太赫兹时域光谱用于混合物定性和定量分析

发布时间：2019-06-10 10:47

【摘要】：太赫兹波是处在电磁波谱中红外与微波之间的一个频段。超快激光在近十几年的发展为其产生提供了便利,因而推动了它发展,也使得太赫兹光谱技术成为国际学术界的前沿研究热点。太赫兹波特有的低能性、穿透性和指纹谱性等优点使其具有良好的应用前景。如低能性可用于无损检测,穿透性能用于安全检查,指纹谱性能用来识别物质的成分或种类。当下生活中,许多安全问题与我们生活息息相关,需要我们为之努力。例如每每触及人们敏感神经的食品安全问题中,有害添加剂、地沟油和假冒伪劣产品需要我们检测出来;令人惶恐不安的反恐安全问题中,毒品、炸药以及枪支道具等需要我们检测出来;日渐突显的环境问题中,大气污染指数需要实时监控。太赫兹波的优良特性,使其在物质检测方面具有很大潜力。太赫兹时域光谱技术是目前太赫兹波应用于物质研究的主要技术,它能获得物质在太赫兹波段的光学参数,通过对光学参数的分析,可以对物质的成分组成进行分析。本文基于太赫兹时域光谱技术,获得阿胶、鹿角胶、四种苯胺、奶粉和三聚氰胺的太赫兹光谱,对物质定性鉴别和混合物定量分析进行研究,成功将模式识别技术用于物质太赫兹光谱的分析中,并应用化学计量方法-偏最小二乘法成功对混合物进行定量分析。本文主要内容和工作创新如下:对本课题的研究背景进行阐述,说明了此项研究的目的。对太赫兹波产生以及探测机理和实验中常用到的透射式时域光谱系统做了详细介绍,并给出了从测得的光谱中提取所需光学参数的公式,详细介绍了样品的制备、测试过程及注意事项。在对物质的定性分析中,先是介绍了进行模式识别前数据预处理的方法如噪声滤除和特征提取,然后对物质模式识别所用到的人工神经网络和支持向量机做了原理上的推导。在获得阿胶和鹿角胶,对氯苯胺、对甲苯胺、对硝基苯胺和对甲氧基苯胺这两组样品的太赫兹光谱后,运用模式识别方法对其进行分析。实验结果表明,通过模式识别方法分析所获得的太赫兹吸收系数可成功对物质进行分类识别。在对物质的定量分析中,先是对定量分析中所用到的化学计量方法-偏最小二乘法做了充分研究,然后对不同混合比例三聚氰胺和奶粉混合物的太赫兹吸收系数谱进行研究,发现其中的线性关系,依据比尔-朗伯定律,结合PLS回归建立三聚氰胺含量回归模型。结果表明,太赫兹时域光谱技术能成功对混合物进行定量分析。
[Abstract]:Terahertz wave is a frequency band between infrared and microwave in the electromagnetic spectrum. The development of ultra-fast laser in recent years has provided convenience for its generation, so it has promoted its development, and terahertz spectroscopy has become a frontier research focus in the international academic community. The unique advantages of terahertz wave, such as low energy, penetration and fingerprint spectrum, make it have a good application prospect. If low energy can be used for nondestructive testing, penetration performance for safety inspection, fingerprint spectrum performance for identifying the composition or type of substances. In the present life, many security problems are closely related to our life, and we need to work hard for it. For example, in food safety problems that touch people's sensitive nerves, harmful additives, trench oil and fake and shoddy products need to be detected. Among the frightening counter-terrorism security problems, drugs, explosives and gun props need to be detected; in the increasingly prominent environmental problems, the air pollution index needs to be monitored in real time. The excellent characteristics of terahertz wave make it have great potential in material detection. Terahertz time domain spectroscopy is the main technology that terahertz wave is used in matter research at present. it can obtain the optical parameters of matter in terahertz band. Through the analysis of optical parameters, the composition of matter can be analyzed. Based on the terahertz time domain spectroscopy, the terahertz spectra of gum, antler gum, four kinds of Aniline, milk powder and melamine were obtained, and the qualitative identification and quantitative analysis of the mixture were studied. The pattern recognition technique was successfully applied to the analysis of terahertz spectra of substances, and the stoichiometric method-partial least square method was used to quantitatively analyze the mixture. The main contents and work innovation of this paper are as follows: the research background of this topic is described, and the purpose of this study is explained. The generation and detection mechanism of terahertz wave and the transmission time domain spectral system commonly used in experiments are introduced in detail, and the formula for extracting the required optical parameters from the measured spectrum is given, and the preparation of the sample is introduced in detail. Test process and matters needing attention. In the qualitative analysis of matter, the methods of data preprocessing before pattern recognition, such as noise filtering and feature extraction, are introduced, and then the artificial neural network and support vector machine used in material pattern recognition are deduced in principle. After obtaining terahertz spectra of gum and antler gum, p-chloranilide, p-toluidine, p-nitroanilide and p-methoxyphenylamine, the samples were analyzed by pattern recognition method. The experimental results show that the terahertz absorption coefficient obtained by pattern recognition method can successfully classify and recognize substances. In the quantitative analysis of substances, the partial least square method, which is the stoichiometric method used in the quantitative analysis, has been fully studied, and then the terahertz absorption coefficient spectra of the mixtures of different mixing ratios of melamine and milk powder have been studied. The linear relationship was found. According to Bill Lambert's law and PLS regression, the regression model of melamine content was established. The results show that terahertz time domain spectroscopy can successfully analyze the mixture.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：O441.4

【参考文献】