Journal Home > Volume 12 , Issue 7
Journal of Advanced Ceramics 2023, 12(7): 1389-1405 https://doi.org/10.26599/JAC.2023.9220762
Research Article | Open Access | Issue | Published: 21 June 2023

Accelerated design of lead-free high-performance piezoelectric ceramics with high accuracy via machine learning

Show Author's Information Wei Gua,†,  Bin Yanga,†,  Dengfeng Lia,  Xunzhong Shanga,  Zhiyong Zhoub( ),  Jinming Guoa( )
Key Laboratory of Green Preparation and Application for Functional Materials, Ministry of Education, School of Materials Science and Engineering, Hubei University, Wuhan 430062, China
Key Laboratory of Inorganic Functional Materials and Devices, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 201899, China

† Wei Gu and Bin Yang contributed equally to this work.

Keywords:
perovskite, piezoelectric properties, machine learning (ML), piezoelectric ceramics, regression learning
Cite this article:
Gu W, Yang B, Li D, et al. Accelerated design of lead-free high-performance piezoelectric ceramics with high accuracy via machine learning. Journal of Advanced Ceramics, 2023, 12(7): 1389-1405. https://doi.org/10.26599/JAC.2023.9220762
Download citation
EndNote(RIS)
BibTeX

1965

Views

346

Downloads

Citations

4

Crossref

3

WoS

3

Scopus

0

CSCD

Abstract Full text Electronic supplementary material About this article

The piezoelectric performance serves as the basis for the applications of piezoelectric ceramics. The ability to rapidly and accurately predict the piezoelectric coefficient (d33) is of much practical importance for exploring high-performance piezoelectric ceramics. In this work, a data-driven approach combining feature engineering, statistical learning, machine learning (ML), experimental design, and synthesis is trialed to investigate its accuracy in predicting d33 of potassium–sodium–niobate ((K,Na)NbO3, KNN)-based ceramics. The atomic radius (AR), valence electron distance (DV) (Schubert), Martynov–Batsanov electronegativity (EN-MB), and absolute electronegativity (EN) are summarized as the four most representative features in describing d33 out of all 27 possible features for the piezoelectric ceramics. These four features contribute greatly to regression learning for predicting d33 and classification learning for distinguishing polymorphic phase boundary (PPB). The ML method developed in this work exhibits a high accuracy in predicting d33 of the piezoelectric ceramics. An example of KNN combined with 6 mol% LiNbO3 demonstrates d33 of 184 pC/N, which is highly consistent with the predicted result. This work proposes a novel feature-oriented guideline for accelerating the design of piezoelectric ceramic systems with large d33, which is expected to be widely used in other functional materials.


menu
Abstract
Full text
Outline
Electronic supplementary material
About this article

Accelerated design of lead-free high-performance piezoelectric ceramics with high accuracy via machine learning

Show Author's information Wei Gua,†, Bin Yanga,†, Dengfeng Lia, Xunzhong Shanga, Zhiyong Zhoub( ), Jinming Guoa( )
Key Laboratory of Green Preparation and Application for Functional Materials, Ministry of Education, School of Materials Science and Engineering, Hubei University, Wuhan 430062, China
Key Laboratory of Inorganic Functional Materials and Devices, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 201899, China

† Wei Gu and Bin Yang contributed equally to this work.

Abstract

The piezoelectric performance serves as the basis for the applications of piezoelectric ceramics. The ability to rapidly and accurately predict the piezoelectric coefficient (d33) is of much practical importance for exploring high-performance piezoelectric ceramics. In this work, a data-driven approach combining feature engineering, statistical learning, machine learning (ML), experimental design, and synthesis is trialed to investigate its accuracy in predicting d33 of potassium–sodium–niobate ((K,Na)NbO3, KNN)-based ceramics. The atomic radius (AR), valence electron distance (DV) (Schubert), Martynov–Batsanov electronegativity (EN-MB), and absolute electronegativity (EN) are summarized as the four most representative features in describing d33 out of all 27 possible features for the piezoelectric ceramics. These four features contribute greatly to regression learning for predicting d33 and classification learning for distinguishing polymorphic phase boundary (PPB). The ML method developed in this work exhibits a high accuracy in predicting d33 of the piezoelectric ceramics. An example of KNN combined with 6 mol% LiNbO3 demonstrates d33 of 184 pC/N, which is highly consistent with the predicted result. This work proposes a novel feature-oriented guideline for accelerating the design of piezoelectric ceramic systems with large d33, which is expected to be widely used in other functional materials.

Keywords: perovskite, piezoelectric properties, machine learning (ML), piezoelectric ceramics, regression learning

References(58)

[1]
Jaffe H, Berlincourt DA. Piezoelectric transducer materials. Proc IEEE 1965, 53: 1372–1386.
DOI Google Scholar
[2]
Berlincourt DA, Cmolik C, Jaffe H. Piezoelectric properties of polycrystalline lead titanate zirconate compositions. Proc IRE 1960, 48: 220–229.
DOI Google Scholar
[3]
Shrout TR, Zhang SJ. Lead-free piezoelectric ceramics: Alternatives for PZT? J Electroceram 2007, 19: 113–126.
DOI Google Scholar
[4]
Egerton L, Dillon DM. Piezoelectric and dielectric properties of ceramics in the system potassium–sodium niobate. J Am Ceram Soc 1959, 42: 438–442.
DOI Google Scholar
[5]
Saito Y, Takao H, Tani T, et al. Lead-free piezoceramics. Nature 2004, 432: 84–87.
DOI Google Scholar
[6]
Wu JG. Perovskite lead-free piezoelectric ceramics. J Appl Phys 2020, 127: 190901.
DOI Google Scholar
[7]
Wang R, Xie RJ, Hanada K, et al. Enhanced piezoelectricity around the tetragonal/orthorhombic morphotropic phase boundary in (Na,K)NbO3–ATiO3 solid solutions. J Electroceram 2008, 21: 263–266.
DOI Google Scholar
[8]
Park HY, Cho KH, Paik DS, et al. Microstructure and piezoelectric properties of lead-free (1−x)(Na0.5K0.5)NbO3–xCaTiO3 ceramics. J Appl Phys 2007, 102: 124101.
DOI Google Scholar
[9]
Du HL, Zhou WC, Luo F, et al. Design and electrical properties’ investigation of (K0.5Na0.5)NbO3–BiMeO3 lead-free piezoelectric ceramics. J Appl Phys 2008, 104: 034104.
DOI Google Scholar
[10]
Zuo RZ, Lv DY, Fu J, et al. Phase transition and electrical properties of lead free (Na0.5K0.5)NbO3–BiAlO3 ceramics. J Alloys Compd 2009, 476: 836–839.
DOI Google Scholar
[11]
Wu WJ, Xiao DQ, Wu JG, et al. Polymorphic phase transition-induced electrical behavior of BiCoO3-modified (K0.48Na0.52)NbO3 lead-free piezoelectric ceramics. J Alloys Compd 2011, 509: L284–L288.
DOI Google Scholar
[12]
Wu JG, Xiao DQ, Zhu JG. Potassium–sodium niobate lead-free piezoelectric materials: Past, present, and future of phase boundaries. Chem Rev 2015, 115: 2559–2595.
DOI Google Scholar
[13]
Xue DZ, Balachandran PV, Hogden J, et al. Accelerated search for materials with targeted properties by adaptive design. Nat Commun 2016, 7: 11241.
DOI Google Scholar
[14]
Raccuglia P, Elbert KC, Adler PDF, et al. Machine-learning-assisted materials discovery using failed experiments. Nature 2016, 533: 73–76.
DOI Google Scholar
[15]
Xue DZ, Balachandran PV, Yuan RH, et al. Accelerated search for BaTiO3-based piezoelectrics with vertical morphotropic phase boundary using Bayesian learning. PNAS 2016, 113: 13301–13306.
DOI Google Scholar
[16]
Ma XY, Lyu HY, Dong XJ, et al. Voting data-driven regression learning for accelerating discovery of advanced functional materials and applications to two-dimensional ferroelectric materials. J Phys Chem Lett 2021, 12: 973–981.
DOI Google Scholar
[17]
Ryu B, Wang LQ, Pu HH, et al. Understanding, discovery, and synthesis of 2D materials enabled by machine learning. Chem Soc Rev 2022, 51: 1899–1925.
DOI Google Scholar
[18]
Balachandran PV, Emery AA, Gubernatis JE, et al. Predictions of new ABO3 perovskite compounds by combining machine learning and density functional theory. Phys Rev Mater 2018, 2: 043802.
DOI Google Scholar
[19]
Jia X, Deng YS, Bao X, et al. Unsupervised machine learning for discovery of promising half-Heusler thermoelectric materials. NPJ Comput Mater 2022, 8: 34.
DOI Google Scholar
[20]
Durodola JF. Machine learning for design, phase transformation and mechanical properties of alloys. Prog Mater Sci 2022, 123: 100797.
DOI Google Scholar
[21]
Xue DZ, Xue DQ, Yuan RH, et al. An informatics approach to transformation temperatures of NiTi-based shape memory alloys. Acta Mater 2017, 125: 532–541.
DOI Google Scholar
[22]
Balachandran PV, Kowalski B, Sehirlioglu A, et al. Experimental search for high-temperature ferroelectric perovskites guided by two-step machine learning. Nat Commun 2018, 9: 1668.
DOI Google Scholar
[23]
Lu SH, Zhou QH, Ouyang YX, et al. Accelerated discovery of stable lead-free hybrid organic–inorganic perovskites via machine learning. Nat Commun 2018, 9: 3405.
DOI Google Scholar
[24]
Gao JH, Liu YB, Wang Y, et al. Designing high dielectric permittivity material in barium titanate. J Phys Chem C 2017, 121: 13106–13113.
DOI Google Scholar
[25]
Song YQ, Lindsay J, Zhao Y, et al. Machine Learning based prediction of noncentrosymmetric crystal materials. Comput Mater Sci 2020, 183: 109792.
DOI Google Scholar
[26]
Sun XY, Chen J, Yu RB, et al. BiScO3 doped (Na0.5K0.5)NbO3 lead-free piezoelectric ceramics. J Am Ceram Soc 2009, 92: 130–132.
DOI Google Scholar
[27]
Zuo RZ, Ye C, Fang XS. Dielectric and piezoelectric properties of lead free Na0.5K0.5NbO3–BiScO3 ceramics. Jpn J Appl Phys 2007, 46: 6733–6736.
DOI Google Scholar
[28]
Guo YP, Kakimoto KI, Ohsato H. (Na0.5K0.5)NbO3–LiTaO3 lead-free piezoelectric ceramics. Mater Lett 2005, 59: 241–244.
DOI Google Scholar
[29]
Chen L, Chen Q, Huang W, et al. Piezoelectric properties of (1−x)(Na0.5K0.5)NbO3–xLiTaO3 lead-free piezoelectric ceramics. In: Proceedings of the 2007 Sixteenth IEEE International Symposium on the Applications of Ferroelectrics, Nara, Japan, 2007: 662–664.
[30]
Yuan RH, Liu Z, Balachandran PV, et al. Accelerated discovery of large electrostrains in BaTiO3-based piezoelectrics using active learning. Adv Mater 2018, 30: 1702884.
DOI Google Scholar
[31]
Li CH, Soh KCK, Wu P. Formability of ABO3 perovskites. J Alloys Compd 2004, 372: 40–48.
DOI Google Scholar
[32]
Pilania G, Balachandran PV, Kim C, et al. Finding new perovskite halides via machine learning. Front Mater 2016, 3: 19.
DOI Google Scholar
[33]
Rubio-Marcos F, Fernandez JF, Ochoa DA, et al. Understanding the piezoelectric properties in potassium–sodium niobate-based lead-free piezoceramics: Interrelationship between intrinsic and extrinsic factors. J Eur Ceram Soc 2017, 37: 3501–3509.
DOI Google Scholar
[34]
Damjanovic D. Contributions to the piezoelectric effect in ferroelectric single crystals and ceramics. J Am Ceram Soc 2005, 88: 2663–2676.
DOI Google Scholar
[35]
He JJ, Li JJ, Liu CB, et al. Machine learning identified materials descriptors for ferroelectricity. Acta Mater 2021, 209: 116815.
DOI Google Scholar
[36]
Song YQ, Siriwardane EMD, Zhao Y, et al. Computational discovery of new 2D materials using deep learning generative models. ACS Appl Mater Interfaces 2021, 13: 53303–53313.
DOI Google Scholar
[37]
Hu J, Song YQ. Piezoelectric modulus prediction using machine learning and graph neural networks. Chem Phys Lett 2022, 791: 139359.
DOI Google Scholar
[38]
Pilania G, Balachandran PV, Gubernatis JE, et al. Classification of ABO3 perovskite solids: A machine learning study. Acta Crystallogr B 2015, 71: 507–513.
DOI Google Scholar
[39]
Zhang H, Li N, Li KY, et al. Structural stability and formability of ABO3-type perovskite compounds. Acta Crystallogr B 2007, 63: 812–818.
DOI Google Scholar
[40]
James G, Witten D, Hastie T, et al. An Introduction to Statistical Learning: With Applications in R. New York: Springer New York, 2013.
DOI
[41]
Zheng T, Wu JG, Xiao DQ, et al. Recent development in lead-free perovskite piezoelectric bulk materials. Prog Mater Sci 2018, 98: 552–624.
DOI Google Scholar
[42]
Duros V, Grizou J, Xuan WM, et al. Human versus robots in the discovery and crystallization of gigantic polyoxometalates. Angew Chem Int Ed 2017, 56: 10815–10820.
DOI Google Scholar
[43]
Balachandran PV, Young J, Lookman T, et al. Learning from data to design functional materials without inversion symmetry. Nat Commun 2017, 8: 14282.
DOI Google Scholar
[44]
Lookman T, Balachandran PV, Xue DZ, et al. Statistical inference and adaptive design for materials discovery. Curr Opin Solid St M 2017, 21: 121–128.
DOI Google Scholar
[45]
Forrester AIJ, Keane AJ. Recent advances in surrogate-based optimization. Prog Aerosp Sci 2009, 45: 50–79.
DOI Google Scholar
[46]
Sharma V, Wang CC, Lorenzini RG, et al. Rational design of all organic polymer dielectrics. Nat Commun 2014, 5: 4845.
DOI Google Scholar
[47]
Balachandran PV, Xue DZ, Theiler J, et al. Adaptive strategies for materials design using uncertainties. Sci Rep 2016, 6: 19660.
DOI Google Scholar
[48]
Jones DR, Schonlau M, Welch WJ. Efficient global optimization of expensive black-box functions. J Global Optim 1998, 13: 455–492.
DOI Google Scholar
[49]
Min K, Cho E. Accelerated discovery of potential ferroelectric perovskite via active learning. J Mater Chem C 2020, 8: 7866–7872.
DOI Google Scholar
[50]
Yuan RH, Xue DQ, Xue DZ, et al. The search for BaTiO3-based piezoelectrics with large piezoelectric coefficient using machine learning. IEEE T Ultrason Ferr 2019, 66: 394–401.
DOI Google Scholar
[51]
Rabe KM, Phillips JC, Villars P, et al. Global multinary structural chemistry of stable quasicrystals, high-TC ferroelectrics, and high-TC superconductors. Phys Rev B 1992, 45: 7650–7676.
DOI Google Scholar
[52]
Pauling L. The nature of the chemical bond. IV. The energy of single bonds and the relative electronegativity of atoms. J Am Chem Soc 1932, 54: 3570–3582.
DOI Google Scholar
[53]
Mulliken RS. A new electroaffinity scale; together with data on valence states and on valence ionization potentials and electron affinities. J Chem Phys 1934, 2: 782–793.
DOI Google Scholar
[54]
Pearson RG. Absolute electronegativity and hardness: Application to inorganic chemistry. Inorg Chem 1988, 27: 734–740.
DOI Google Scholar
[55]
Oh SHV, Hwang W, Kim K, et al. Using feature-assisted machine learning algorithms to boost polarity in lead-free multicomponent niobate alloys for high-performance ferroelectrics. Adv Sci 2022, 9: 2104569.
DOI Google Scholar
[56]
Sukkha U, Vittayakorn N, Guo RY, et al. Phase transition behavior of Ba(Mg1/3Nb2/3)O3 modified PbZrO3 solid solution. J Mater Chem C 2014, 2: 2929–2938.
DOI Google Scholar
[57]
Li S, Nie HC, Wang GS, et al. Novel AgNbO3-based lead-free ceramics featuring excellent pyroelectric properties for infrared detecting and energy-harvesting applications via antiferroelectric/ferroelectric phase-boundary design. J Mater Chem C 2019, 7: 4403–4414.
DOI Google Scholar
[58]
Wang J, Xiao B, Liu Y. Machine learning assisted high-throughput experiments accelerates the composition design of hard high-entropy alloy CoxCryTizMouWv. Mater China 2020, 39: 269–277. (in Chinese)
Google Scholar
File
JAC0762_ESM.pdf (365.5 KB)
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 22 December 2022
Revised: 20 March 2023
Accepted: 30 April 2023
Published: 21 June 2023
Issue date: July 2023

Copyright

© The Author(s) 2023.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant No. 52001117). It is also supported by the Opening Project of Key Laboratory of Inorganic Functional Materials and Devices, Chinese Academy of Sciences (Grant No. KLIFMD202305).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.

The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Return