节点文献

近红外光谱分析技术在烟草化学中的应用研究

Researches on Applications of Near Infrared Spectroscopy in Tobacco Chemistry

【作者】 陈国辉

【导师】 王保兴;

【作者基本信息】 云南师范大学 , 生物化学与分子生物学, 2007, 硕士

【摘要】 烟草生产中,烟草中的糖、氮、碱等化学成分是配方设计和质量控制必不可少的指标,通常需花费大量的时间及经费进行分析,特别是在视质量为生命的市场竞争中及自动化程度日益提高的今天,发展适合卷烟工业生产的快速、高效、简便的分析方法,实现烟草生产的过程优化控制,从而保证烟草及烟草制品质量的稳定和均一是亟待解决的问题之一。由于近红外(NIR)光谱分析技术具有无损和多特征同时分析的特点,使原料和产品在线检测和实时监控成为可能,特别适合烟草自动化生产和控制的要求。因此,近红外光谱分析技术有望且正在发展成为烟草化学分析的重要技术。为此,本文对近红外光谱分析技术在烟草化学中的应用进行了比较系统的研究,具体包括以下几个内容:1.通过大量的中外文献阅读,简单阐述了烟草质量和烟草化学概念、近红外光谱分析技术发展概况及其在烟草领域的应用进展、论文选题背景及意义;介绍了近红外光谱分析仪器发展;介绍了近红外光谱预处理方法、波长选择方法、定性定量分析方法及模型转移等NIR中的化学计量学方法等技术。2.本文首先研究了NIRS分析烟草及其制品的准备实验方法。其中包括实验样品收集、样品制各及样品NIR光谱采集;烟草及其制品中主要化学成分测定方法确定;烟叶中多种有机酸的基于气质联用的扫描与选择离子法(GC/MS-Scan-SIM)快速同时测定方法的建立等。上述准备实验方法的研究,确保了样品的收集与制备、样品NIR光谱采集以及建模标准值测定的一致性,为后续建立烟草及其制品的近红外光谱校正模型奠定了良好的基础。3.以233个陈化烤烟烟叶样本为研究对象,利用近红外漫反射光谱技术,运用偏最小二乘法建立了烟草中21种有机酸含量以及挥发性有机酸总量快速测定的校正模型。结果表明,该方法快速、准确、无损,可以应用到大批量烟叶样品的快速定量测定。4.选取161个不同陈化质量(不同陈化时间)烤烟烟叶样品作为研究对象,在综合考虑烟叶常规化学成分、烟叶关键致香成分以及烟叶中有机酸含量的基础上结合陈化时间对烟叶进行质量分类,应用主成分分析-马氏距离方法建立了陈化烤烟烟叶质量评价的NIR定性判别模型。结果表明,应用本方法辅助烤烟陈化质量评估,相比化学成分分析以及感官质量评价等研究陈化的经典方法具有操作简便、速度快和精度高等特点,而且应用的是样品光谱的整体信息,充分体现了现代分析化学整体性的思想。5.以红云集团昆明卷烟厂四个卷烟品牌的多个批次卷烟成品作为研究对象,利用近红外漫反射光谱和近红外透射技术,运用主成分分析-马氏距离法,分别对品牌卷烟叶组配方烟叶和卷烟纸建立了定型判别模型。同时,考察了已建立烟叶常规化学成份测定的NIR模型对卷烟叶组配方烟叶的适配性检验。结果表明,在严格的实验条件下,原有NIR模型适用于卷烟品牌叶组配方烟叶定量分析,与标准方法测定结果相比,吻合度较高。在应用于卷烟品牌定性分析,具有良好的判别能力。该方法具有简便易行,快速的特点,可以应用于辅助卷烟品牌质量评价当中。6.选取不同生产配方的造纸法再造烟叶样品600个作为研究对象,分别建立了再造烟叶中总糖、还原糖、烟碱及Cl定量测定模型,并且,建立了三种不同生产配方(分别以A、B及C来表示)的主成分-马氏距离法定性模型,进而应用所建立的NIR定量定性模型对三种配方模式下的再造烟叶质量评价。结果表明,应用多变量分析方法评价再造烟叶质量能够更好的表征再造烟叶质量的特征,克服了单变量分析方法存在的局限性,能够更大程度上的提取再造烟叶质量特征的信息。7.选用巨豆三烯酮生产的3个中间体、粗提物以及成品作为其合成过程的控制点,应用近红外透射采样技术采集它们的近红外透射光谱,采用主成分分析-马氏距离法结合多变量统计过程控制(MSPC)原理,对巨豆三烯酮合成的过程进行监控与评价,结果表明,该方法适合于烟用香精巨豆三烯酮合成的过程质量控制,能够及时发现异常,迅速排除异因,使过程受控。本方法应用到实际生产当中获得良好的效果。

【Abstract】 As the growing use of tobacco,developing rapid,powerful,simple and convenient methods for tobacco manufacture process analysis and quality control has been gotten more and more attention.In recent years,near-infrared spectroscopy(NIRS) has grown dramatically and is widely used in many areas.The on-line and real time analysis,especially for process analysis and quality control of tobacco roboticized manufacture,can be true by reason of the lossless,simultaneous strongpoints of NIRS analysis technique.So the NIRS technique is properly suitable for tobacco industry manufacture.In the present paper,the application techniques of NIRS in tobacco chemistry are studied.The main contributions of this thesis are as follows:1.Firstly describe the conceptions of tobacco quality and tobacco chemistry,and review the applications and developments of NIRS in tobacco chemistry analysis,expatiate the significance and the background of the present paper,and then introduce the concepts and the principles of NIRS,NIR spectra pretreatment,NIR spectra region option,quantitative and qualitative analysis methods and medel transfer.2.Studies on the preparation experimental methods of tobacco chemistry analysis for tobacco manufacture using NIRS were first carried out.The present methods ensured the superiority of sample collection,sample pretreatment, sample NIR spectra collection and the accuracy of determination the active components in tobacco.These methods can be not only applied to process analysis and quality control of tobacco manufacture,but also effectively supported the further investigation about process analysis for tobacco products based on NIRS method.3.The calibration medels were established for determination of 21 organic acids and total volatile organic acids of tobacco leaves by partial least squares(PLS) and Fourier transform near infrared spectrometry(FT-NIRS) of 233 representative tobacco samples.The method has been applied to determination of organic acids of tobacco leaves with satisfactory results. 4.The classification distinguished models were developed for evaluating different flue-cured aging quality of Yunnan flue-cured tobacco during the 4-year aging time by pricinpal component analysis-Mahalanobis distance(PCA-MD) and Fourier transform near infrared spectrometry(FT-NIRS) considering the general characters of tobacco routine chemistry,tobacco key aroma compomemts and tobacco organic acids combined with aging time.The results of the distinguished power were satisfied.5.The qualitative discriminant analysis models were developed for 4 cigarette trademarks leaf blend and cigarette paper of Kunming Cigarette Factory,Hongyun Group by principal component analysis-Mahalanobis distance(PCA-MD) and FT-NIRS.The method is simple,rapid and can be used to evaluate quality of cigarette trademarks.6.The calibration models were developed for the analysis of total sugar,total reduce sugar,nicotine and chlorin in Paper-process reconstituted tobacco by PLS and FT-NIRS of 600 representative samples.Furthermore,the classification distinguished models were established for quality control of 3 different manufacture blends(A,B and C) by PCA-MD and FT-NIRS.The results showed that this method can be used to estimate the quality of Paper-process reconstituted tobacco by multivariate analysis.7.The process analysis and quality control model was developed for megastigmatrienone manufacture by PCA-MD and FT-NIRS combined with multivariate statistical process control(MSPC) using 3 intermediates,raw products and final products as objects.The results showed that this method can be used for process analysis and quality control in good time.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络