【作者】 付绍晨；

【导师】 肖夏；

【作者基本信息】 天津大学 ， 微电子学与固体电子学， 2009， 硕士

【摘要】 LSAWs技术是利用声表面波在多层薄膜介质中传播时产生色散的原理来进行样品测量的。本文研究声表面波在分层媒质中传播模型及色散理论,着重对LSAWs实验系统优化,数据采集及处理等的研究进展进行阐述。在理论方面,建立了各向同性薄膜/硅衬底的计算模型,通过理论推导,得到了声表面波频散曲线和各参数之间的关系。本文以薄膜的硬度为主要切入点,重点研究了薄膜的杨氏模量同声表面波频散曲线之间的关系,并通过一定的仿真,获得了理论频散曲线的特性。在实验中,分析各个因素对实验信号的影响,深入研究了信号的屏蔽处理方法,完成了对实验系统的优化,实现了实验微弱信号的采集,并获得了具有可观信噪比的声表面波信号。考虑到匹配的精确性,对实验信号进行了进一步的滤波处理,本文通过对数字滤波器的研究和比较,设计了以凯瑟窗为基础的滤波器,并通过实验数据对窗函数参数进行调试,防止了频谱的泄漏,通过一定的仿真获得了很好的实验信号。为获得实验的频散曲线,研究了快速傅立叶变换的相关理论并随后进行了频谱分析,就DFT变换容易出现的问题,通过末尾增零的方法来减弱频谱的栅栏效应,最终得到了符合要求的实验频散曲线。本文采用了基于最小二乘原理的曲线平滑算法对实验频散曲线进行处理,并基于实验研究中发现的频散曲线的特殊规律提出了一种基于二分法的快速查找算法,大大提高了查找速度和匹配精度,完成对实验曲线和理论曲线的匹配,自动得到被测样品的杨氏模量。更多还原

【Abstract】 LSAWs (Laser-generated Surface Acoustic Waves) technology measures samples based on the dispersion of surface acoustic waves when propagating along thin multilayer film media.This article studies the dissemination models and dispersion theory of surface acoustic waves progating along layered media, focusing on system optimization of LSAWs experiment and research of data acquisition and processing.In theoretical aspect, the "isotropic thin film / silicon substrate" calculation model was established, the SAW dispersion curve and the relationship between each parameter were obtained. In this paper, taking the hardness of thin films as the main entry point, the relationship between Young’s modulus of thin film and the surface wave dispersion curve was investigated, the theoretical dispersion curve characteristics was obtained through simulation.During experimental procedural, influence of various factors on the experimental signal was analyzed, the shielding and processing methods of the signal were studied, the optimization of the experimental system was completed, the experimental weak signal was acquired, and surface acoustic wave signal with a substantial signal to noise ratio was obtained. Through research and comparison of digital filter, Kaiser window-based filter was designed, and the experimental data were taken to debug window function parameters, thus preventing the leakage of the spectrum. good experimental signal was obtained through simulation. In order to obtain the experimental dispersion curve, relevant theory on the fast Fourier transform was studied and a frequency spectral analysis was carried out. In view of common problems in DFT transform, the fence effect of spectrum was weaken by adding zero at the end of the signal, and ultimately the experimental dispersion curve was obtained in accordance with the requirements.In this paper, based on the principle of least squares curve smoothing algorithm, the experimental dispersion curve was handled. According to the special law of dispersion curves found during experimental study, a dichotomy-based fast search algorithm was proposed which greatly improved the search speed and matching accuracy. This method can complete the match of experimental curve and theoretical curve and acquire the Young’s modulus of measured sample automatically.更多还原