Abstract
Phase transition is one of the most important phenomena in nature and plays a central role in materials design. All phase transitions are characterized by suitable order parameters, including the order–disorder phase transition. However, finding a representative order parameter for complex systems is non-trivial, such as for high-entropy alloys. Given the strength of dimensionality reduction of a variational autoencoder (VAE), we introduce a VAE-based order parameter. We propose that the Manhattan distance in the VAE latent space can serve as a generic order parameter for order–disorder phase transitions. The physical properties of our order parameter are quantitatively interpreted and demonstrated by multiple refractory high-entropy alloys. Using this order parameter, a generally applicable alloy design concept is proposed by mimicking the natural mixing process of elements. Our physically interpretable VAE-based order parameter provides a computational technique for understanding chemical ordering in alloys, which can facilitate the development of rational alloy design strategies.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$99.00 per year
only $8.25 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The sample training and validation data, along with a pre-trained model and VASP input files (version 5.4.4) for Ti38V16Nb23Hf24, are available at https://doi.org/10.6084/m9.figshare.14417225.v449. Source data are provided with this paper.
Code availability
The VAE model training and order parameter inference codes are available at https://code.ornl.gov/jqyin/deepthermo. The training code along with sample data can also be found on our Code Ocean capsule50. The figures are plotted with the notebook at https://code.ornl.gov/jqyin/deepthermo/-/blob/master/utils/hea_vae_analysis.ipynb.
References
Gibbs, J. W. On the equilibrium of heterogeneous substances. Am. J. Sci. s3-16, 441–458 (1878).
Thomas, H. Über Widerstandslegierungen. Z. Phys. 129, 219–232 (1951).
Sourmail, T. Near equiatomic FeCo alloys: constitution, mechanical and magnetic properties. Prog. Mater. Sci. 50, 816–880 (2005).
Yang, T. et al. Ultrahigh-strength and ductile superlattice alloys with nanoscale disordered interfaces. Science 369, 427–432 (2020).
Chen, H. et al. Unprecedented non-hysteretic superelasticity of [001]-oriented NiCoFeGa single crystals. Nat. Mater. 19, 712–718 (2020).
Yeh, J.-W. et al. Nanostructured high-entropy alloys with multiple principal elements: novel alloy design concepts and outcomes. Adv. Eng. Mater. 6, 299–303 (2004).
Cantor, B., Chang, I., Knight, P. & Vincent, A. Microstructural development in equiatomic multicomponent alloys. Mater. Sci. Eng. A 375–377, 213–218 (2004).
Zhang, Y. et al. Microstructures and properties of high-entropy alloys. Progr. Mater. Sci. 61, 1–93 (2014).
Pei, Z. An overview of modeling the stacking faults in lightweight and high-entropy alloys: theory and application. Mater. Sci. Eng. A 737, 132–150 (2018).
Pei, Z., Li, R., Gao, M. C. & Stocks, G. M. Statistics of the NiCoCr medium-entropy alloy: novel aspects of an old puzzle. npj Comput. Mater. 6, 122 (2020).
Pei, Z., Yin, J., Hawk, J. A., Alman, D. E. & Gao, M. C. Machine-learning informed prediction of high-entropy solid solution formation: beyond the Hume–Rothery rules. npj Comput. Mater. 6, 50 (2020).
Zhang, R. et al. Short-range order and its impact on the CrCoNi medium-entropy alloy. Nature 581, 283–287 (2020).
Landau, L. D. On the theory of phase transitions. Zh. Eksp. Teor. Fiz. 7, 19–32 (1937).
Owen, L., Playford, H., Stone, H. & Tucker, M. A new approach to the analysis of short-range order in alloys using total scattering. Acta Mater. 115, 155–166 (2016).
Cowley, J. M. An approximate theory of order in alloys. Phys. Rev. 77, 669–675 (1950).
Kingma, D. P. & Welling, M. Auto-encoding variational Bayes. In Proc. 2nd International Conference on Learning Representations (ICLR 2014) (eds Bengio, Y. and LeCun, Y.) (ICLR, 2014); https://arxiv.org/pdf/1312.6114v10.pdf
Doersch, C. Tutorial on variational autoencoders. Preprint at https://arxiv.org/abs/1606.05908 (2016).
van Nieuwenburg, E. P. L., Liu, Y.-H. & Huber, S. D. Learning phase transitions by confusion. Nat. Phys. 13, 435–439 (2017).
Lee, H. et al. DeepDriveMD: deep-learning driven adaptive molecular simulations for protein folding. In Proc. Third IEEE/ACM Workshop on Deep Learning on Supercomputers 2019, 12–19 (IEEE, 2019); https://doi.org/10.1109/DLS49591.2019.00007
Pol, A. A., Berger, V., Germain, C., Cerminara, G. & Pierini, M. Anomaly detection with conditional variational autoencoders. In Proc. 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA) 1651–1657 (IEEE, 2019); https://doi.org/10.1109/ICMLA.2019.00270
Wetzel, S. J. Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders. Phys. Rev. E 96, 022140 (2017).
Walker, N., Tam, K.-M. & Jarrell, M. Deep learning on the 2-dimensional Ising model to extract the crossover region with a variational autoencoder. Sci. Rep. 10, 13047 (2020).
Alexandrou, C., Athenodorou, A., Chrysostomou, C. & Paul, S. The critical temperature of the 2D-Ising model through deep learning autoencoders. Eur. Phys. J. B 93, 226 (2020).
Hu, W., Singh, R. R. P. & Scalettar, R. T. Discovering phases, phase transitions and crossovers through unsupervised machine learning: a critical examination. Phys. Rev. E 95, 062122 (2017).
Wang, R. et al. Nuclear liquid–gas phase transition with machine learning. Phys. Rev. Res. 2, 043202 (2020).
Käming, N. et al. Unsupervised machine learning of topological phase transitions from experimental data. Mach. Learn. Sci. Technol. 2, 035037 (2021).
Aggarwal, C. C., Hinneburg, A. & Keim, D. A. in Database Theory—ICDT 2001 (eds Van den Bussche, J. & Vianu, V.) 420–434 (Springer, 2001).
Wei, S. et al. Natural-mixing guided design of refractory high-entropy alloys with as-cast tensile ductility. Nat. Mater. 19, 1175–1181 (2020).
Wang, F. et al. Multiplicity of dislocation pathways in a refractory multiprincipal element alloy. Science 370, 95–101 (2020).
Kostiuchenko, T., Körmann, F., Neugebauer, J. & Shapeev, A. Impact of lattice relaxations on phase transitions in a high-entropy alloy studied by machine-learning potentials. npj Comput. Math. 5, 55 (2019).
Liu, X. et al. Monte Carlo simulation of order–disorder transition in refractory high entropy alloys: a data-driven approach. Comput. Mater. Sci. 187, 110135 (2021).
Feng, R. et al. High-throughput design of high-performance lightweight high-entropy alloys. Nat. Commun. 12, 4329 (2021).
Santodonato, L., Liaw, P., Unocic, R., Bei, H. & Morris, J. Predictive multiphase evolution in Al-containing high-entropy alloys. Nat. Commun. 9, 4520 (2018).
Körmann, F., Ruban, A. V. & Sluiter, M. H. Long-ranged interactions in bcc NbMoTaW high-entropy alloys. Mater. Res. Lett. 5, 35–40 (2017).
Senkov, O., Wilks, G., Scott, J. & Miracle, D. Mechanical properties of Nb25Mo25Ta25W25 and V20Nb20Mo20Ta20W20 refractory high entropy alloys. Intermetallics 19, 698–706 (2011).
Ducastelle, F. in Nanoalloys: Synthesis, Structure and Properties (eds Alloyeau, D. et al.) 197–225 (Springer, 2012).
Singh, P., Smirnov, A. V. & Johnson, D. D. Atomic short-range order and incipient long-range order in high-entropy alloys. Phys. Rev. B 91, 224204 (2015).
Hohenberg, P. & Kohn, W. Inhomogeneous electron gas. Phys. Rev. 136, B864 (1964).
Kohn, W. & Sham, L. J. Self-consistent equations including exchange and correlation effects.Phys. Rev. 140, A1133 (1965).
Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B 50, 17953–17979 (1994).
Monkhorst, H. J. & Pack, J. D. Special points for Brillouin-zone integrations. Phys. Rev. B 13, 5188–5192 (1976).
Zhang, J. et al. Robust data-driven approach for predicting the configurational energy of high entropy alloys. Mater. Design 185, 108247 (2020).
Swendsen, R. H. & Wang, J.-S. Replica Monte Carlo simulation of spin-glasses. Phys. Rev. Lett. 57, 2607–2609 (1986).
Misra, P. K. in Physics of Condensed Matter (ed. Misra, P. K.) Ch. 1, 1–35 (Academic, 2012); https://doi.org/10.1016/B978-0-12-384954-0.00001-3
Vasudevan, R. K. et al. Mapping mesoscopic phase evolution during E-beam induced transformations via deep learning of atomically resolved images. npj Comput. Mater. 4, 760 (2018).
Santodonato, L. J. et al. Deviation from high-entropy configurations in the atomic distributions of a multi-principal-element alloy. Nat. Commun. 6, 5964 (2015).
Yin, J. & Pei, Z. (2021) Ti38V15Nb23Hf24. figshare https://doi.org/10.6084/m9.figshare.14417225.v4 (2021).
Yin, J. DeepThermo: neural network based order parameter for phase transitions. Code Ocean https://doi.org/10.24433/CO.8787331.v1 (2021).
Acknowledgements
This research was sponsored by and used resources of the Oak Ridge Leadership Computing Facility (OLCF), which is a US Department of Energy (DOE) Office of Science User Facility at the Oak Ridge National Laboratory supported by the US DOE under contract no. DE-AC05-00OR22725. This work was also performed in support of the US DOE’s Fossil Energy Crosscutting Technology Research Program, and in part by an appointment to the US DOE Postgraduate Research Program at the National Energy Technology Laboratory (NETL) administered by the Oak Ridge Institute for Science and Education. Neither the United States Government nor any agency thereof, nor any of its employees, nor the support contractor, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
Author information
Authors and Affiliations
Contributions
The initial project idea was formulated by J.Y. J.Y. and Z.P. designed the experiments. J.Y. performed the data generation, VAE training and order parameter inference. Z.P. performed all DFT calculations, interpreted the total order parameters, proposed the alloy design concept and realized the initial version of the method. J.Y. and Z.P. wrote the manuscript. J.Y., Z.P. and M.C.G. analyzed the results and finalized the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Computational Science thanks Evert van Nieuwenburg, Sebastian Wetzel and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Handling editor: Jie Pan, in collaboration with the Nature Computational Science team.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary Information
Supplementary Figs. 1–7, Table 1 and Discussion.
Source data
Source Data Fig. 2
VAE embeddings and order parameter, and specific heat data for MoNbTaW.
Source Data Fig. 3
Second-order moment of the VAE order parameter and VAE embeddings for AlxCoCrFeNi, where x = 1, 1.6, and 2.
Source Data Fig. 4
Configurations of eight subsystems of TiVNbHf and the VAE order parameter histogram and element distribution for TiVNbHf alloy design, and specific heat for Ti25V25Nb25Hf25, Ti38V15Nb23Hf24, Ti29V15Nb28Hf28 and Ti28V12Nb30Hf30.
Source Data Fig. 5
SRO, LRO and VAE order parameter, and total order parameter for MoNbTaW.
Source Data Fig. 6
Training and inference time for up to 96 GPUs on the Summit supercomputer.
Rights and permissions
About this article
Cite this article
Yin, J., Pei, Z. & Gao, M.C. Neural network-based order parameter for phase transitions and its applications in high-entropy alloys. Nat Comput Sci 1, 686–693 (2021). https://doi.org/10.1038/s43588-021-00139-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s43588-021-00139-3
This article is cited by
-
Toward the design of ultrahigh-entropy alloys via mining six million texts
Nature Communications (2023)
-
Order Parameter Engineering for Random Systems
High Entropy Alloys & Materials (2023)
-
To V or not to V
Nature Computational Science (2021)
-
For the greatest benefit to humankind
Nature Computational Science (2021)