves MACE performance on structural, elastic, and finite-temperature properties of magnetic alloys, while cleanly isolating a failure mode in high-pressure phase stability. The explicit comparison to both DFT references and experiment, together with the identification of missing physics rather than data artifacts, strengthens the utility of the findings for future model development in magnetoelastic systems.

major comments (2)
  1. [Abstract; bcc-to-hcp transition results] Abstract and the bcc-to-hcp transition discussion: the claim that the incorrect (increasing) trend of transition pressure with Ni content demonstrates that 'high-pressure magnetic collapse and composition-dependent magnetoelastic effects are not yet fully captured' by the MACE models rests on the assumption that spin-polarized PBE supplies the correct target trend. Because the reference dataset is restricted to bcc/fcc SQS structures and volumetric/shear deformations with no hcp or collapsed-moment configurations included, and no direct PBE-computed transition pressures versus composition are reported for comparison with experiment, the attribution risks conflating possible deficiencies of the DFT reference with limitations specific to the ML potentials.
  2. [Methods (training dataset); Results (phase stability)] Methods and results on model training: while validation errors are reported for MACE-sqs, the manuscript does not quantify how the absence of hcp or high-pressure magnetic data in the training set propagates into the extrapolated transition-pressure predictions, leaving the load-bearing negative conclusion without a direct test of whether the observed discrepancy originates in the MACE architecture or in the underlying PBE reference.
minor comments (2)
  1. [Abstract; Introduction] The distinction between 'MACE foundation models' and the 'MACE-sqs' model should be stated more explicitly in the abstract and introduction to avoid reader confusion about which models were retrained versus used off-the-shelf.
  2. [Figures on bcc-hcp transitions] Figure captions for the transition-pressure plots could usefully include the experimental trend line for direct visual comparison.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and valuable feedback on our manuscript. We address the major comments point by point below, and have revised the manuscript accordingly where appropriate.

read point-by-point responses

  1. Referee: [Abstract; bcc-to-hcp transition results] Abstract and the bcc-to-hcp transition discussion: the claim that the incorrect (increasing) trend of transition pressure with Ni content demonstrates that 'high-pressure magnetic collapse and composition-dependent magnetoelastic effects are not yet fully captured' by the MACE models rests on the assumption that spin-polarized PBE supplies the correct target trend. Because the reference dataset is restricted to bcc/fcc SQS structures and volumetric/shear deformations with no hcp or collapsed-moment configurations included, and no direct PBE-computed transition pressures versus composition are reported for comparison with experiment, the attribution risks conflating possible deficiencies of the DFT reference with limitations specific to the ML potentials.

    Authors: We agree with the referee that our training set does not include hcp structures or high-pressure magnetic collapse configurations, and that we have not computed PBE transition pressures for hcp as a function of Ni content. The bcc-to-hcp predictions are therefore extrapolations from the bcc/fcc data. While the models fail to match the experimental trend of decreasing transition pressure with Ni, we cannot rule out that part of the discrepancy arises from the PBE functional itself under these conditions. In the revised manuscript, we will modify the abstract and discussion to state that the models do not reproduce the experimental composition dependence, and that this highlights the challenge in capturing high-pressure magnetoelastic effects, without definitively attributing it solely to the MACE architecture. We will also note the limitation of the training data. revision: partial

  2. Referee: [Methods (training dataset); Results (phase stability)] Methods and results on model training: while validation errors are reported for MACE-sqs, the manuscript does not quantify how the absence of hcp or high-pressure magnetic data in the training set propagates into the extrapolated transition-pressure predictions, leaving the load-bearing negative conclusion without a direct test of whether the observed discrepancy originates in the MACE architecture or in the underlying PBE reference.

    Authors: We acknowledge that a direct quantification would require additional PBE calculations on hcp structures at varying pressures and compositions to compare against MACE predictions. Such calculations were not performed as the primary focus was on benchmarking within the bcc and fcc phases relevant to the SQS training data. We will add a paragraph in the discussion section explicitly stating this limitation and recommending that future work include hcp and magnetic collapse data to better isolate the source of the discrepancy. revision: partial

Circularity Check

0 steps flagged

No significant circularity; all performance claims benchmarked to external DFT and experiment.

full rationale

The paper trains MACE models on spin-polarized PBE DFT data for SQS structures and reports validation errors plus comparisons for EOS, volumes, elastic constants, thermal expansion, and bcc-hcp transition pressures. All metrics are direct numerical comparisons to independent external references (DFT calculations and experimental values). No equations reduce a claimed prediction to a fitted parameter by construction, no self-citations are load-bearing for the central claims, and the training/validation split does not create self-definitional loops. The negative finding on transition-pressure trends is presented as a discrepancy with experiment, not as a quantity derived from the model's own inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the accuracy of the PBE DFT reference data and on the MACE architecture whose parameters are fitted to that data; no new physical entities are postulated.

free parameters (1)
  • MACE-sqs model parameters
    Hundreds of neural-network weights and radial basis parameters are optimized during training on the DFT SQS dataset to achieve the reported validation errors.
axioms (1)
  • domain assumption Spin-polarized PBE DFT provides an adequate reference for magnetic and structural properties of Fe-Ni alloys
    Invoked as the ground-truth data source for both training and validation across all compositions and deformations.

pith-pipeline@v0.9.1-grok · 5794 in / 1269 out tokens · 44394 ms · 2026-06-29T11:05:51.889435+00:00 · methodology

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

103 extracted references · 4 canonical work pages · 1 internal anchor

  1. [1]

    Kuwayama, K

    Y. Kuwayama, K. Hirose, N. Sata, and Y. Ohishi, Phase relations of iron and iron–nickel alloys up to 300 GPa: Implicationsforcompositionandstructureoftheearth’s inner core, Earth and Planetary Science Letters273, 379 (2008)

    2008

  2. [2]

    W. L. Mao, A. J. Campbell, D. L. Heinz, and G. Shen, PhaserelationsofFe–Nialloysathighpressureandtem- perature, Physics of the Earth and Planetary Interiors 155, 146 (2006)

    2006

  3. [3]

    van Schilfgaarde, I

    M. van Schilfgaarde, I. A. Abrikosov, and B. Johansson, Origin of the invar effect in iron–nickel alloys, Nature 400, 46 (1999)

    1999

  4. [4]

    Dubrovinsky, N

    L. Dubrovinsky, N. Dubrovinskaia, I. A. Abrikosov, M. Vennström, F. Westman, S. Carlson, M. van Schilf- gaarde, and B. Johansson, Pressure-induced invar effect in Fe-Ni alloys, Phys. Rev. Lett.86, 4851 (2001)

    2001

  5. [5]

    Mushnikov, A

    N. Mushnikov, A. Popov, V. Gaviko, A. Protasov, N. Kleinerman, O. Golovnya, and S. Naumov, Pecu- liarities of phase diagram of Fe–Ni system at Ni con- centrations 0 – 20 at.%., Acta Materialia240, 118330 (2022)

    2022

  6. [6]

    A. A. Ahles, J. D. Emery, and D. C. Dunand, Me- chanical properties of meteoritic Fe–Ni alloys for in-situ extraterrestrial structures, Acta Astronautica189, 465 (2021)

    2021

  7. [7]

    Huang, W

    E. Huang, W. A. Bassett, and M. S. Weathers, Phase relationships in Fe–Ni alloys at high pressures and tem- peratures,JournalofGeophysicalResearch: SolidEarth 93, 7741 (1988)

    1988

  8. [8]

    Akahama, Y

    Y. Akahama, Y. Fujimoto, T. Terai, T. Fukuda, S. Kawaguchi, N. Hirao, Y. Ohishi, and T. Kakeshita, Pressure–composition phase diagram of Fe–Ni alloy, MATERIALS TRANSACTIONS61, 1058 (2020)

    2020

  9. [9]

    Ramakrishna, M

    K. Ramakrishna, M. Lokamani, A. Baczewski, J. Vor- berger, and A. Cangi, Electrical conductivity of iron in earth’s core from microscopic ohm’s law, Phys. Rev. B 107, 115131 (2023)

    2023

  10. [10]

    Takahashi and W

    T. Takahashi and W. A. Bassett, High-pressure poly- morph of iron, Science145, 483 (1964)

    1964

  11. [11]

    Mathon, F

    O. Mathon, F. Baudelet, J. P. Itié, A. Polian, M. d’Astuto, J. C. Chervin, and S. Pascarelli, Dynamics of the magnetic and structuralα−ϵphase transition in iron, Phys. Rev. Lett.93, 255503 (2004)

    2004

  12. [12]

    Ramakrishna, M

    K. Ramakrishna, M. Lokamani, A. Baczewski, J. Vor- berger, and A. Cangi, Impact of electronic corre- lations on high-pressure iron: insights from time- dependent density functional theory, Electronic Struc- ture5, 045002 (2023)

    2023

  13. [13]

    Behler and M

    J. Behler and M. Parrinello, Generalized neural-network representationofhigh-dimensionalpotential-energysur- faces, Phys. Rev. Lett.98, 146401 (2007)

    2007

  14. [14]

    Behler, Four generations of high-dimensional neu- ral network potentials, Chemical Reviews121, 10037 (2021)

    J. Behler, Four generations of high-dimensional neu- ral network potentials, Chemical Reviews121, 10037 (2021)

    2021

  15. [15]

    Thompson, L

    A. Thompson, L. Swiler, C. Trott, S. Foiles, and G. Tucker, Spectral neighbor analysis method for auto- mated generation of quantum-accurate interatomic po- tentials, Journal of Computational Physics285, 316 (2015)

    2015

  16. [16]

    Y. Zuo, C. Chen, X. Li, Z. Deng, Y. Chen, J. Behler, G. Csányi, A. V. Shapeev, A. P. Thompson, M. A. Wood, and S. P. Ong, Performance and cost assessment of machine learning interatomic potentials, The Journal of Physical Chemistry A124, 731 (2020)

    2020

  17. [17]

    Nikolov, M

    S. Nikolov, M. A. Wood, A. Cangi, J.-B. Maillet, M.- C. Marinica, A. P. Thompson, M. P. Desjarlais, and J. Tranchida, Data-driven magneto-elastic predictions with scalable classical spin-lattice dynamics, npj Com- putational Materials7, 153 (2021)

    2021

  18. [18]

    Nikolov, J

    S. Nikolov, J. Tranchida, K. Ramakrishna, M. Loka- mani, A. Cangi, and M. A. Wood, Dissociating the phononic, magneticandelectroniccontributionstother- mal conductivity: a computational study in alpha-iron, Journal of Materials Science57, 10535 (2022)

    2022

  19. [19]

    Nikolov, K

    S. Nikolov, K. Ramakrishna, A. Rohskopf, M. Loka- mani, J. Tranchida, J. Carpenter, A. Cangi, and M. A. Wood, Probing iron in earth’s core with molecular-spin dynamics, Proceedings of the National Academy of Sci- ences121, e2408897121 (2024)

    2024

  20. [20]

    A. P. Bartók, M. C. Payne, R. Kondor, and G. Csányi, Gaussian approximation potentials: The accuracy of quantum mechanics, without the electrons, Phys. Rev. Lett.104, 136403 (2010)

    2010

  21. [21]

    Dragoni, T

    D. Dragoni, T. D. Daff, G. Csányi, and N. Marzari, Achieving DFT accuracy with a machine-learning inter- atomic potential: Thermomechanics and defects in bcc ferromagnetic iron, Phys. Rev. Mater.2, 013808 (2018)

    2018

  22. [22]

    Shenoy, C

    L. Shenoy, C. D. Woodgate, J. B. Staunton, A. P. Bartók, C. S. Becquart, C. Domain, and J. R. Kermode, Collinear-spin machine learned interatomic potential for Fe7Cr2Ni alloy, Phys. Rev. Mater.8, 033804 (2024)

    2024

  23. [23]

    Fellman, J

    A. Fellman, J. Byggmästar, F. Granberg, K. Nordlund, and F. Djurabekova, Fast and accurate machine-learned interatomic potentials for large-scale simulations of Cu, Al, and Ni, Phys. Rev. Mater.9, 053807 (2025)

    2025

  24. [24]

    A. V. Shapeev, Moment tensor potentials: A class of systematically improvable interatomic potentials, Mul- tiscale Modeling & Simulation14, 1153 (2016)

    2016

  25. [25]

    I.Novikov, B.Grabowski, F.Körmann,andA.Shapeev, Magnetic moment tensor potentials for collinear spin- polarized materials reproduce different magnetic states of bcc Fe, npj Computational Materials8, 13 (2022)

    2022

  26. [26]

    S. Hao, P. Singh, A. V. Smirnov, D. D. Johnson, D. E. Alman, and M. C. Gao, Developing reliable machine learning interatomic potential for Fe–Cr–Ni austenitic alloys, Journal of Applied Physics138, 085103 (2025)

    2025

  27. [27]

    Drautz, Atomic cluster expansion for accurate and transferable interatomic potentials, Phys

    R. Drautz, Atomic cluster expansion for accurate and transferable interatomic potentials, Phys. Rev. B99, 014104 (2019)

    2019

  28. [28]

    Rinaldi, M

    M. Rinaldi, M. Mrovec, A. Bochkarev, Y. Lysogorskiy, and R. Drautz, Non-collinear magnetic atomic cluster expansion for iron, npj Computational Materials10, 12 (2024)

    2024

  29. [29]

    C. J. Owen, S. B. Torrisi, Y. Xie, S. Batzner, K. Bystrom, J. Coulter, A. Musaelian, L. Sun, and B. Kozinsky, Complexity of many-body interactions in transition metals via machine-learned force fields from the TM23 data set, npj Computational Materials10, 92 (2024)

    2024

  30. [30]

    Zhang, J

    L. Zhang, J. Han, H. Wang, R. Car, and W. E, Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics, Phys. Rev. Lett. 14 120, 143001 (2018)

    2018

  31. [31]

    Zhang, H

    Y. Zhang, H. Wang, W. Chen, J. Zeng, L. Zhang, H. Wang, and W. E, DP-GEN: A concurrent learn- ing platform for the generation of reliable deep learn- ing based potential energy models, Computer Physics Communications253, 107206 (2020)

    2020

  32. [32]

    X. Gong, Z. Li, A. S. L. S. Pattamatta, T. Wen, and D. J. Srolovitz, An accurate and transferable machine learning interatomic potential for nickel, Communica- tions Materials5, 157 (2024)

    2024

  33. [33]

    E. O. Khazieva, N. M. Chtchelkatchev, N. N. Katkov, and R. E. Ryltsev, Accuracy and limitations of machine- learned interatomic potentials for magnetic systems: A case study on Fe-Cr-C, Phys. Rev. E112, 055302 (2025)

    2025

  34. [34]

    Batzner, A

    S. Batzner, A. Musaelian, L. Sun, M. Geiger, J. P. Mailoa, M. Kornbluth, N. Molinari, T. E. Smidt, and B. Kozinsky, E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials, Na- ture Communications13, 2453 (2022)

    2022

  35. [35]

    Batatia, D

    I. Batatia, D. P. Kovács, G. Simm, C. Ortner, and G. Csányi, MACE: Higher order equivariant message passing neural networks for fast and accurate force fields, inAdvances in Neural Information Processing Systems, Vol. 35, edited by S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (Curran Associates, Inc., 2022) pp. 11423–11436

    2022

  36. [36]

    B. Deng, P. Zhong, K. Jun, J. Riebesell, K. Han, C. J. Bartel, and G. Ceder, Chgnet as a pretrained universal neural network potential for charge-informed atomistic modelling, Nature Machine Intelligence5, 1031 (2023)

    2023

  37. [37]

    Schmidt, H.-C

    J. Schmidt, H.-C. Wang, T. F. T. Cerqueira, S. Botti, and M. A. L. Marques, A dataset of 175K stable and metastable materials calculated with the PBEsol and SCAN functionals, Scientific Data9, 64 (2022)

    2022

  38. [38]

    Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models

    L. Barroso-Luque, M. Shuaibi, X. Fu, B. M. Wood, M. Dzamba, M. Gao, A. Rizvi, C. L. Zitnick, and Z. W. Ulissi, Open materials 2024 (omat24) inorganic materi- als dataset and models (2024), arXiv:2410.12771 [cond- mat.mtrl-sci]

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  39. [39]

    A. D. Kaplan, R. Liu, J. Qi, T. W. Ko, B. Deng, J. Riebesell, G. Ceder, K. A. Persson, and S. P. Ong, A foundational potential energy surface dataset for mate- rials (2025), arXiv:2503.04070 [cond-mat.mtrl-sci]

    work page arXiv 2025

  40. [40]

    van de Walle, M

    A. van de Walle, M. D. Asta, and G. Ceder, The Alloy Theoretic Automated Toolkit: A user guide, Calphad 26, 539 (2002)

    2002

  41. [41]

    van de Walle, P

    A. van de Walle, P. Tiwary, M. de Jong, D. Olmsted, M.Asta, A.Dick, D.Shin, Y.Wang, L.-Q.Chen,andZ.- K. Liu, Efficient stochastic generation of special quasir- andom structures, Calphad42, 13 (2013)

    2013

  42. [42]

    J. M. Cowley, Short-range order and long-range order parameters, Phys. Rev.138, A1384 (1965)

    1965

  43. [43]

    J. M. Cowley, An approximate theory of order in alloys, Phys. Rev.77, 669 (1950)

    1950

  44. [44]

    Kresse and J

    G. Kresse and J. Hafner, Ab initio molecular dynamics for liquid metals, Phys. Rev. B47, 558 (1993)

    1993

  45. [45]

    Kresse and D

    G. Kresse and D. Joubert, From ultrasoft pseudopoten- tials to the projector augmented-wave method, Phys. Rev. B59, 1758 (1999)

    1999

  46. [46]

    Kresse and J

    G. Kresse and J. Furthmüller, Efficiency of ab-initio to- tal energy calculations for metals and semiconductors using a plane-wave basis set, Computational Materials Science6, 15 (1996)

    1996

  47. [47]

    G.KresseandJ.Furthmüller,Efficientiterativeschemes forabinitiototal-energycalculationsusingaplane-wave basis set, Phys. Rev. B54, 11169 (1996)

    1996

  48. [48]

    J. P. Perdew, K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77, 3865 (1996)

    1996

  49. [49]

    Tesch and P

    R. Tesch and P. M. Kowalski, Hubbarduparameters for transition metals from first principles, Phys. Rev. B 105, 195153 (2022)

    2022

  50. [50]

    Hausoel, M

    A. Hausoel, M. Karolak, E. Şaşιoğlu, A. Lichtenstein, K. Held, A. Katanin, A. Toschi, and G. Sangiovanni, Local magnetic moments in iron and nickel at ambient and earth’s core conditions, Nature Communications8, 16062 (2017)

    2017

  51. [51]

    T. D. Warford, F. L. Thiemann, and G. Csányi, Better without U: Impact of selective Hubbard U correction on foundational MLIPs, Machine Learning: Science and Technology 10.1088/2632-2153/ae6be5 (2026)

    work page doi:10.1088/2632-2153/ae6be5 2026

  52. [52]

    Xiong, H

    W. Xiong, H. Zhang, L. Vitos, and M. Selleby, Magnetic phase diagram of the Fe–Ni system, Acta Materialia59, 521 (2011)

    2011

  53. [53]

    Peschard, Contribution à l’étude des ferro-nickels, Re- vue de Métallurgie22, 490 (1925)

    1925

  54. [54]

    Crangle, G

    J. Crangle, G. C. Hallam, and W. Sucksmith, The mag- netization of face-centred cubic and body-centred cubic iron + nickel alloys, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 272, 119 (1963)

    1963

  55. [55]

    C. S. Tian, D. Qian, D. Wu, R. H. He, Y. Z. Wu, W. X. Tang, L. F. Yin, Y. S. Shi, G. S. Dong, X. F. Jin, X. M. Jiang, F. Q. Liu, H. J. Qian, K. Sun, L. M. Wang, G. Rossi, Z. Q. Qiu, and J. Shi, Body-centered-cubic Ni and its magnetic properties, Phys. Rev. Lett.94, 137210 (2005)

    2005

  56. [56]

    Sumiyama, M

    K. Sumiyama, M. Kadono, and Y. Nakamura, Metastable bcc phase in sputtered Fe–Ni alloys, Trans- actions of the Japan institute of metals24, 190 (1983)

    1983

  57. [57]

    Asano, Magnetism ofγFe-Ni invar alloys with low nickel concentration, Journal of the Physical Society of Japan27, 542 (1969)

    H. Asano, Magnetism ofγFe-Ni invar alloys with low nickel concentration, Journal of the Physical Society of Japan27, 542 (1969)

    1969

  58. [58]

    Bando, The magnetization of face centered cubic iron-nickel alloys in the vicinity of invar region, Jour- nal of the Physical Society of Japan19, 237 (1964)

    Y. Bando, The magnetization of face centered cubic iron-nickel alloys in the vicinity of invar region, Jour- nal of the Physical Society of Japan19, 237 (1964)

    1964

  59. [59]

    Glaubitz, S

    B. Glaubitz, S. Buschhorn, F. Brüssing, R. Abrudan, and H. Zabel, Development of magnetic moments in Fe1−xNix-alloys, Journal of Physics: Condensed Mat- ter23, 254210 (2011)

    2011

  60. [60]

    I. A. Tsoukalas and K. G. Melidis, Measurements on the magnetic properties of Fe–Ni alloys, International Journal of Materials Research78, 498 (1987)

    1987

  61. [61]

    Dunaev, and E

    V.M.Kalinin, V.A.Kornyakov, O.A.Khomenko, F.N. Dunaev, and E. E. Yurchikov, Mössbauer investigation of iron-nickel invar alloys, Bulletin of the Academy of Sciences of the USSR, Physical Series36, 1420 (1972)

    1972

  62. [62]

    Biewald, Experiment tracking with weights and bi- ases (2020), software available from wandb.com

    L. Biewald, Experiment tracking with weights and bi- ases (2020), software available from wandb.com

    2020

  63. [63]

    Riebesell, R

    J. Riebesell, R. E. A. Goodall, P. Benner, Y. Chiang, B. Deng, G. Ceder, M. Asta, A. A. Lee, A. Jain, and K. A. Persson, A framework to evaluate machine learn- ing crystal stability predictions, Nature Machine Intel- ligence7, 836 (2025)

    2025

  64. [64]

    Hjorth Larsen, J

    A. Hjorth Larsen, J. Jørgen Mortensen, J. Blomqvist, I. E. Castelli, R. Christensen, M. Dułak, J. Friis, M. N. Groves, B. Hammer, C. Hargus, E. D. Hermes, 15 P. C. Jennings, P. Bjerre Jensen, J. Kermode, J. R. Kitchin, E. Leonhard Kolsbjerg, J. Kubal, K. Kaasb- jerg, S. Lysgaard, J. Bergmann Maronsson, T. Max- son, T. Olsen, L. Pastewka, A. Peterson, C. R...

    2017

  65. [65]

    Cohen, J

    O. Cohen, J. Riebesell, R. Goodall, A. Kolluru, S. Fall- etta, J. Krause, J. Colindres, G. Ceder, and A. S. Gan- gan, Torchsim: an efficient atomistic simulation engine in pytorch, AI for Science1, 025003 (2025)

    2025

  66. [66]

    Birch, Finite elastic strain of cubic crystals, Phys

    F. Birch, Finite elastic strain of cubic crystals, Phys. Rev.71, 809 (1947)

    1947

  67. [67]

    F. D. Murnaghan, The compressibility of media un- der extreme pressures, Proceedings of the National Academy of Sciences30, 244 (1944)

    1944

  68. [68]

    Dewaele, M

    A. Dewaele, M. Torrent, P. Loubeyre, and M. Mezouar, Compression curves of transition metals in the mbar range: Experiments and projector augmented-wave cal- culations, Phys. Rev. B78, 104102 (2008)

    2008

  69. [69]

    R. A. Morrison, J. M. Jackson, W. Sturhahn, D. Zhang, and E. Greenberg, Equations of state and anisotropy of Fe-Ni-Si alloys, Journal of Geophysical Research: Solid Earth123, 4647 (2018)

    2018

  70. [70]

    C. Zhou, J. Buchen, V. V. Dobrosavljevic, B. Strozewski, O. S. Pardo, W. Sturhahn, T. Ishii, T. S. Toellner, J. D. Wilding, S. Chariton, B. Kalkan, M. Kunz, and J. M. Jackson, Equation of state of a low-sulfur iron–nickel alloy up to 143 GPa, Physics of the Earth and Planetary Interiors372, 107515 (2026)

    2026

  71. [71]

    Hirao, Y

    N. Hirao, Y. Akahama, and Y. Ohishi, Equations of state of iron and nickel to the pressure at the center of the earth, Matter and Radiation at Extremes7, 038403 (2022)

    2022

  72. [72]

    Zwell, D

    L. Zwell, D. E. Carnahan, and G. R. Speich, Lattice parameter of ferritic and martensitic Fe-Ni alloys, Met- allurgical Transactions1, 1007 (1970)

    1970

  73. [73]

    Hayase, M

    M. Hayase, M. Shiga, and Y. Nakamura, Spontaneous volume magnetostriction and lattice constant of face- centered cubic Fe-Ni and Ni-Cu alloys, Journal of the Physical Society of Japan34, 925 (1973)

    1973

  74. [74]

    M. Acet, H. Zähres, E. F. Wassermann, and W. Pepper- hoff, High-temperature moment-volume instability and anti-invar ofγ-Fe, Phys. Rev. B49, 6012 (1994)

    1994

  75. [75]

    Kądziołka-Gaweł, W

    M. Kądziołka-Gaweł, W. Zarek, E. Popiel, and A. Chrobak, The crystal structure and magnetic prop- erties of selected fcc FeNi and Fe40Ni40B20 alloys, Acta Physica Polonica A117(2010)

    2010

  76. [76]

    N. S. Kanhe, A. Kumar, S. Yusuf, A. Nawale, S. Gaik- wad, S. A. Raut, S. Bhoraskar, S. Y. Wu, A. Das, and V. Mathe, Investigation of structural and magnetic properties of thermal plasma-synthesized Fe1−xNix al- loy nanoparticles, Journal of Alloys and Compounds 663, 30 (2016)

    2016

  77. [77]

    Q. Wei, S. A. Gilder, W. Ertel-Ingrisch, F. Guillou, and F. Wilhelm, Magnetism of body-centered cubic Fe–Ni alloys under pressure: Strain-enhanced ferromagnetism at the phase transitions, Journal of Geophysical Re- search: Solid Earth125, e2020JB020922 (2020)

    2020

  78. [78]

    Ekman, B

    M. Ekman, B. Sadigh, K. Einarsdotter, and P. Blaha, Ab initio study of the martensitic bcc-hcp transforma- tion in iron, Phys. Rev. B58, 5296 (1998)

    1998

  79. [79]

    J.J.Adams, D.S.Agosta, R.G.Leisure,andH.Ledbet- ter, Elastic constants of monocrystal iron from 3to500k, Journal of Applied Physics100, 113530 (2006)

    2006

  80. [80]

    J. A. Rayne and B. S. Chandrasekhar, Elastic constants of iron from 4.2 to 300°K, Phys. Rev.122, 1714 (1961)

    1961

Showing first 80 references.