期刊文献+

基于SVR的脉冲激光沉积TiN/AlN多层薄膜的工艺优化 被引量:2

Process parameters optimization for TiN/AlN multilayer films fabricated by pulsed laser deposition via SVR
下载PDF
导出
摘要 根据脉冲激光沉积(PLD)法在单晶Si试样表面沉积制备多层TiN/AlN硬质膜实验数据,应用基于粒子群算法(PSO)寻优的支持向量回归(SVR)方法,建立不同工艺参数下沉积的TiN/AlN多层膜的AlN膜厚及TiN薄膜硬度的SVR预测模型。在相同的训练与测试样本集下,将SVR所得的AlN膜厚预测值与免疫径向基函数(IRBF)神经网络的计算结果进行比较。结果表明,SVR模型训练和预测结果的平均绝对百分误差要比IRBFNN模型的小,其预测精度更高,预测效果更好。应用SVR的TiN薄膜硬度模型对PLD法沉积TiN薄膜的工艺参数进行了优化,分析了多因素对PLD法沉积TiN薄膜硬度的交互作用和影响。该方法可为人们利用PLD法沉积TiN/AlN多层功能薄膜提供科学的理论指导,具有重要的理论意义和实用价值。 Based on the experimental dataset of TiN/AlN multilayer films ablated on the monocrystalline silicon substrate via pulse laser deposition(PLD) technique,the support vector regression(SVR) combined with particle swarm optimization(PSO) is proposed to construct models for prediction the thickness of AlN thin films and hardness of TiN thin films in TiN/AlN multilayer films deposited under different process parameters.The predicted hardness of AlN films via established SVR model is compared with that obtained by immune radial basis function(IRBF) neural network using identical training and test samples.It is demonstrated that the mean absolute percentage errors achieved by SVR both for the training set and test set are smaller than those of IRBFNN,respectively.SVR possesses higher prediction accuracy and a better forecast capacity than IRBFNN.The established SVR model for TiN hardness was further employed to analysis and optimized the PLD deposition process,and was utilized to depict the interaction influences of multi-factor on the hardness of deposited TiN films.The method introduced in this study can provide scientific and theoretical guide in fabrication of TiN/AlN multilayer films via PLD,and it would be of important practical application value.
出处 《功能材料》 EI CAS CSCD 北大核心 2013年第14期2074-2078,共5页 Journal of Functional Materials
基金 中央高校基本科研业务费资助项目(WLYJSBJRCTD201102) 教育部新世纪优秀人才支持计划资助项目(NCET-07-0903) 留学回国人员科研启动基金资助项目(2008101-1) 重庆市自然科学基金资助项目(CSTC2006BB5240)
关键词 脉冲激光沉积 TIN AlN硬质多层膜 支持向量回归 回归分析 工艺优化 pulse laser deposition TiN/AlN multilayer films support vector regression regression and analysis process optimization
  • 相关文献

参考文献4

二级参考文献58

共引文献19

同被引文献83

  • 1郭延龙,王小兵,万强,卢常勇.超快脉冲激光沉积类金刚石膜的研究进展[J].激光与光电子学进展,2005,42(7):22-25. 被引量:10
  • 2Serbezov V, Sotirov S, Benkhouj K. Investigation of superfast deposition of metal oxide and diamond-like carbon thin films by nanosecond Ytterbium (Yb) fiber laser[J]. Optical Materials, 2014, 36(1): 53-59.
  • 3Marcu A, Stokker F, Zamani R R, et al.. High repetition rate laser ablation for vapor-liquid-solid nanowire growth[J]. Current Applied Physics, 2014, 14(5): 614-620.
  • 4Bonisa A D, Santagata A, Sansone M, et al.. Femtosecond pulsed laser ablation of molybdenum carbide: Nanoparticles and thin film characteristics[J]. Applied Surface Science, 2013, 278(2): 321-324.
  • 5Usmana A, Rafiquea M S, Rahmana M K U, et al.. Growth and characterization of Ni:DLC composite films using pulsed laser deposition technique[J]. Materials Chemistry and Physics, 2011, 126(6): 649-654.
  • 6Zhang K J. Formation of complex bis (fl-mercaptobenzothiazole)-zinc (11) films by pulsed laser deposition[J]. Applied Surface Science, 2013, 273(7): 836-840.
  • 7Junji H, Shintaro I, Kilner J A, et al.. Electronic and oxide ion conductivity in PrNi0.7Cu02Ga005OJCe0.sSm0.O2 laminated film [J]. Solid State Ionics, 2013, 230(1): 16-20.
  • 8Jing C, Yang Y J, Wang X L, et al.. Epitaxial growth of single-crystalline Ni46Co4Mn37In13 thin film and investigation of its magnetoresistance[J]. Progressin Natural Science: Materials International, 2014, 24(1): 19-23.
  • 9Han S, Shao Y K, Lu Y M, et al.. Effect of oxygen pressure on preferred deposition orientations and optical properties of cubic MgZnO thin films on amorphous quartz substrate[J]. Journal of Alloys and Compounds, 2013, 559(2): 209-213.
  • 10Yang N, Tebano A, Dicastro D, et al.. Deposition and electrochemical characterization of Yttrium doped Barium cerate and zirconate hetero structures[J]. Thin Solid Films, 2014, 562(3): 264-268.

引证文献2

二级引证文献13

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部