A deep neural network potential model for transition metal diborides

Abstract

Transition metal diborides (TMB₂s) are renowned for their high melting point and exceptional wear, corrosion, and erosion resistance, making them promising candidate materials for applications in extreme environments. As such, there is an urgent need for reliable material design tools for TMB₂s to improve efficiency in developing new materials. To address this need, we have developed a domain-specific medium-scale interatomic potential model for TMB₂s that encompasses elements Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, and B. The prediction errors in energy and force of our model are 8.8 meV/atom and 387 meV/Å, respectively. Furthermore, the model demonstrates high accuracy in predicting various material properties, including lattice parameters, elastic constants, equations of states, and melting points, as well as grain boundary segregations. By providing a reliable and efficient tool for material design, this model will play a crucial role in the development of new, high-performance TMB₂s for use in extreme environments.

Keywords

Machine learning potential / atomic-scale simulation / transition metal diborides / melting point / grain boundary segregation

References

[1] Opeka MM,Zaykoski JA.Oxidation-based materials selection for 2000 °C + hypersonic aerosurfaces: theoretical considerations and historical experience.J Mater Sci2004;39:5887-904

[2] Fahrenholtz WG,Talmy IG.Refractory diborides of zirconium and hafnium.J Am Ceram Soc2007;90:1347-64

[3] Ni D,Zhang J.Advances in ultra-high temperature ceramics, composites, and coatings.J Adv Ceram2022;11:1-56

[4] Zou J,Hu CF.Strong ZrB₂-SiC-WC ceramics at 1600 °C.J Am Ceram Soc2012;95:874-8

[5] Wang Y,Jiao X.The highest melting point material: searched by Bayesian global optimization with deep potential molecular dynamics.J Adv Ceram2023;12:803-14

[6] Ma HB,Liu JX,Zhang GJ.Microstructures, solid solution formation and high-temperature mechanical properties of ZrB₂ ceramics doped with 5vol.% WC.Mater Design2015;81:133-40

[7] Silvestroni L,Migliori A.A simple route to fabricate strong boride hierarchical composites for use at ultra-high temperature.Compos Part B Eng2020;183:107618

[8] Silvestroni L,Vinokurov V,Grigoriev O.Core-shell structure: an effective feature for strengthening ZrB₂ ceramics.Scr Mater2019;160:1-4

[9] Silvestroni L,Sciti D,Savino R.Effect of hypersonic flow chemical composition on the oxidation behavior of a super-strong UHTC.Corros Sci2019;159:108125

[10] Chen H,Dai FZ,Zhou Y.Porous high entropy (Zr_0.2Hf_0.2Ti_0.2Nb_0.2Ta_0.2)B₂: a novel strategy towards making ultrahigh temperature ceramics thermal insulating.J Mater Sci Technol2019;35:2404-8

[11] Li RZ,Yuan JH.Enhanced high-temperature strength in textured (Ti_1/3Zr_1/3Hf_1/3)B₂ medium-entropy ceramics via strong magnetic field.J Am Ceram Soc2023;106:5440-53

[12] Yang QQ,Wu P.Ultra-high strength medium-entropy (Ti,Zr,Ta)C ceramics at 1800 °C by consolidating a core-shell structured powder.J Am Ceram Soc2022;105:823-9

[13] Feng L,Fahrenholtz WG.Strength of single-phase high-entropy carbide ceramics up to 2300 °C.J Am Ceram Soc2021;104:419-27

[14] Gild J,Harrington T.High-entropy metal diborides: a new class of high-entropy materials and a new type of ultrahigh temperature ceramics.Sci Rep2016;6:37946 PMCID:PMC5126569

[15] Xiang H,Dai F.High-entropy ceramics: present status, challenges, and a look forward.J Adv Ceram2021;10:385-441

[16] Wen Z,Liu Y,Yu H.Ultrastrong and high thermal insulating porous high-entropy ceramics up to 2000 °C.Adv Mater2024;36:e2311870

[17] Zhang L,Wang H,Weinan E.Deep potential molecular dynamics: a scalable model with the accuracy of quantum mechanics.Phys Rev Lett2018;120:143001

[18] Behler J.Generalized neural-network representation of high-dimensional potential-energy surfaces.Phys Rev Lett2007;98:146401

[19] Bartók AP,Kondor R.Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons.Phys Rev Lett2010;104:136403

[20] Deng B,Jun K.CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling.Nat Mach Intell2023;5:1031-41

[21] Takamoto S,Motoki D.Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements.Nat Commun2022;13:2991 PMCID:PMC9151783

[22] Chen C.A universal graph deep learning interatomic potential for the periodic table.Nat Comput Sci2022;2:718-28

[23] Dai FZ,Sun Y,Zhou Y.Theoretical prediction on thermal and mechanical properties of high entropy (Zr_0.2Hf_0.2Ti_0.2Nb_0.2Ta_0.2)C by deep learning potential.J Mater Sci Technol2020;43:168-74

[24] Dai FZ,Wen B,Zhou Y.Temperature dependent thermal and elastic properties of high entropy (Ti_0.2Zr_0.2Hf_0.2Nb_0.2Ta_0.2)B₂: molecular dynamics simulation by deep learning potential.J Mater Sci Technol2021;72:8-15

[25] Dai FZ,Sun Y,Xiang H.Grain boundary segregation induced strong UHTCs at elevated temperatures: a universal mechanism from conventional UHTCs to high entropy UHTCs.J Mater Sci Technol2022;123:26-33

[26] Dai FZ,Xiang H.Grain boundary strengthening in ZrB₂ by segregation of W: atomistic simulations with deep learning potential.J Eur Ceram Soc2020;40:5029-36

[27] Zhang D,Zhang X. DPA-2: towards a universal large atomic model for molecular and material simulation. arXiv. [Preprint.] Dec 24, 2023 [accessed 2024 Aug 8]. Available from: https://arxiv.org/abs/2312.15492.

[28] Zhang D,Dai FZ.Pretraining of attention-based deep learning potential model for molecular simulation.npj Comput Mater2024;10:94

[29] Zhang L,Wang H,Weinan E.Active learning of uniformly accurate interatomic potentials for materials simulation.Phys Rev Mater2019;3:023804

[30] Kresse G.Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set.Comput Mater Sci1996;6:15-50

[31] Blöchl PE.Projector augmented-wave method.Phys Rev B Condens Matter1994;50:17953-79

[32] Perdew JP,Ernzerhof M.Generalized gradient approximation made simple.Phys Rev Lett1996;77:3865-8

[33] Pack JD.“Special points for Brillouin-zone integrations” - a reply.Phys Rev B1977;16:1748-9

[34] Wang H,Han J.DeePMD-kit: a deep learning package for many-body potential energy representation and molecular dynamics.Comput Phys Commun2018;228:178-84

[35] Zeng J,Lu D.DeePMD-kit v2: a software package for deep potential models.J Chem Phys2023;159:054801

[36] Plimpton S.Fast parallel algorithms for short-range molecular dynamics.J Comput Phys1995;117:1-19

[37] Ma HB,Liu HL,Lu Y.Effect of WC or ZrC addition on thermal residual stresses in ZrB₂-SiC ceramics.Mater Design2016;110:340-5

[38] Ruby E,Chang Y. Ternary phase equilibria in transition metal-boron-carbon-silicon systems. Available from: https://www.semanticscholar.org/paper/Ternary-phase-equilibria-in-transition-systems-Ruby-Windsch/2d30dae05430ca0696e725f9a05b98aa47a02e88. [Last accessed on 8 Aug 2024]

[39] Ma HB,Zhu JT,Zhang GJ.Segregation of tungsten atoms at ZrB₂ grain boundaries in strong ZrB₂-SiC-WC ceramics.Scr Mater2018;157:76-80

[40] Su W,Zhang W,Wang Y.Insights into grain boundary segregation and solubility limit of Cr in (TiZrNbTaCr)C.J Mater Sci Technol2023;139:1-9

[41] Su W,Huo S,Wang Y.Fracture mode transition from intergranular to transgranular in (TiZrNbTaCr)C: the grain boundary purification effect of Cr carbide.J Eur Ceram Soc2024;44:1881-9

[42] Dai FZ,Sun W.Segregation of solute atoms (Y, Nb, Ta, Mo and W) in ZrB₂ grain boundaries and their effects on grain boundary strengths: a first-principles investigation.Acta Mater2017;127:312-8

[43] Silvestroni L,Fahrenholtz WG.Super-strong materials for temperatures exceeding 2000 °C.Sci Rep2017;7:40730 PMCID:PMC5244373

[44] Mo P,Zhao D.Accurate and efficient molecular dynamics based on machine learning and non von Neumann architecture.npj Comput Mater2022;8:107

[45] Jain A,Hautier G.Commentary: the materials project: a materials genome approach to accelerating materials innovation.APL Mater2013;1:011002

[46] Curtarolo S,Wang S.AFLOWLIB.ORG: a distributed materials properties repository from high-throughput ab initio calculations.Comput Mater Sci2012;58:227-35

[47] Saal JE,Aykol M,Wolverton C.Materials design and discovery with high-throughput density functional theory: the Open Quantum Materials Database (OQMD).JOM2013;65:1501-9