【作者】 朱绍华；

【导师】 吴海龙； 俞汝勤；

【作者基本信息】 湖南大学 ， 分析化学， 2008， 博士

【摘要】 化学计量学是一门新兴发展的交叉学科,它运用数学、统计学和计算机科学等方法设计最优化学量测策略,通过对化学数据的分析处理最大限度地获取化学及相关信息。当前化学计量学的研究主要分为两个方面:化学计量学基础理论与方法的研究;化学计量学方法在化学及其相关领域中应用,可以解决传统方法无法处理的问题,体现化学计量学的优势。本文作者在仔细分析了当前国内外化学计量学发展方向及其研究动态的基础上,选取高维数据分析中的几个重要问题进行了深入系统的理论、方法和应用研究。随着可产生高阶数据的高阶分析仪器的不断发展,高维数据的分辨与校正已成为化学计量学领域的一大热点,在处理复杂化学体系的无扰动、实时在线分析以及传统分析化学难于处理的复杂多组份平衡与动力学体系的解析方面有着独特的优势。本论文主要涉及以下几个方面:1.多线性成分模型(第二章)为了进一步理解基于多线性成分模型而提出的分解算法以及这些算法在实际中的应用,有必要对多线性成分模型的优势及特点进行阐述和探索。本章从多线性成分模型的数学表述及图形表述入手,阐述了其相应模型的本质及相互联系。这些工作也可为数学学科多线性代数的发展提供很有意义的参考。2.多线性数据分析在动力学研究方面的应用(第三章-第四章)第三章应用三线性平行因子分析方法直接检测了河水和自来水中西维因及其降解产物1-萘酚的含量,并应用此方法研究西维因在河水和自来水中的降解过程,得到了满意的结果。第四章应用四线性平行因子分析方法检测了农业污水中西维因的含量,同时利用四线性分解的优势,同时对其水解过程进行研究,得到了满意的结果。对化学反应的动态过程进行研究,是化学的一个重要方面,用常规的分析方法,很难在未知成分干扰下对研究对象在复杂体系中的动力学过程进行研究,而这两个工作充分利用了二阶校正和三阶校正的优势,能够直接对其动态过程进行解析,体现了多线性数据分析在动力学研究方面的潜力和优势。3.四线性分解方法的研究(第五章)随着研究体系日益复杂、分析仪器更加智能化、分析手段更加丰富,使得处理四维响应数据阵也越来越迫切。当四维数据基于四线性成分模型进行定量预测分析时,这一方法可被称作“三阶校正”,其具有的优势叫做“三阶优势”。理论上,三阶校正的优势不仅包括“二阶优势”,即能够在未知干扰共存的情况下也能对感兴趣组分进行定量分析,而且还会包括更多的优势。然而,目前我们还仍然缺乏对四维数阵四线性成分模型的理论优势的整体理解,因此对于四线性成分模型理论和三阶优势的探索已变得十分迫切而必要。本文提出了交替不对称四线性分解算法。为了探索高阶数据的分析特征,我们运用分别基于该算法和四维平行因子分析算法的三阶校正方法用于处理模拟的和实际的四维数阵。结果显示,这两种四维算法都能在定量分析上获得满意的结果。新算法与四维平行因子分析算法相比较,新算法收敛速度要明显快于四维平行因子分析算法,并且对模型的预估成分数不敏感,这种特性避免了在三阶校正中模型需要选择正确组分数的尴尬。4.三线性数据分析在农药分析方面的应用(第六章-第八章)农药主要是指用来防治危害农林牧业生产的有害生物(害虫、害螨、线虫、病原菌、杂草及鼠类)和调节植物生长的化学药品,但通常也把改善有效成分物理、化学性状的各种助剂包括在内。由于农药的滥用,产生了许多环境问题和公共卫生问题,对自然界的生态平衡和人的生存健康都产生了很大的危害,因此农药残留的问题引起了世界各国和国际组织的重视。第六章采用基于自加权三线性分解(SWATLD)的二阶校正方法,来定量检测蜂蜜中的多菌灵、西维因和1-萘酚,并与二阶标准加入法和液相色谱与质谱联用方法的结果相一致。第七章采用三种二阶校正方法:平行因子分析法(PARAFAC)、自加权交替三线性分解方法(SWATLD)和交替惩罚三线性分解方法(APTLD)的二阶优势,利用数学分离来代替化学分离,实现严重干扰下对多菌灵定量测定。第八章采用交替三线性分解法检测苹果和香蕉中西维因的含量。更多还原

【Abstract】 Chemometrics is a developing composite discipline. It uses the methods of mathematics, statistics and computer sciences to extracting the optimal scheme for chemical measurements and to elucidating the data collected from chemical measurements. There are two important aspects in study on chemometrics: one is the research on basic theories and methods, another is on applications of chemometrics to chemistry and relative scientific fields.The author has been analyzing carefully the development direction and the research focus of chemometrics, and carried out the research on multi-way chemometric methodologies and their applications. With the development of high-order analytical instrumentation, multi-way data analysis has become an active domain with practical significance. Chemometric methodologies not only comprehensively expand the fundamental theory of modern analytical chemistry, but also provide a variety of powerful techniques for direct and on-line analysis of complex chemical systems, which are generally hard to handle by conventional analytical techniques.Studies presented in the thesis primarily deals with the following aspects of multi-way data analysis in chemometrics.1. The multilinear component model (Chapter 2):In order to further understand the decomposition algorithms based on multilinear component model and the applications of these algorithms in practice, it is necessary to realize and explore the characteristics and advantages of multilinear decomposition algorithms. Based on the mathematical and graphic expressions of multilinear component model, the nature of these decomposition algorithms were analyzed in details. The algorithms of multi-way arrays and methodologies provide interesting hints to develop multilinear algebra in mathematics.2. Application of multilinear data analysis to the study of dynamics (Chapter 3– Chapter 4):In Chapter 3 it provided a novel method to determine carbaryl and 1-naphthol in river water and tap water and used this method to investigate the hydrolysis of carbaryl. In Chapter 4 a new method to determine carbaryl in effluent was given and the hydrolysis of carbaryl in effluent with the aid of three-order calibarion was studied. Monitoring a chemical reaction is an important aspect in chemistry. But it is difficult to achieve this aim in presence of uncalibrated interference in complex system with traditional methods. It was convenient to implement the aim using the methods in this thesis and validates the power and potential of second-order calibration and three-order calibration.3. The quadrilinear decomposition method (Chapter 5):The instruments that generate quadrilinear data array are available to chemist, it is necessary to develop a quadrilinear decomposition method to analyze quadrilinear data array. It can be used to three-order calibration and it not only retains the second-order advantage of second-order calibration but also holds additional advantages. They can be defined the‘third-order advantage’. However, the complete third-order advantages are still unknown. So it is pressing to explore the theory of quadrilinear decomposition and third-order advantage. In Chapter 5 a novel method, alternating asymmetric quadrilinear decomposition (AAQLD), is developed for decomposition of quadrilinear data and applied to third-order calibration. By treating simulated and real data sets, the results indicated that both AAQLD and PARAFAC work well. It was shown a much higher convergence rate compared with quadrilinear PARAFAC. Moreover, it is generally insensitive to the overestimates of the component number chosen. This offers the advantage that in third-order calibration one need not pay much attention to determining a proper component number for the model, and it is difficult for quadrilinear PARAFAC to avoid it.5. The applications of three-way data analysis in Pesticide analysis (Chapter 6 to Chapter 8):Pesticides were used to control mainly harmful organisms (such as pests, pest mites, nematodes, pathogens, weeds and rodents), which harm agriculture, forestry and animal husbandry and regulate plant growth. However, there are a lot of environmental issues and public health issues for the abuse of pesticides. The issue of pesticide residue has aroused attention of many countries and international organizations. In Chapter 6 it provided a novel method to determine carbendazim, carbaryl and 1-naphthol in honey, and validate it by second-order standard addition method (SOSAM) and high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS). In Chapter 7 with the aid of three second-order calibaration methods, parallel factor analysis (PARAFAC), selfweighted alternating trilinear decomposition (SWATLD) and alternating penalty trilinear decomposition algorithm (APTLD), quantifying carbendazim in banana was performed. In Chapter 8 a new method for determination of carbofuran in apple and banana is proposed with the aid of alternating trilinear decomposition (ATLD) coupled with excitation–emission matrix fluorescence.更多还原