基于机器学习筛选低热导率稀土钽/铌酸盐材料

廖梦婷; 黎峻利; 皮智鹏; 张帆

doi:10.19976/j.cnki.43-1448/TF.2024109

粉末冶金材料科学与工程 >

2025 , Vol. 30 >Issue 3: 179 - 192

DOI: https://doi.org/10.19976/j.cnki.43-1448/TF.2024109

理论研究

基于机器学习筛选低热导率稀土钽/铌酸盐材料

廖梦婷 ,
黎峻利 ,
皮智鹏 ,
张帆

展开

湘潭大学材料科学与工程学院,湘潭 411105

皮智鹏,副教授,博士。电话：17700238582;E-mail: pizhipeng@xtu.edu.cn;张帆,教授,博士。电话：15874107569;E-mail: zhangfan15@xtu.edu.cn

收稿日期: 2024-12-13

修回日期: 2025-04-16

网络出版日期: 2025-07-28

基金资助

国家自然科学基金资助项目(52171015)

收起

Screening of low thermal conductivity rare earth tantalate/niobate materials based on machine learning

LIAO Mengting ,
LI Junli ,
PI Zhipeng ,
ZHANG Fan

Expand

School of Materials Science and Engineering, Xiangtan University, Xiangtan 411105, China

Received date: 2024-12-13

Revised date: 2025-04-16

Online published: 2025-07-28

Fold

摘要

稀土(rare earth, RE)钽/铌酸盐(RE(Ta/Nb)O₄)作为新一代的热障涂层备选材料,因其优异的性能而备受关注。本文基于机器学习法分析RE(Ta/Nb)O₄ (RE=Sc、Y、Yb、Dy、Gd、Sm、Ho、La、Lu、Tm、Er、Ce、Eu)材料的热导率,并结合贪心算法寻找具有更低热导率的热障涂层材料。利用元素组成、原子特性与参数、晶体结构信息和热力学数据作为特征参数,采用梯度提升决策树作为预测模型对材料进行筛选,并通过实验进行验证。结果表明：(Y_2/7Yb_5/7)(Ta_1/2Nb_1/2)O₄的实验值与预测模型符合较好,梯度提升决策树作为后续热导率预测的机器学习模型,筛选出了多种低热导率的RE(Ta_1/2Nb_1/2)O₄热障涂层材料。多稀土共掺杂和高熵化RE(Ta/Nb)O₄的热学性能比某些单组分RE(Ta/Nb)O₄更突出,是具有潜力的新型热障涂层材料。

关键词： 机器学习; 热导率; 稀土钽/铌酸盐; 材料设计; 梯度提升决策树; 生成对抗网络

本文引用格式

廖梦婷 , 黎峻利 , 皮智鹏 , 张帆 . 基于机器学习筛选低热导率稀土钽/铌酸盐材料[J]. 粉末冶金材料科学与工程, 2025 , 30(3) : 179 -192 . DOI: 10.19976/j.cnki.43-1448/TF.2024109

Abstract

Rare earth tantalate/niobate (RE(Ta/Nb)O₄) materials have attracted significant attention as promising candidates for next-generation thermal barrier coatings due to their excellent properties. The machine learning was used to analyze the thermal conductivity of RE(Ta/Nb)O₄ (RE=Sc, Y, Yb, Dy, Gd, Sm, Ho, La, Lu, Tm, Er, Ce, Eu) materials, and combined a greedy algorithm to identify materials with lower thermal conductivity for thermal barrier coatings in this study. Using feature parameters such as element composition, atomic properties and parameters, crystal structure information, and thermodynamic data, the gradient boosting decision tree model was employed for material screening and validated through experiments. The results show that the experimental values of (Y_2/7Yb_5/7)(Ta_1/2Nb_1/2)O₄ align well with the predicted model, and gradient boosting decision tree proves to be an effective machine learning model for future thermal conductivity predictions. Several low thermal conductivity RE(Ta_1/2Nb_1/2)O₄ thermal barrier coating materials are successfully identified. Co-doping with multiple rare earth elements and high-entropy RE(Ta/Nb)O₄ exhibit superior thermal performance compared to certain single- component RE(Ta/Nb)O₄ materials, making them promising new materials for thermal barrier coatings.

Key words： machine learning; thermal conductivity; rare earth tantalate/niobate; material design; gradient boosting decision tree; generative adversarial network

参考文献

[1] 赵云松, 张迈, 戴建伟, 等. 航空发动机涡轮叶片热障涂层研究进展[J]. 材料导报, 2023, 37(6): 77-83.
ZHAO Yunsong, ZHANG Mai, DAI Jianwei, et al.Research progress of thermal barrier coatings for aeroengine turbine blades[J]. Materials Review, 2023, 37(6): 77-83.
[2] 赵娟利, 杨岚, 张成冠, 等. 热障涂层材料研究进展[J]. 现代技术陶瓷, 2020, 41(3): 148-170.
ZHAO Juanli, YANG Lan, ZHANG Chengguan, et al.Recent progress in thermal barrier coatings[J]. Advanced Ceramics, 2020, 41(3): 148-170.
[3] RAHIMI S, SHARIFIANJAZI F, ESMAEILKHANIAN A, et al.Effect of SiO₂ content on Y-TZP/Al₂O₃ ceramic-nanocomposite properties as potential dental applications[J]. Ceramics International, 2020, 46(8): 10910-10916.
[4] BROIDO D A, MALORNY M, BIRNER G, et al.Intrinsic lattice thermal conductivity of semiconductors from first principles[J]. Applied Physics Letters, 2007, 91(23): 457-462.
[5] FAN Z, PEREIRA L F C, WANG H Q, et al. Force and heat current formulas for many-body potentials in molecular dynamics simulations with applications to thermal conductivity calculations[J]. Physical Review B, 2015, 92(9): 094301.
[6] LUO Y, LI M, YUAN H, et al.Predicting lattice thermal conductivity via machine learning: a mini review[J]. NPJ Computational Materials, 2023, 9(1): 4.
[7] JAAFREH R, KANG Y S, HAMAD K.Lattice thermal conductivity: an accelerated discovery guided by machine learning[J]. ACS Applied Materials & Interfaces, 2021, 13(48): 57204-57213.
[8] CHAKRABORTY P, LIU Y D, MA T F, et al.Quenching thermal transport in aperiodic superlattices: a molecular dynamics & machine learning study[J]. ACS Applied Materials & Interfaces, 2020, 12(7): 8795-8804.
[9] RACCUGLIA P, ELBERT K C, ADLER P D F, et al. Machine-learning-assisted materials discovery using failed experiments[J]. Nature, 2016, 533(7601): 73-76.
[10] QIN J C, LIU Z F, MA M S, et al.Machine learning-assisted materials design and discovery of low-melting-point inorganic oxides for low-temperature cofired ceramic applications[J]. ACS Sustainable Chemistry & Engineering, 2022, 10(4): 1554-1564.
[11] CHEN L, FENG J.Thermal and mechanical properties optimization of ABO₄ type EuNbO₄ by the B-site substitution of Ta[J]. Engineering, 2020, 6(2): 178-185.
[12] CHEN L, LI B H, FENG J.Rare-earth tantalates for next-generation thermal barrier coatings[J]. Progress in Materials Science, 2024, 144: 101265.
[13] WU P, CHONG X Y, WU F S, et al.Investigation of the thermophysical properties of (Y₁-_xYb_x)TaO₄ ceramics[J]. Journal of the European Ceramic Society, 2020, 40(8): 3111-3121.
[14] Royal Society of Chemistry. Periodic table[Z/OL]. (2024-09) [2024-10-29]. https://www.rsc.org/periodic-table.
[15] Chemical Aid. Periodic table[Z/OL]. (2024-06-09)[2024- 10-29]. https://periodictable.chemicalaid.com.
[16] The Materials Project. Crystal structure[DB/OL]. (2024-10- 25) [2024-10-29]. https://materialsproject.org.
[17] SCHLICHTING K W, PADTURE N P, KLEMENS P G.Thermal conductivity of dense and porous yttria-stabilized zirconia[J]. Materal Science, 2001, 36(12): 3003-3010.
[18] CHEN L, WANG J K, LI B H, et al.Simultaneous manipulations of thermal expansion and conductivity in symbiotic ScTaO₄/SmTaO₄ composites via multiscale effects[J]. Journal of Advanced Ceramics, 2023, 12(8): 1625-1640.
[19] WANG J, CHONG X Y, ZHOU R, et al.Microstructure and thermal properties of RETaO₄ (RE=Nd, Eu, Gd, Dy, Er, Yb, Lu) as promising thermal barrier coating materials[J]. Scripta Materialia, 2017, 126: 24-28.
[20] WANG J, ZHENG Q, SHI X L, et al.Microstructural evolution and thermal-physical properties of YTaO₄ coating after high-temperature exposure[J]. Surface and Coatings Technology, 2023, 456: 129222.
[21] WU F S, WU P, ZHOU Y X, et al.The thermo‐mechanical properties and ferroelastic phase transition of RENbO₄ (RE=Y, La, Nd, Sm, Gd, Dy, Yb) ceramics[J]. Journal of the American Ceramic Society, 2020, 103(4): 2727-2740.
[22] CHEN L, WU P, SONG P, et al.Potential thermal barrier coating materials: RE₃NbO₇ (RE=La, Nd, Sm, Eu, Gd, Dy) ceramics[J]. Journal of the American Ceramic Society, 2018, 101(10): 4503-4508.
[23] WANG J, WU F S, ZOU R A, et al.High‐entropy ferroelastic rare‐earth tantalite ceramic: (Y_0.2Ce_0.2Sm_0.2Gd_0.2Dy_0.2)TaO₄[J]. Journal of the American Ceramic Society, 2021, 104(11): 5873-5882.
[24] CHEN L, LI B H, LUO K R, et al.Origins of high fracture toughness and glass‐like thermal conductivity in Zr-Ta-O composites[J]. Journal of the American Ceramic Society, 2022, 105(11): 6508-6516.
[25] ZHU J T, XU J, ZHANG P, et al.Enhanced mechanical and thermal properties of ferroelastic high-entropy rare-earth- niobates[J]. Scripta Materialia, 2021, 200: 113912.
[26] WU P, CHONG X Y, FENG J.Effect of Al³⁺ doping on mechanical and thermal properties of DyTaO₄ as promising thermal barrier coating application[J]. Journal of the American Ceramic Society, 2018, 101(5): 1818-1823.
[27] XUE T, ZHANG Y A, GAO Y, et al.Synthesis and thermophysical properties of (Al_xY₁-_x)TaO₄ ceramics as thermal barrier coating materials[J]. Ceramics International, 2023, 49(16): 27577-27588.
[28] WU P, HU M Y, CHEN L, et al.Investigation on microstructures and thermo-physical properties of ferroelastic (Y₁-_xDy_x)TaO₄ ceramics[J]. Materialia, 2018, 4: 478-486.
[29] ZHANG P, FENG Y J, LI Y, et al.Thermal and mechanical properties of ferroelastic RENbO₄ (RE=Nd, Sm, Gd, Dy, Er, Yb) for thermal barrier coatings[J]. Scripta Materialia, 2020, 180: 51-56.
[30] WANG P P, WANG J, CAI H Y, et al.Effect of multi-component at the A site on the thermophysical properties of high entropy niobates[J]. Journal of the European Ceramic Society, 2024, 44(5): 2954-2964.
[31] WANG J, CHONG X Y, LV L, et al.High-entropy ferroelastic (10RE_0.1)TaO₄ ceramics with oxygen vacancies and improved thermophysical properties[J]. Journal of Materials Science & Technology, 2023, 157: 98-106.
[32] DINSDALE A T.SGTE data for pure elements[J]. Calphad, 1991, 15(4): 317-425.
[33] GAN M D, LAI L P, WANG J K, et al.Suppressing the oxygen-ionic conductivity and promoting the phase stability of the high-entropy rare earth niobates via Ta substitution[J]. Journal of Materials Science & Technology, 2025, 209: 79-94.
[34] CHEN L, HU M Y, FENG J.Defect-dominated phonon scattering processes and thermal transports of ferroelastic (Sm₁-_xYb_x)TaO₄ solid solutions[J]. Materials Today Physics, 2023, 35: 101118.
[35] LUO Y, CHEN L, WU P, et al.Synthesis and thermophysics properties of ferroelastic SmNb₁-_xTa_xO₄ ceramics[J]. Ceramics International, 2018, 44(12): 13999-14006.
[36] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
ZHOU Zhihua.Machine Learning[M]. Beijing: Tsinghua University Press, 2016.
[37] YANG J S, ZHAO C Y, YU H T, et al.Use GBDT to predict the stock market[J]. Procedia Computer Science, 2020, 174: 161-171.
[38] NICHOLLS J R, LAWSON K J, JOHNSTONE A, et al. Methods to reduce the thermal conductivity of EB-PVD TBCs[J]. Surface and Coatings Technology, 2002, 151/152: 383-391.
[39] 黄昆. 固体物理学[M]. 北京: 高等教育出版社, 1988.
HUANG Kun.Solid State Physics[M]. Beijing: Higher Education Press, 1988.
[40] LECUN Y, BENGIO Y, HINTON. Deep learning[J]. Nature, 2015, 521: 436-444.
[41] ZHANG X, LI H.A robust kernel principal component analysis method and its applications[J]. Neurocomputing, 2020, 395: 191-201.
[42] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al.Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144.
[43] TSAI M H, YEH J W.High-entropy alloys: a critical review[J]. Materials Research Letters, 2014, 2(3): 107-123.
[44] PARKER W J, JENKINS R J, BUTLER C P, et al.Flash method of determining thermal diffusivity, heat capacity, and thermal conductivity[J]. Journal of Applied Physics, 1961, 32(9): 1679-1684.
[45] LEITNER J, VOŇKA P, SEDMIDUBSKÝ D, et al. Application of Neumann-Kopp rule for the estimation of heat capacity of mixed oxides[J]. Thermochimica Acta, 2010, 497(1/2): 7-13.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献