鉴于成本效益、高能量密度及资源丰富性,研发高储能锂离子电池钒酸盐(MVO)负极材料成为当前科研热点。采用数据驱动的方法,融合数据收集与特征筛选、相关性分析、多元模型构建与对比,特征贡献度解析等手段,旨在揭示决定钒酸盐负极材料储能性能的关键因素。根据变量相关度分析可知,M元素离子电负性与首圈、稳定后放电比容量具有较强相关性。通过不同算法的综合比较,Adaboosting+决策树算法模型预测性能最为出色,决定系数分别高达0.89、0.81,shap分析显示离子电负性对该模型预测输出贡献度极高。经相关性研究与机器学习模型预测综合判断,确证M元素离子电负性与储能能力呈较强正相关。基于数据驱动方式可以加速钒酸盐负极材料研发,并为其他领域新型材料研发提供了思路。
Vanadate (MVO),as an excellent negative electrode material for lithium-ion batteries,has been attracted considerable attention due to its quality-price ratio,high energy density and abundant natural resources.In order to reveal the key factors determining the energy storage performance of vanadate anode materials,the data collection and feature screening,the correlation analysis,the multivariate model construction,and the feature contribution resolution were applied based on data-driven approach.According to the analysis of variable correlation,it could be seen that there was a close correlation between the ionic electronegativity of M element and the discharge specific capacity of the first circle after stabilization.Among the various algorithms,the Adaboosting+decision tree algorithm model exhibited the optimal forecasting,with the decision coefficients of 0.89 and 0.81,respectively.The SHAP analysis showed that the ionic electronegativity was main contribution for the predicted output of the model.In short,it was confirmed that the electronegativity of M ions showed a strong positive correlation with the energy storage capacity of vanadate through the comprehensive judgment of correlation study and machine learning model prediction.The data-driven approach could accelerate the exploration of vanadate anode materials for lithium-ion batteries and provide ideas for the development of new materials in other fields.
[1] Sendek A D,Ransom B,Cubuk E D,et al.Machine learning modeling for accelerated battery materials design in the small data regime[J].Advanced Energy Materials,2022,12(31):2200553.
[2] Zhang Y,Ling C.A strategy to apply machine learning to small datasets in materials science[J].Npj Computational Materials,2018,4(1):25.
[3] 闫家辉,乔庆东,于廷云,等.无机材料对于锂电池隔膜改性研究进展[J].化工新型材料,2022,50(9):1-5.
[4] Joshi R P,Eickholt J,Li L,et al.Machine learning the voltage of electrode materials in metal-ion batteries[J].ACS Applied Materials & Interfaces,2019,11(20):18494-18503.
[5] Wei J,Chu X,Sun X Y,et al.Machine learning in materials science[J].InfoMat,2019,1(3):338-358.
[6] 施思齐,涂章伟,邹欣欣,等.数据驱动的机器学习在电化学储能材料研究中的应用[J].储能科学与技术,2022,11(3):739-759.
[7] Chen C,Zuo Y,Ye W,et al.A critical review of machine learning of energy materials[J].Advanced Energy Materials,2020,10(8):1903242.
[8] Min K,Choi B,Park K,et al.Machine learning assisted optimization of electrochemical properties for Ni-rich cathode materials[J].Scientific Reports,2018,8(1):15778.
[9] Honrao S J,Yang X,Radhakrishnan B,et al.Discovery of novel Li SSE and anode coatings using interpretable machine learning and high-throughput multi-property screening[J].Scientific Reports,2021,11(1):16484.
[10] Adam M L,Moses O A,Mailoa J P,et al.Navigating materials chemical space to discover new battery electrodes using machine learning[J].Energy Storage Materials,2024,65:103090.
[11] Zhao Y,Otto S K,Lombardo T,et al.Identification of lithium compounds on surfaces of lithium metal anode with machine-learning-assisted analysis of ToF-SIMS spectra[J].ACS Applied Materials & Interfaces,2023,15(43):50469-50478.
[12] Luo Y,Chen H,Wang J,et al.Machine-learning-assisted discovery of boron-doped graphene with high work function as an anode material for Li/Na/K-ion batteries[J].Physical Chemistry Chemical Physics,2023,25(17):12200-12206.
[13] Mistry A,Verma A,Sripad S,et al.A minimal information set to enable verifiable theoretical battery research[J].ACS Energy Letters,2021,6(11):3831-3835.
[14] Lv C,Zhou X,Zhong L,et al.Machine learning:an advanced platform for materials development and state prediction in lithium-ion batteries[J].Advanced Materials,2022,34(25):2101474.
[15] 李想,曹卓,但雅波,等.基于机器学习与多尺度特征的金属氧化物性能预测[J].化工新型材料,2019,47(9):146-150.
[16] Zhao J,Feng X,Pang Q,et al.Battery safety:machine learning-based prognostics[J].Progress in Energy and Combustion Science,2024,102:101142.
[17] Nguyen T,Jewik J,Bansal H,et al.Climatelearn:benchmarking machine learning for weather and climate modeling[J].Advances in Neural Information Processing Systems,2024,36,DOI:10.48550/arXiv.2307.01909.
[18] 牛程程,李少波,但雅波,等.基于机器学习的热导率预测模型研究[J].化工新型材料,2020,48(3):134-137,142.
[19] Zhao X,Li L,Zheng L,et al.3d-Orbital regulation of transition metal intercalated vanadate as optimized cathodes for calcium-ion batteries[J].Advanced Functional Materials,2024,34(2):2309753.
[20] Todorov Y M,Numata K.Effects of the Li∶(Mn+Co+Ni) molar ratio on the electrochemical properties of LiMn1/3Co1/3Ni1/3O2 cathode material[J].Electrochimica Acta,2004,50(2-3):495-499.
[21] Xia D,Gao H,Li M,et al.Transition metal vanadates electrodes in lithium-ion batteries:a holistic review[J].Energy Storage Materials,2021,35:169-191.
[22] Quispe-Garrido L V,Monje I E,López E O,et al.Influence of the molar ratio of Co and V in bimetallic oxides on their pseudocapacitive properties[J].ACS Omega,2022,7(48):43522-43530.
[23] Guo X,Chen C,Ong S P.Intercalation chemistry of the disordered rocksalt Li3V2O5 anode from cluster expansions and machine learning interatomic potentials[J].Chemistry of Materials,2023,35(4):1537-1546.
[24] Liu X,Wang T,Ji T,et al.Using machine learning to screen non-graphite carbon materials based on Na-ion storage properties[J].Journal of Materials Chemistry A,2022,10(14):8031-8046.
[25] Thakkar P,Khatri S,Dobariya D,et al.Advances in materials and machine learning techniques for energy storage devices:a comprehensive review[J].Journal of Energy Storage,2024,81:110452.
[26] Agrawal A,Choudhary A.Perspective:materials informatics and big data:realization of the “fourth paradigm” of science in materials science[J].Apl Materials,2016,4(5):053208.
[27] Mittal H,Kushwaha O S.Machine learning in commercialized coatings[J].Functional Coatings:Innovations and Challenges,2024:450-474.
[28] Xia D,Gao H,Li M,et al.Transition metal vanadates electrodes in lithium-ion batteries:a holistic review[J].Energy Storage Materials,2021,35:169-191.
[29] Manna S,Das A,Das S,et al.Machine learning assisted screening of MXene with superior anchoring effect in Al-S batteries[J].ACS Materials Letters,2024,6(2):572-582.
[30] Ahmad Z,Xie T,Maheshwari C,et al.Machine learning enabled computational screening of inorganic solid electrolytes for suppression of dendrite formation in lithium metal anodes[J].ACS Central Science,2018,4(8):996-1006.
[31] Jiang M,Zhang Y,Yang Z,et al.A data-driven interpretable method to predict capacities of metal ion doped TiO2 anode materials for lithium-ion batteries using machine learning classifiers[J].Inorganic Chemistry Frontiers,2023,10(22):6646-6654.
基金资助
国家自然青年科学基金项目(52104330)
PDF
