pith. sign in

arxiv: 2606.21796 · v1 · pith:74MXFLMXnew · submitted 2026-06-19 · ❄️ cond-mat.mtrl-sci

The FAST Framework: Developing a Data-Efficient Machine Learning Potential to Decode Superionic Transition-Induced Thermophysical and Kinetic Anomalies in UO2 under Extreme Conditions

Fengnian Zhuang , Gaosheng Yan , Hong Chen , Yi Zhang , Wenshan Yu , Minglong Xu , Shengping Shen This is my paper

Pith reviewed 2026-06-26 13:13 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci
keywords machine learning interatomic potentialUO2superionic transitionneuroevolution potentialthermal expansionionic diffusionactive learning
0
0 comments X

The pith

The FAST framework trains a neuroevolution potential on 500 configurations that reproduces the lambda peak in UO2 thermal expansion and non-Arrhenius anionic diffusion triggered by superionic transition.

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

The paper develops an efficient training strategy called FAST that combines superionic transition-targeted sampling with active learning to fine-tune a foundation model into a DFT-accurate neuroevolution potential for UO2 using only 500 configurations. This potential is shown to match experimental and ab initio trends in mechanical, defect, thermophysical, and diffusion properties across a wide temperature range. A sympathetic reader would care because it enables detailed atomic-scale simulation of nuclear fuel behavior under extreme conditions where direct experiments are difficult. The work focuses on capturing the specific anomalies that appear when the oxygen sublattice undergoes pre-melting.

Core claim

The neuroevolution potential trained via the FAST framework accurately reproduces both the lambda-peak in the linear thermal expansion coefficient and the non-Arrhenius temperature dependence of anionic diffusion that accompany the superionic transition in UO2; these anomalies originate from the pre-melting of the oxygen sublattice, which produces a kinetic decoupling between uranium and oxygen ions.

What carries the argument

The FAST (Fine-tuning via Active-learning and Superionic-Targeting) framework, which integrates superionic transition-targeted sampling with active learning to generate a compact 500-configuration dataset for fine-tuning a foundation model into a neuroevolution potential while enforcing proper treatment of uranium 5f correlations and the antiferromagnetic ground state.

If this is right

  • The same NEP can be used to compute mechanical properties, defect formation energies, and ionic diffusivities over an extended temperature range at near-DFT accuracy.
  • Large-scale molecular dynamics runs become feasible that directly link the microscopic pre-melting of the oxygen sublattice to the observed macroscopic anomalies.
  • The approach supplies a data-efficient route for modeling UO2 behavior under accident-relevant extreme conditions without requiring millions of DFT calculations.

Where Pith is reading between the lines

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

  • The same sampling-plus-active-learning recipe could be tested on other fluorite-structured superionic conductors to check whether 500 configurations remain sufficient.
  • The observed kinetic decoupling between cation and anion sublattices may offer a general design principle for tailoring ionic conductivity in related solid electrolytes.
  • Direct comparison of the predicted lambda-peak temperature and height against new high-temperature dilatometry measurements on UO2 single crystals would provide an immediate experimental test.

Load-bearing premise

The 500 configurations produced by superionic transition-targeted sampling and active learning are representative enough to capture the full range of atomic arrangements that control the thermophysical and kinetic anomalies.

What would settle it

A side-by-side comparison in which the trained NEP fails to produce a lambda-shaped peak in the linear thermal expansion coefficient or fails to recover the non-Arrhenius curvature in oxygen diffusivity near the superionic transition temperature would falsify the central claim.

read the original abstract

Uranium dioxide ($UO_2$) serves as the predominant nuclear fuel globally. Despite its widespread application, evaluating its mechanical, thermophysical, and species transport behaviors under extreme accident scenarios remains a formidable challenge for conventional experimental and computational methods. To address this, we develop a versatile machine learning interatomic potential (MLIP) for $UO_2$ by proposing an efficient training strategy, termed the "FAST" (Fine-tuning via Active-learning and Superionic-Targeting) framework. Our "FAST" framework integrates superionic transition-targeted sampling with active learning-enhanced exploration to efficiently construct a highly compact dataset comprising only 500 configurations for fine-tuning a foundation model. By rigorously accounting for the strong correlation of uranium 5f electrons and antiferromagnetic (AFM) ground state during DFT labeling, we train a robust DFT-level neuroevolution potential (NEP) for $UO_2$. We demonstrate that this NEP exhibits superior predictive capability for various physical properties, encompassing mechanical, defect, thermophysical, and ionic diffusion over an extensive temperature range. Moreover, this NEP accurately captures the anomalous thermophysical and kinetic behaviors triggered by superionic transition. Specifically, it reproduces both the $\lambda$-peak in linear thermal expansion coefficient (LTEC) and "non-Arrhenius" anionic diffusion. Crucially, NEP-based simulations elucidate the microscopic origins underlying these anomalies: the pre-melting of oxygen sublattice and resultant kinetic decoupling between U and O ions.

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 / 1 minor

Summary. The manuscript introduces the FAST (Fine-tuning via Active-learning and Superionic-Targeting) framework to construct a compact 500-configuration DFT-labeled dataset for fine-tuning a foundation model into a neuroevolution potential (NEP) for UO2. The central claim is that this NEP reproduces the λ-peak in the linear thermal expansion coefficient (LTEC) and non-Arrhenius anionic diffusion across the superionic transition, with the anomalies originating from oxygen sublattice pre-melting and kinetic decoupling between U and O ions; the NEP is also asserted to show superior performance on mechanical, defect, thermophysical, and diffusion properties over a wide temperature range.

Significance. If the reproduction of the emergent anomalies holds with quantitative validation, the work would demonstrate a data-efficient route to ML potentials that capture complex, non-fitted behaviors in actinide oxides under extreme conditions, with potential utility for nuclear fuel modeling where conventional methods struggle.

major comments (2)
  1. [Abstract] Abstract: the claim that the NEP 'accurately captures' the λ-peak in LTEC and non-Arrhenius diffusion is presented without any quantitative validation metrics, error bars, direct comparisons to experiment, or benchmarks against other potentials or DFT, leaving the strength of the reproduction unassessable.
  2. [FAST framework] FAST framework description (dataset construction): the assertion that the 500-configuration dataset obtained via superionic-targeted sampling plus active learning is representative of configurations relevant to the thermophysical and kinetic anomalies lacks supporting quantitative coverage metrics, such as the distribution of oxygen mean-square displacements, radial distribution functions, or phonon spectra sampled across the transition temperature; without these, it is unclear whether the observed anomalies in NEP MD arise from learned physics or from extrapolation.
minor comments (1)
  1. [Abstract] Abstract: the phrase 'non-Arrhenius' anionic diffusion should be accompanied by a brief indication of the observed deviation (e.g., specific temperature dependence or activation energy change).

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and constructive feedback on our manuscript arXiv:2606.21796. We address each of the major comments point by point below and outline the revisions we plan to make.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that the NEP 'accurately captures' the λ-peak in LTEC and non-Arrhenius diffusion is presented without any quantitative validation metrics, error bars, direct comparisons to experiment, or benchmarks against other potentials or DFT, leaving the strength of the reproduction unassessable.

    Authors: We agree that including quantitative metrics in the abstract would better allow readers to assess the strength of our claims. In the revised version, we will modify the abstract to incorporate specific validation metrics, such as the position and magnitude of the λ-peak in LTEC with comparison to experimental data, and the activation energies or deviation measures for the non-Arrhenius diffusion behavior. These details are provided in the main text with error bars from multiple simulations, but adding them to the abstract addresses this concern directly. revision: yes

  2. Referee: [FAST framework] FAST framework description (dataset construction): the assertion that the 500-configuration dataset obtained via superionic-targeted sampling plus active learning is representative of configurations relevant to the thermophysical and kinetic anomalies lacks supporting quantitative coverage metrics, such as the distribution of oxygen mean-square displacements, radial distribution functions, or phonon spectra sampled across the transition temperature; without these, it is unclear whether the observed anomalies in NEP MD arise from learned physics or from extrapolation.

    Authors: We acknowledge the value of providing explicit quantitative coverage metrics to demonstrate the representativeness of the dataset. While the active learning and superionic-targeted sampling were designed to ensure coverage of relevant configurations, we will include in the revised manuscript additional figures or supplementary information showing the distributions of oxygen mean-square displacements, radial distribution functions, and phonon spectra across the temperature range of the superionic transition. This will provide evidence that the key physics is captured within the training data rather than through extrapolation. revision: yes

Circularity Check

0 steps flagged

No significant circularity; emergent properties from MD are independent of training targets.

full rationale

The FAST framework constructs a 500-configuration DFT-labeled dataset via targeted sampling and active learning, then trains an NEP on local energies/forces. The λ-peak in LTEC and non-Arrhenius diffusion are obtained from subsequent MD runs and are collective, long-timescale observables not present in the per-atom training labels. No equations, self-citations, or ansatzes in the abstract reduce these outputs to the inputs by construction; the derivation chain remains self-contained against external DFT benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the representativeness of the actively learned and superionic-targeted 500 configurations plus the accuracy of the DFT labels that incorporate 5f electron correlations and antiferromagnetic order; no new physical entities are postulated.

free parameters (1)
  • Training dataset size
    The number 500 is chosen by hand as the target size for the compact fine-tuning set.
axioms (1)
  • domain assumption DFT calculations with explicit treatment of uranium 5f electron correlations and antiferromagnetic ground state provide sufficiently accurate labels for the configurations
    Invoked during the data labeling step of the FAST framework.

pith-pipeline@v0.9.1-grok · 5833 in / 1484 out tokens · 31309 ms · 2026-06-26T13:13:22.630460+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

111 extracted references

  1. [1]

    Periodic boundary conditions (PBCs) are imposed to all three mutually orthogonal directions of models

    Results: All MD simulations are conducted using the GPUMD software [46] with an integration ` time step of 1 fs. Periodic boundary conditions (PBCs) are imposed to all three mutually orthogonal directions of models. Notably, GPUMD is selected as the MD engine primarily due to its native compatibility with the NEP model, enabling exceptional computational ...

  2. [2]

    and stochastic cell rescaling (SCR) barostat [62] are adopted for the NVT and NPT ensembles, respectively. 3.1 Elastic and acoustic properties The elastic and acoustic properties of single-crystalline UO2 are evaluated via static calculations at 0 K using the Atomic Simulation Environment (ASE) [63] and calorine

  3. [3]

    λ-shape

    packages. Firstly, the equilibrium lattice constant ( a ) is determined by fully optimizing the bulk cell until the maximum atomic forces fall below 10-4 eV/Å. Based on this optimized configuration, the elastic stiffness tensor ( ijC ) is calculated, which is used to obtain Zener’s anisotropy factor (Eq.(1)). Furthermore, the long-wavelength acoustic phon...

  4. [4]

    non-Arrhenius

    Discussions: 4.1 Oxygen sublattice pre-melting: The pair distribution function (PDF) describes the probability density of finding a particle at a specific radial distance from a reference particle. Serving as a direct and fundamental measure, it accurately delineates the local structural order and coordination environment within a many-particle system. In...

    2000

  5. [5]

    non-Arrhenius

    Conclusions: ` (1) In this study, we propose a novel training workflow termed the “FAST” (Fine- tuning via Active-learning and Superionic-Targeting) framework. This training strategy is designed to develop a highly robust and versatile neuroevolution potential (NEP) for UO₂, which remains applicable across a broad temperature range up to its melting point...

  6. [6]

    Cappia, D

    F. Cappia, D. Pizzocri, M. Marchetti, A. Schubert, P. Van Uffelen, L. Luzzi, D. Papaioannou, R. Macián-Juan, V. Rondinella, Microhardness and Young's modulus of high burn-up UO2 fuel, J. Nucl. Mater. 479 (2016) 447-454

    2016

  7. [7]

    Henry, I

    R. Henry, I. Zacharie-Aubrun, T. Blay, N. Tarisien, S. Chalal, X. Iltis, J.-M. Gatt, C. Langlois, S. Meille, Irradiation effects on the fracture properties of UO2 fuels studied by micro-mechanical testing, J. Nucl. Mater. 536 (2020) 152179

    2020

  8. [8]

    Horlait, J

    D. Horlait, J. Domange, M.-L. Amany, M. Gérardin, M.-F. Barthe, G. Carlot, E. Gilabert, Experimental investigation of Kr diffusion in UO2+x: slight deviations from stoichiometry, significant effects on diffusion kinetics and mechanisms, J. Nucl. Mater. 574 (2023) 154191

    2023

  9. [9]

    Saidy, W.H

    M. Saidy, W.H. Hocking, J.F. Mouris, P. Garcia, G. Carlot, B. Pasquet, Thermal diffusion of iodine in UO2 and UO2+ x, J. Nucl. Mater. 372(2-3) (2008) 405-415

    2008

  10. [10]

    Harding, D

    J. Harding, D. Martin, A recommendation for the thermal conductivity of UO2, J. Nucl. Mater. 166(3) (1989) 223-226

    1989

  11. [11]

    Lucuta, I

    P. Lucuta, I. Hastings, A pragmatic approach to modelling thermal conductivity of irradiated UO2 fuel: review and recommendations, J. Nucl. Mater. 232(2-3) (1996) 166-180

    1996

  12. [12]

    Dorado, B

    B. Dorado, B. Amadon, M. Freyss, M. Bertolus, DFT+ U calculations of the ground state and metastable states of uranium dioxide, Phys. Rev. B 79(23) (2009) 235125

    2009

  13. [13]

    Vathonne, J

    E. Vathonne, J. Wiktor, M. Freyss, G. Jomard, M. Bertolus, DFT+ U investigation of charged point defects and clusters in UO2, J. Phys. Condens. Matter 26(32) (2014) 325501

    2014

  14. [14]

    Dorado, D.A

    B. Dorado, D.A. Andersson, C.R. Stanek, M. Bertolus, B.P. Uberuaga, G. Martin, M. Freyss, P. Garcia, First-principles calculations of uranium diffusion in uranium dioxide, Phys. Rev. B 86(3) (2012) 035110

    2012

  15. [15]

    Stippell, L

    E. Stippell, L. Alzate-Vargas, K.N. Subedi, R.M. Tutchton, M.W. Cooper, S. Tretiak, T. Gibson, R.A. Messerly, Building a DFT+ U machine learning interatomic potential for uranium dioxide, ` Artif. Intell. Chem. 2(1) (2024) 100042

    2024

  16. [16]

    Basak, A

    C. Basak, A. Sengupta, H. Kamath, Classical molecular dynamics simulation of UO2 to predict thermophysical properties, J. Alloy. Compd. 360(1-2) (2003) 210-216

    2003

  17. [17]

    Morelon, D

    N.-D. Morelon, D. Ghaleb, J.-M. Delaye, L. Van Brutzel, A new empirical potential for simulating the formation of defects and their mobility in uranium dioxide, Philos. Mag. 83(13) (2003) 1533-1555

    2003

  18. [18]

    Yakub, C

    E. Yakub, C. Ronchi, D. Staicu, Molecular dynamics simulation of premelting and melting phase transitions in stoichiometric uranium dioxide, J. Chem. Phys. 127(9) (2007)

    2007

  19. [19]

    Cooper, M

    M. Cooper, M. Rushton, R. Grimes, A many-body potential approach to modelling the thermomechanical properties of actinide oxides, J. Phys. Condens. Matter 26(10) (2014) 105401

    2014

  20. [20]

    Soulié, J.-P

    A. Soulié, J.-P. Crocombette, A. Kraych, F. Garrido, G. Sattonnay, E. Clouet, Atomistically- informed thermal glide model for edge dislocations in uranium dioxide, Acta Mater. 150 (2018) 248-261

    2018

  21. [21]

    Liu, M.W.D

    X.-Y. Liu, M.W.D. Cooper, K.J. McClellan, J.C. Lashley, D.D. Byler, B. Bell, R. Grimes, C.R. Stanek, D.A. Andersson, Molecular dynamics simulation of thermal transport in UO₂ containing uranium, oxygen, and fission-product defects, Phys. Rev. Appl. 6(4) (2016) 044015

    2016

  22. [22]

    Balboa, L

    H. Balboa, L. Van Brutzel, A. Chartier, Y. Le Bouar, Assessment of empirical potential for MOX nuclear fuels and thermomechanical properties, J. Nucl. Mater. 495 (2017) 67-77

    2017

  23. [23]

    Zhuang, G

    F. Zhuang, G. Yan, H. Chen, Y. Zhang, W. Yu, M. Xu, S. Shen, Superionic Transition-Induced Diffusional Creep Behavior Transitions in Nanocrystalline (U₁₋ₓ, Thₓ)O₂ Nuclear Fuels, Acta Mech. Solida Sin. (2026) 1-15

    2026

  24. [24]

    Zhuang, G

    F. Zhuang, G. Yan, W. Yu, M. Xu, S. Shen, The high‐temperature creep behavior transition in stoichiometric nanocrystalline (Puₓ, U₁₋ₓ)O₂ mixed oxides, J. Am. Ceram. Soc. 108(4) (2025) e20275

    2025

  25. [25]

    C. Chil, A. Pivano, B. Michel, E. Bourasseau, Fracture behavior of grain boundaries in (U, Pu) O₂ fuels: An atomistic study, J. Am. Ceram. Soc. 109(1) (2026) e70326

    2026

  26. [26]

    Behler, Perspective: Machine learning potentials for atomistic simulations, J

    J. Behler, Perspective: Machine learning potentials for atomistic simulations, J. Chem. Phys. 145(17) (2016)

    2016

  27. [27]

    Dubois, J

    E.T. Dubois, J. Tranchida, J. Bouchet, J.-B. Maillet, Atomistic simulations of nuclear fuel UO₂ with machine learning interatomic potentials, Phys. Rev. Mater. 8(2) (2024) 025402

    2024

  28. [28]

    Konashi, N

    K. Konashi, N. Kato, K. Mori, K. Kurosaki, Neural network potential for molecular dynamics calculation of UO2, J. Nucl. Mater. 607 (2025) 155660

    2025

  29. [29]

    Zhong, L

    J. Zhong, L. Zhang, T. Bo, Active learning-enhanced neuroevolution potential for predictive modeling of UO₂ thermophysical properties, Phys. Chem. Chem. Phys. 28(1) (2026) 671-682

    2026

  30. [30]

    Hurley, A

    D.H. Hurley, A. El-Azab, M.S. Bryan, M.W. Cooper, C.A. Dennett, K. Gofryk, L. He, M. Khafizov, G.H. Lander, M.E. Manley, Thermal energy transport in oxide nuclear fuel, Chem. Rev. 122(3) (2022) 3711-3762

    2022

  31. [31]

    Zhang, X

    H. Zhang, X. Wang, A. Chremos, J.F. Douglas, Superionic UO2: A model anharmonic crystalline material, J. Chem. Phys. 150(17) (2019)

    2019

  32. [32]

    Naito, High-temperature heat capacities of UO2 and doped UO2, J

    K. Naito, High-temperature heat capacities of UO2 and doped UO2, J. Nucl. Mater. 167 (1989) 30-35

    1989

  33. [33]

    Hiernaut, G

    J. Hiernaut, G. Hyland, C. Ronchi, Premelting transition in uranium dioxide, Int. J. Thermophys. 14(2) (1993) 259-283

    1993

  34. [34]

    Lunev, B.A

    A.V. Lunev, B.A. Tarasov, A classical molecular dynamics study of the correlation between the ` Bredig transition and thermal conductivity of stoichiometric uranium dioxide, J. Nucl. Mater. 415(2) (2011) 217-221

    2011

  35. [35]

    Fossati, P.A

    P.C. Fossati, P.A. Burr, M.W. Cooper, C.O. Galvin, R.W. Grimes, Superionic transition in uranium dioxide: Insights from molecular dynamics and lattice dynamics simulations, Phys. Rev. Mater. 8(11) (2024) 115404

    2024

  36. [36]

    P. Ying, C. Qian, R. Zhao, Y. Wang, K. Xu, F. Ding, S. Chen, Z. Fan, Advances in modeling complex materials: The rise of neuroevolution potentials, Chem. Phys. Rev. 6(1) (2025)

    2025

  37. [37]

    X. Liu, K. Zeng, Z. Luo, Y. Wang, T. Zhao, Z. Xu, Fine-tuning universal machine-learned interatomic potentials: A tutorial on methods and applications, J. Appl. Phys. 139(4) (2026)

    2026

  38. [38]

    Liang, K

    T. Liang, K. Xu, E. Lindgren, Z. Chen, R. Zhao, J. Liu, E. Berger, B. Tang, B. Zhang, Y. Wang, NEP89: Universal neuroevolution potential for inorganic and organic materials across 89 elements, arXiv preprint arXiv:2504.21286 (2025)

    arXiv 2025

  39. [39]

    Z. Fan, Y. Wang, P. Ying, K. Song, J. Wang, Y. Wang, Z. Zeng, K. Xu, E. Lindgren, J.M. Rahm, GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations, J. Chem. Phys. 157(11) (2022)

    2022

  40. [40]

    Galvin, M

    C.O. Galvin, M. Machida, H. Nakamura, D.A. Andersson, M.W. Cooper, Correlations for the specific heat capacity of (UₓPu₁₋ₓ)₁₋ᵧGdᵧO₂₋z derived from molecular dynamics, J. Nucl. Mater. 572 (2022) 154028

    2022

  41. [41]

    Ushakov, A

    S.V. Ushakov, A. Navrotsky, R.J. Weber, J.C. Neuefeind, Structure and thermal expansion of YSZ and La₂Zr₂O₇ above 1500 °C from neutron diffraction on levitated samples, J. Am. Ceram. Soc. 98(10) (2015) 3381-3388

    2015

  42. [42]

    Yin, J.-T

    Y.-C. Yin, J.-T. Yang, J.-D. Luo, G.-X. Lu, Z. Huang, J.-P. Wang, P. Li, F. Li, Y.-C. Wu, T. Tian, A LaCl₃-based lithium superionic conductor compatible with lithium metal, Nature 616(7955) (2023) 77-83

    2023

  43. [43]

    K. Jun, Y. Chen, G. Wei, X. Yang, G. Ceder, Diffusion mechanisms of fast lithium-ion conductors, Nat. Rev. Mater. 9(12) (2024) 887-905

    2024

  44. [44]

    Y. Kato, S. Hori, T. Saito, K. Suzuki, M. Hirayama, A. Mitsui, M. Yonemura, H. Iba, R. Kanno, High-power all-solid-state batteries using sulfide superionic conductors, Nat. Energy 1(4) (2016) 16030

    2016

  45. [45]

    W. Liu, Y. Zhou, Energy transport in superionic crystals, Phys. Rev. Lett. 134(14) (2025) 146301

    2025

  46. [46]

    B. Li, H. Wang, Y. Kawakita, Q. Zhang, M. Feygenson, H. Yu, D. Wu, K. Ohara, T. Kikuchi, K. Shibata, Liquid-like thermal conduction in intercalated layered crystalline solids, Nat. Mater. 17(3) (2018) 226-230

    2018

  47. [47]

    Millot, F

    M. Millot, F. Coppari, J.R. Rygg, A. Correa Barrios, S. Hamel, D.C. Swift, J.H. Eggert, Nanosecond X-ray diffraction of shock-compressed superionic water ice, Nature 569(7755) (2019) 251-255

    2019

  48. [48]

    Y. He, W. Zhang, Q. Hu, S. Sun, J. Hu, D. Liu, L. Zhou, L. Dai, D.Y. Kim, S.A. Redfern, Absence of dehydration due to superionic transition at Earth’s core-mantle boundary, Sci. Adv. 12(5) (2026) eaeb3006

    2026

  49. [49]

    C. Chen, Y. Li, R. Zhao, Z. Liu, Z. Fan, G. Tang, Z. Wang, NepTrain and NepTrainKit: Automated active learning and visualization toolkit for neuroevolution potentials, Comput. Phys. Commun. (2025) 109859

    2025

  50. [50]

    Thompson, H.M

    A.P. Thompson, H.M. Aktulga, R. Berger, D.S. Bolintineanu, W.M. Brown, P.S. Crozier, P.J. ` In't Veld, A. Kohlmeyer, S.G. Moore, T.D. Nguyen, LAMMPS-a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales, Comput. Phys. Commun. 271 (2022) 108171

    2022

  51. [51]

    K. Xu, H. Bu, S. Pan, E. Lindgren, Y. Wu, Y. Wang, J. Liu, K. Song, B. Xu, Y. Li, GPUMD 4.0: a high‐performance molecular dynamics package for versatile materials simulations with machine ‐learned potentials, Mater. Genome Eng. Adv. 3(3) (2025) e70028

    2025

  52. [52]

    Cooper, S

    M. Cooper, S. Murphy, M. Rushton, R. Grimes, Thermophysical properties and oxygen transport in the (Uₓ, Pu₁₋ₓ)O₂ lattice, J. Nucl. Mater. 461 (2015) 206-214

    2015

  53. [53]

    Z. Yan, D. Li, X. Wu, Z. Liu, C. Hua, B. Situ, H. Yang, S. Tang, B. Tang, Z. Wang, GPUMDkit: A User‐Friendly Toolkit for GPUMD and NEP, Mater. Genome Eng. Adv. (2026) e70074

    2026

  54. [54]

    Lysogorskiy, A

    Y. Lysogorskiy, A. Bochkarev, M. Mrovec, R. Drautz, Active learning strategies for atomic cluster expansion models, Phys. Rev. Mater. 7(4) (2023) 043801

    2023

  55. [55]

    Zhang, D.-Y

    L. Zhang, D.-Y. Lin, H. Wang, R. Car, W. E, Active learning of uniformly accurate interatomic potentials for materials simulation, Phys. Rev. Mater. 3(2) (2019) 023804

    2019

  56. [56]

    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(16) (1996) 11169

    1996

  57. [57]

    Njifon, E

    I.C. Njifon, E. Torres, Atomistic investigation of uranium oxycarbide (UCO) phase stability at finite temperatures: DFT+ U and AIMD+ U approaches, J. Nucl. Mater. 554 (2021) 153046

    2021

  58. [58]

    Kresse, D

    G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B 59(3) (1999) 1758

    1999

  59. [59]

    Perdew, K

    J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77(18) (1996) 3865

    1996

  60. [60]

    Monkhorst, J.D

    H.J. Monkhorst, J.D. Pack, Special points for Brillouin-zone integrations, Phys. Rev. B 13(12) (1976) 5188

    1976

  61. [61]

    Dudarev, D.N

    S. Dudarev, D.N. Manh, A. Sutton, Effect of Mott-Hubbard correlations on the electronic structure and structural stability of uranium dioxide, Philos. Mag. B 75(5) (1997) 613-628

    1997

  62. [62]

    Allen, G.W

    J.P. Allen, G.W. Watson, Occupation matrix control of d-and f-electron localisations using DFT+ U, Phys. Chem. Chem. Phys. 16(39) (2014) 21016-21031

    2014

  63. [63]

    Claisse, M

    A. Claisse, M. Klipfel, N. Lindbom, M. Freyss, P. Olsson, GGA+ U study of uranium mononitride: A comparison of the U-ramping and occupation matrix schemes and incorporation energies of fission products, J. Nucl. Mater. 478 (2016) 119-124

    2016

  64. [64]

    Ziegler, J.P

    J.F. Ziegler, J.P. Biersack, The stopping and range of ions in matter, Treat. Heavy-Ion Sci. 6, Springer, 1985, pp. 93-129

    1985

  65. [65]

    J. Liu, J. Byggmästar, Z. Fan, P. Qian, Y. Su, Large-scale machine-learning molecular dynamics simulation of primary radiation damage in tungsten, Phys. Rev. B 108(5) (2023) 054312

    2023

  66. [66]

    Bussi, D

    G. Bussi, D. Donadio, M. Parrinello, Canonical sampling through velocity rescaling, J. Chem. Phys. 126(1) (2007)

    2007

  67. [67]

    Bernetti, G

    M. Bernetti, G. Bussi, Pressure control using stochastic cell rescaling, J. Chem. Phys. 153(11) (2020)

    2020

  68. [68]

    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, The atomic simulation environment — a Python library for working with atoms, J. Phys. Condens. Matter 29(27) (2017) 273002

    2017

  69. [69]

    Lindgren, M

    E. Lindgren, M. Rahm, E. Fransson, F. Eriksson, N. Österbacka, Z. Fan, P. Erhart, calorine: A Python package for constructing and sampling neuroevolution potential models, J. Open Source ` Softw. 9(95) (2024) 6264

    2024

  70. [70]

    Fink, Thermophysical properties of uranium dioxide, J

    J. Fink, Thermophysical properties of uranium dioxide, J. Nucl. Mater. 279(1) (2000) 1-18

    2000

  71. [71]

    Brando, E

    E. Brando, E. Dubois, M. Rochedy, J. Bouchet, F. Bruneval, J. Crocombette, J. Tranchida, Atomistic investigation of the temperature and stoichiometry dependence of the heat-capacity in nuclear fuel UO2+ x, J. Nucl. Mater. (2025) 156280

    2025

  72. [72]

    J.-L. Chen, N. Kaltsoyannis, DFT+ U study of uranium dioxide and plutonium dioxide with occupation matrix control, J. Phys. Chem. C 126(27) (2022) 11426-11435

    2022

  73. [73]

    Z.-G. Mei, M. Stan, J. Yang, First-principles study of thermophysical properties of uranium dioxide, J. Alloy. Compd. 603 (2014) 282-286

    2014

  74. [74]

    Wachtman Jr, M

    J. Wachtman Jr, M. Wheat, H. Anderson, J. Bates, Elastic constants of single crystal UO₂ at 25 °C, J. Nucl. Mater. 16(1) (1965) 39-41

    1965

  75. [75]

    Leinders, T

    G. Leinders, T. Cardinaels, K. Binnemans, M. Verwerft, Accurate lattice parameter measurements of stoichiometric uranium dioxide, J. Nucl. Mater. 459 (2015) 135-142

    2015

  76. [76]

    Fritz, Elastic properties of UO₂ at high pressure, J

    I. Fritz, Elastic properties of UO₂ at high pressure, J. Appl. Phys. 47(10) (1976) 4353-4358

    1976

  77. [77]

    Parrinello, A

    M. Parrinello, A. Rahman, Strain fluctuations and elastic constants, J. Chem. Phys. 76(5) (1982) 2662-2666

    1982

  78. [78]

    Hutchings, High-temperature studies of UO₂ and ThO₂ using neutron scattering techniques, J

    M.T. Hutchings, High-temperature studies of UO₂ and ThO₂ using neutron scattering techniques, J. Chem. Soc. Faraday Trans. 2 83(7) (1987) 1083-1103

    1987

  79. [79]

    Martyna, D.J

    G.J. Martyna, D.J. Tobias, M.L. Klein, Constant pressure molecular dynamics algorithms, J. Chem. Phys. 101(4177) (1994) 10.1063

    1994

  80. [80]

    Zhang, X.-Y

    Y. Zhang, X.-Y. Liu, P.C. Millett, M. Tonks, D.A. Andersson, B. Biner, Crack tip plasticity in single crystal UO₂: Atomistic simulations, J. Nucl. Mater. 430(1-3) (2012) 96-105

    2012

Showing first 80 references.