| 161 | 0 | 86 |
| 下载次数 | 被引频次 | 阅读次数 |
Sm-Co基稀土永磁合金在高温用永磁材料领域具有不可替代的重要地位。然而，目前对Sm-Co基永磁合金的成分设计仍以实验试错法为主，使得新型合金的研究开发效率较低。本文系统综述了数据驱动研究范式在Sm-Co基稀土永磁合金高性能化设计研究中的关键进展，阐述了如热力学计算、第一性原理计算等传统计算材料学方法在Sm-Co基合金物相分析和磁性能预测方面的进展，重点介绍和讨论了基于机器学习方法的Sm-Co基永磁合金成分高通量设计与多目标性能协同优化的新进展。最后展望了多尺度计算模拟、物理可解释机器学习、逆向设计闭环系统等融合人工智能的研究新范式推动本领域数字化高效研究的发展趋势。
Abstract：Samarium-Cobalt(Sm-Co) based rare-earth permanent magnets are critical materials for high-technology applications, particularly in aerospace and defense systems, due to their unmatched high-temperature performance, including an exceptional Curie temperature and superior magnetic stability under extreme operating conditions. However, optimizing their comprehensive magnetic properties remains a challenge, hindering by the inefficiency of traditional trial-and-error methodologies. The emergence of a new research paradigm, integrating computational modeling and data-driven science, is now powerfully addressing the trial-and-error limitations. This review comprehensively examines key advancements in this field. Traditional computational materials science methods, including thermodynamic calculations via the CALPHAD approach and nanoscale thermodynamic modeling, have enabled accurate predictions of phase equilibria and metastable phase stability in multicomponent alloys. First-principles calculations have provided atomic-level insights into the electronic structure origins of magnetocrystalline anisotropy and the mechanistic role of transition metal doping in tuning magnetic properties and phase stability. This review highlights the transformative impact of data-driven and AI strategies. By constructing high-quality databases for Sm-Co-based alloys and employing advanced machine learning(ML) algorithms, researchers have achieved high-throughput prediction and multi-objective optimization of key properties such as saturation magnetization, coercivity, Curie temperature, and phase constitution. The application of multi-objective optimization strategies successfully breaks the inherent trade-off between saturation magnetization and coercivity, leading to the discovery of novel compositions with superior comprehensive magnetic properties. The development of user-friendly intelligent design systems has significantly lowered the barrier for employing these AI methods and tools in practical magnet design. The coupling of multi-scale simulations(e.g., DFT, micromagnetic modeling) with active learning frameworks also presents a promising pathway for building physically informed and data-efficient models, marking a significant step towards inverse design. These data-driven models have successfully guided the experimental synthesis of novel nanocrystalline Sm-Co-based alloys with superior magnetic performance, validating this modern AI approach. Summary and prospects Despite the significant progress, the full potential of data-driven and AI methodologies in Sm-Co-based permanent magnet research is yet to be realized. To overcome existing challenges and harness future opportunities, the following prospects are proposed. 1) Enhancing Physical Interpretability of ML Models. Future research should go beyond treating machine learning as a "black-box" predictive tool and place greater emphasis on the physical interpretability of models. On one hand, efforts should be made to actively develop feature descriptors that integrate physical priors and domain knowledge such as magnetism theory and crystal field theory. On the other hand, while continuing to deepen the application of interpretable AI techniques like SISSO and SHAP to quantify the contributions of different features to performance, in-depth research should be conducted to explore the application of new algorithms and optimization strategies such as neural networks. Exploring hybrid models that embed constraints and relationships derived from first-principles calculations and micromagnetic simulations will be crucial for predictions in physical reality. 2) Developing Frameworks for Multi-Source Data and Transfer Learning. The data generated from research on rare-earth permanent magnetic alloys are typical multi-source heterogeneous data, including numerically computed and experimentally measured values, microstructural images, where the microstructure significantly influencing magnetic properties. Future work should leverage deep learning to create unified, machine-readable representations of this diverse data. On this basis, cross-scale transfer learning methods and active learning approaches should be developed to prevent traditional machine learning methods from falling into the traps of overfitting and local optima in small-sample scenarios. 3) Establishing Closed-Loop Inverse Design Systems. The ultimate goal of data-driven approaches is to achieve target-performance-oriented intelligent research and development. Future efforts should focus on constructing an automated intelligent pipeline for inverse design and automated validation. Driven by target performance, this system utilizes conditional generative models to automatically generate candidate compositions and process schemes that meet the requirements. These candidate solutions are then fed into a multi-scale computational chain for simulation validation and optimization. The computational results are fed back as new data into the database and machine learning loops, enabling continuous iteration and optimization of design strategies. 4) Pursuing Data-Assisted Sustainable and Economic Alternatives. By constructing a database incorporating substitution elements, this approach predicts the impact of partially replacing Sm with high-abundance rare-earth elements such as Ce/La on magnetic properties. The goal is to enhance the utilization of high-abundance rare-earth elements while ensuring high performance, thereby developing new cost-effective composition systems. Furthermore, by comprehensively considering factors such as manufacturing processes, raw material costs, and performance merits, novel low-carbon and environmentally friendly rare-earth permanent magnetic alloy systems can be designed.
[1]COEY J M D. Perspective and prospects for rare earth permanent magnets[J]. Engineering, 2020, 6(2):119–131.
[2]LI Y C, XING Z C, YU N J, et al. High-performance anisotropic SmCo-based magnet with novel microstructure[J]. J Rare Earths, 2024,42(8):1539–1545.
[3]HAO Y H, WANG Y, ZHANG T T, et al. Cu docking-activated Nb incorporation in multivariate CuO-Nb2O5/CeO2 catalysts for selective reduction of NOx with NH3[J]. Appl Catal B Environ, 2024, 340:123254.
[4]LIU Y H, XUE Y, LIU R H, et al. White-light emitting single-phase phosphor La3Si6N11:Dy3+, Tb3+:Color tunable emission, thermal stability and energy transfer process[J]. J Rare Earths, 2019, 37(4):350–355.
[5]JOUBERT J M, PAUL-BONCOUR V, CUEVAS F, et al. LaNi5 related AB5 compounds:Structure, properties and applications[J]. J Alloys Compd, 2021, 862:158163.
[6]NAKAMURA Y, SAKAKI K, KIM H, et al. Reaction paths via a new transient phase in non-equilibrium hydrogen absorption of LaNi2Co3[J].Int J Hydrog Energy, 2020, 45(41):21655–21665.
[7]FENG X D, AL ET AL E. Converting ceria polyhedral nanoparticles into single-crystal nanospheres[J]. ChemInform, 2006, 37(36):200636200.
[8]MIYAKE T, AKAI H. Quantum theory of rare-earth magnets[J]. J Phys Soc Jpn, 2018, 87(4):041009.
[9]JIANG C B, AN S Z. Recent progress in high temperature permanent magnetic materials[J]. Rare Met, 2013, 32(5):431–440.
[10]BRÜCK E. Handbook of Magnetic Materials[M]. Amsterdam:North Holland, 2018:269–372.
[11]YANG J B, HAN J Z, TIAN H D, et al. Structural and magnetic properties of nanocomposite Nd–Fe–B prepared by rapid thermal processing[J]. Engineering, 2020, 6(2):132–140.
[12]HONO K, SEPEHRI-AMIN H. Strategy for high-coercivity Nd–Fe–B magnets[J]. Scr Mater, 2012, 67(6):530–535.
[13]HONO K, SEPEHRI-AMIN H. Prospect for HRE-free high coercivity Nd-Fe-B permanent magnets[J]. Scr Mater, 2018, 151:6–13.
[14]SAITO H, TAKAHASHI M, WAKIYAMA T, et al. Magnetocrystalline anisotropy for Sm2(Co-Mn)17 compound with TbCu7 type disordered structure[J]. J Magn Magn Mater, 1989, 82(2–3):322–326.
[15]LIU J F, ZHANG Y, DIMITROV D, et al. Microstructure and high temperature magnetic properties of Sm(Co, Cu, Fe, Zr)z (z=6.7–9.1)permanent magnets[J]. J Appl Phys, 1999, 85(5):2800–2804.
[16]SHANG M Y, LU H, XU G J, et al. Nanocrystalline Sm Co12main-phase alloys with V-doping:Structure stability and magnetic performance[J]. J Mater Sci Technol, 2025, 210:254–264.
[17]LUO J, LIANG J K, GUO Y Q, et al. Effects of the doping element on crystal structure and magnetic properties of Sm(Co, M)7 compounds(M=Si, Cu, Ti, Zr, and Hf)[J]. Intermetallics, 2005, 13(7):710–716.
[18]LUO J, LIANG J K, GUO Y Q, et al. Crystal structure and magnetic properties of SmCo7–xHfx compounds[J]. Appl Phys Lett, 2004, 85(22):5299–5301.
[19]ZHOU J, AL-OMARI I A, LIU J P, et al. Structure and magnetic properties of SmCo7–xTix with TbCu7-type structure[J]. J Appl Phys,2000, 87(9):5299–5301.
[20]WANG P J, KAO H, ZHANG C W, et al. Electronic structures and magnetic properties in SmCo7–xMx[J]. Chin Phys B, 2009, 18(10):4490–4496.
[21]LUO J, LIANG J K, GUO Y Q, et al. Crystal structure and magnetic properties of SmCo5.85Si0.90 compound[J]. Appl Phys Lett, 2004, 84(16):3094–3096.
[22]GUO Y Q, FENG W C, LI W, et al. Magnetism and phase stability of R(Co, M)7 pseudobinary intermetallics with TbCu7-type structure[J]. J Appl Phys, 2007, 101(2):023919.
[23]AGRAWAL A, CHOUDHARY A. Perspective:Materials informatics and big data:Realization of the“fourth paradigm” of science in materials science[J]. APL Mater, 2016, 4(5):053208.
[24]李国杰.智能化科研(AI4R):第五科研范式[J].中国科学院院刊,2024, 39(1):1–9.LI Guojie. Bull Chin Acad Sci, 2024, 39(1):1–9.
[25]CURTAROLO S, HART G L W, NARDELLI M B, et al. The high-throughput highway to computational materials design[J]. Nat Mater, 2013, 12(3):191–201.
[26]NYSHADHAM C, OSES C, HANSEN J E, et al. A computational high-throughput search for new ternary superalloys[J]. Acta Mater,2017, 122:438–447.
[27]WANG A Y, MURDOCK R J, KAUWE S K, et al. Machine learning for materials scientists:An introductory guide toward best practices[J].Chem Mater, 2020, 32(12):4954–4965.
[28]WANG J Y, XU P C, JI X B, et al. Feature selection in machine learning for perovskite materials design and discovery[J]. Materials, 2023, 16(8):3134.
[29]TAO Q L, XU P C, LI M J, et al. Machine learning for perovskite materials design and discovery[J]. NPJ Comput Mater, 2021, 7:23.
[30]BUSCHOW K H J, VAN DER GOOT A S. Intermetallic compounds in the system samarium-cobalt[J]. J Less Common Met, 1968, 14(3):323–328.
[31]BUSCHOW K H J, DEN BROEDER F J A. The cobalt-rich regions of the samarium-cobalt and gadolinium-cobalt phase diagrams[J]. J Less Common Met, 1973, 33(2):191–201.
[32]DERKAOUI S, ALLIBERT C H. Redetermination of the phase equilibria in the system Sm-Co-Cu for Sm content 0–20 at.%at 850℃[J].J Less Common Met, 1989, 154(2):309–315.
[33]GE W Q, WU C H, CHUANG Y C. Reinvestigation of the Sm-Co binary system[J]. Int J Mater Res, 1993, 84(3):165–169.
[34]CATALDO L, LEFÈVRE A, DUCRET F, et al. Binary system SmCo:Revision of the phase diagram in the Co rich field[J]. J Alloys Compd,1996, 241(1–2):216–223.
[35]YUAN Y, DELSANTE S, YI J, et al. A revision of the Sm-rich region of the Sm–Co system[J]. J Alloys Compd, 2010, 508(2):309–314.
[36]SCHNEIDER G, HENIG E T, LUKAS H L, et al. Phase relations in the samarium-poor Sm–Co–Fe system[J]. J Less Common Met, 1985,110(1–2):159–170.
[37]SU X P, ZHANG W J, LIU G Q, et al. A thermodynamic assessment of the Co–Sm system[J]. J Alloys Compd, 1998, 267(1–2):149–153.
[38]YUAN Y, YI J H, BORZONE G, et al. Thermodynamic modeling of the Co–Sm system[J]. Calphad, 2011, 35(3):416–420.
[39]CHEN D K, LUO L, DAI F L, et al. Phase equilibria in Sm-Co-Ti ternary system[J]. J Phase Equilib Diffus, 2022, 43(3):317–331.
[40]FARTUSHNA I, BAJENOVA I, SHAKIROVA J, et al. Experimental investigation of the phase equilibria in Co–Fe-Sm system with special attention to the effect of Fe substitution in structure and magnetic properties of intermetallic phases[J]. Intermetallics, 2022, 143:107502.
[41]ZHANG L L, ZHENG Q J, JIANG H X, et al. Effect of La addition on microstructure evolution of hypoeutectic Al–6Si alloys[J]. J Mater Sci,2020, 55(17):7546–7554.
[42]DONG G Y, YOU X G, XU Z H, et al. A new model for studing the evaporation behavior of alloy elements in DD98M alloy during electron beam smelting[J]. Vacuum, 2022, 195:110641.
[43]ZHU Z, FANG Y K, WU W, et al. Prediction of Sm-(Co, Cu, Fe, Zr)binary alloys’ thermodynamic parameters by improved miedema model[J]. Mater Res Express, 2024, 11(9):096513.
[44]ZHU Z, FANG Y K, WU W, et al. Computation of the thermodynamic parameters of Sm–Co–Fe ternary alloys[J]. Mater Res Express, 2025,12(4):046101.
[45]LI C, YUAN Y, LI F, et al. Modification and verification of Miedema model for predicating thermodynamic properties of binary precipitates in multi-element alloys[J]. Phys B Condens Matter, 2022, 627:413540.
[46]LI R Q. The application of R function to new asymmetric model and toop model[J]. Calphad, 1989, 13(1):67–70.
[47]LIU P P, DAI F L, LUO L, et al. Experimental study and thermodynamic calculation of the Sm-Co-Fe system[J]. Calphad, 2022,78:102447.
[48]OKAMOTO H. Co-Sm(cobalt-samarium)[J]. J Phase Equilib Diffus,2011, 32(2):165–166.
[49]LESLIE-PELECKY D L, RIEKE R D. Magnetic properties of nanostructured materials[J]. Chem Mater, 1996, 8(8):1770–1783.
[50]LU N D, SONG X Y, ZHANG J X. Crystal structure and magnetic properties of ultrafine nanocrystalline[J]. Nanotechnology, 2010, 21(11):115708.
[51]XU W W, SONG X Y, LU N D, et al. Thermodynamic and experimental study on phase stability in nanocrystalline alloys[J]. Acta Mater, 2010, 58(2):396–407.
[52]XU W W, SONG X Y, LU N D, et al. Nanoscale thermodynamic study on phase transformation in the nanocrystalline Sm2Co17 alloy[J].Nanoscale, 2009, 1(2):238–244.
[53]宋晓艳,徐文武,张哲旭.亚稳相的纳米尺度稳定化:热力学模型与实验研究[J].物理学报, 2012, 61(20):151–158.SONG Xiaoyan, XU Wenwu, ZHANG Zhexu. Acta Phys Sin, 2012,61(20):151–158.
[54]SEYRING M, SONG X Y, ZHANG Z X, et al. Concurrent ordering and phase transformation in SmCo7 nanograins[J]. Nanoscale, 2015, 7(28):12126–12132.
[55]LARSON P, MAZIN I I, PAPACONSTANTOPOULOS D A.Calculation of magnetic anisotropy energy in SmCo5[J]. Phys Rev B,2003, 67(21):214405.
[56]DAS B, CHOUDHARY R, SKOMSKI R, et al. Anisotropy and orbital moment in Sm-Co permanent magnets[J]. Phys Rev B, 2019, 100(2):024419.
[57]MEENA L, AHUJA U, KUMARI MEENA S, et al. Electronic properties and quantitative mapping of magnetic moments of rare-earth hard magnet SmCo5:First principles computations and inelastic scattering[J]. J Magn Magn Mater, 2024, 603:172272.
[58]SOBOLEV A N, GOLOVNIA O A, POPOV A G. Embedded atom potential for Sm–Co compounds obtained by force-matching[J]. J Magn Magn Mater, 2019, 490:165468.
[59]FANG C, YAN Z, ZHANG X J, et al. Enhanced structural stability and magnetism of SmCo3 permanent magnet doped with 3d transition metals:An ab initio study[J]. Rare Met, 2025, 44(2):1256–1266.
[60]LIU X B, ALTOUNIAN Z. Magnetic moments and exchange interaction in Sm(Co, Fe)5 from first-principles[J]. Comput Mater Sci,2011, 50(3):841–846.
[61]GABAY A M, LARSON P, MAZIN I I, et al. Magnetic states and structural transformations in Sm(Co, Cu)5 and Sm(Co, Fe, Cu)5permanent magnets[J]. J Phys D Appl Phys, 2005, 38(9):1337.
[62]MAO F, LU H, LIU D, et al. Structural stability and magnetic properties of SmCo5 compounds doped with transition metal elements[J]. J Alloys Compd, 2019, 810:151888.
[63]YAN Z, FANG C, ZHANG X J, et al. Synergistic 3d-orbital modulation in dual-doped SmCo5-based alloys:Quantifying electronic synergy for enhanced permanent magnetic performance[J]. J Alloys Compd, 2025,1034:181408.
[64]SABIRIANOV R F, KASHYAP A, SKOMSKI R, et al. First principles study of transition-metal substitutions in Sm–Co permanent magnets[J].Appl Phys Lett, 2004, 85(12):2286–2288.
[65]FANG C, YAN Z, ZHANG X J, et al. Stable geometries and magnetic properties in transition-metal-doped Sm5Co19 permanent magnet alloys:Insights from DFT[J]. Sci China Phys Mech Astron, 2025, 68(4):247512.
[66]毛斐,吕皓,唐法威,等. Mn和In添加对SmCo7结构稳定性及磁矩影响的第一性原理计算[J].金属学报, 2021, 57(7):948–958.MAO Fei, LU Hao, TANG Fawei, et al. Acta Metall Sin, 2021, 57(7):948–958.
[67]SÖDERLIND P, LANDA A, LOCHT I L M, et al. Prediction of the new efficient permanent magnet SmCoNiFe3[J]. Phys Rev B, 2017, 96(10):100404.
[68]SABIRYANOV R, JASWAL S S. Magnetic properties of hard/soft composites:SmCo5/Co1–xfex[J]. Phys Rev B, 1998, 58:12071–12074.
[69]宋晓艳,刘东,刘雪梅,等.面向数据驱动材料设计的专用数据库及信息管理系统[J].中国科学:技术科学, 2020, 50(6):786–800.SONG Xiaoyan, LIU Dong, LIU Xuemei, et al. Sci Sin Technol, 2020,50(6):786–800.
[70]SONG X Y, GUO K, LU H, et al. Modelling of phase stability:Integrating computational materials science and materials informatics[J]. J Mater Inform, 2021:7.
[71]LIU D, GUO K, TANG F W, et al. Selecting doping elements by data mining for advanced magnets[J]. Chem Mater, 2019, 31(24):10117–10125.
[72]GUO K, LU H, ZHAO Z, et al. Magnetic performance oriented composition design of Sm-Co based alloys by machine learning and experimental studies[J]. Comput Mater Sci, 2022, 205:111232.
[73]OUYANG R H, CURTAROLO S, AHMETCIK E, et al. SISSO:A compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates[J]. Phys Rev Materials,2018, 2(8):083802.
[74]LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA.ACM, 2017:4768–4777.
[75]LIU P X, LU H, XU G J, et al. Breaking through the trade-off between saturation magnetization and coercivity:A data-driven strategy[J]. Acta Mater, 2025, 289:120945.
[76]CAO Z W, ZHANG T L, TONG X K. Quality evaluation of chicken soup based on entropy weight method and grey correlation degree method[J]. Sci Rep, 2024, 14(1):13038.
[77]NGUYEN D N, PHAM T L, NGUYEN V C, et al. Ensemble learning reveals dissimilarity between rare-earth transition-metal binary alloys with respect to the Curie temperature[J]. J Phys Mater, 2019, 2(3):034009.
[78]LONG T, FORTUNATO N M, ZHANG Y X, et al. An accelerating approach of designing ferromagnetic materials via machine learning modeling of magnetic ground state and Curie temperature[J]. Mater Res Lett, 2021, 9(4):169–174.
[79]NELSON J, SANVITO S. Predicting the Curie temperature of ferromagnets using machine learning[J]. Phys Rev Materials, 2019,3(10):104405.
[80]BELOT J F, TAUFOUR V, SANVITO S, et al. Machine learning predictions of high-Curie-temperature materials[J]. Appl Phys Lett,2023, 123(4):042405.
[81]XU G J, CHENG F, LU H, et al. Predicting the Curie temperature of Sm-Co-based alloys via data-driven strategy[J]. Acta Mater, 2024, 274:120026.
[82]XU G J, LU H, GUO K, et al. Predictions on the phase constitution of SmCo7–xMx alloys by data mining[J]. Nanomaterials, 2022, 12(9):1452.
[83]HAN C Y, LU H, XU G J, et al. Magnetic properties enhancement of multi-element-doped SmCo7 nanocrystalline alloys[J]. Mater Today Phys, 2024, 40:101306.
[84]WANG Z, ZHANG S Y, WANG J, et al. Machine learning guided design of RE-Fe-B(RE=Pr, Nd, La, Ce)with comprehensive high performance[J]. Acta Mater, 2025, 292:121031.
[85]TANG X, LI J, SRINITHI A K, et al. Role of V on the coercivity of SmFe12-based melt-spun ribbons revealed by machine learning and microstructure characterizations[J]. Scr Mater, 2021, 200:113925.
[86]HOSOKAWA H, CALVERT E L, SHIMOJIMA K. Machine learning prediction for magnetic properties of Sm-Fe-N based alloys produced by melt spinning[J]. J Magn Magn Mater, 2021, 526:167651.
[87]LU S H, ZHOU Q H, GUO Y L, et al. On-the-fly interpretable machine learning for rapid discovery of two-dimensional ferromagnets with high Curie temperature[J]. Chem, 2022, 8(3):769–783.
[88]YANG Y W, KÜHN P, FATHIDOOST M, et al. Coercivity influence of nanostructure in SmCo-1:7 magnets:Machine learning of high-throughput micromagnetic data[J]. arXiv, 2024, 2408.03198v2.
基本信息:
DOI：10.14062/j.issn.0454-5648.20250793
中图分类号:TG146.45;TM273
引用信息:
[1]吕皓,许国婧,刘培鑫,等.数据驱动的Sm-Co基稀土永磁合金研究进展[J].硅酸盐学报,2026,54(01):49-69.DOI:10.14062/j.issn.0454-5648.20250793.
基金信息:
国家重点研发计划项目（2021YFB3501502,2021YFB3501504）; 国家自然科学基金（52494942）
2025-10-31
2025
2025-12-10
2025
1
2025-12-31
2025-12-31
2025-12-31