大气中痕量挥发性有机物气相色谱保留时间的预估  

Estimation and forecasting of chromatographic retention time of the volatile organic trace compounds in the atmosphere

作　　者：廖立敏[1] 梅虎[1] 李建凤[1] 朱万平[1] 舒茂[1] 郑怀礼[1] 李志良[2] 

机构地区：[1]重庆大学化学化工学院/三峡库区生态环境教育部重点实验室 [2]1重庆大学化学化工学院/三峡库区生态环境教育部重点实验室

出　　处：《安全与环境学报》2008年第1期11-14,共4页Journal of Safety and Environment

基　　金：重庆市自然科学基础项目(CSTC-2004BA7019);国家自然科学基金项目(20577072)

摘　　要：为探寻环境中挥发性有机物气相色谱保留行为与结构的关系,应用三维原子场全息作用矢量(3D-HoVAIF),对39个大气中痕量挥发性有机物进行了结构表征。采用逐步回归对变量进行筛选,建立了8变量定量结构色谱保留相关(QSRR)模型,模型的复相关系数为R1=0.985;进而将3D-HoVAIF矢量与相对分子质量结合,建立了9变量QSRR模型,模型的复相关系数为R2=0.991;采用留一法交互检验,结果分别为RCV,1=0.970、RCV,2=0.978,表明模型具有良好的估计能力和稳定性。The present paper is aimed at introducing the author＇s newly developed three-dimensional holographic vector of atomic interaction and the corresponding QSRR modal of 9 variables for predicting the chromatographic retention data. As is known, in order to explore the relationship between volatile organic compounds （VOC） in the environment and their structures, we are using a 3D-HoVAIF field in our laboratory to describe the chemical composition of 39 VOC in the atmosphere. On the basis of the so-called QSRR model of 8 variables set up by the stepwise multiple regression （SMR） with the correlation coefficient （R） of 0.985, a new model of 9 variables has been deduced from the 3D-HoVAIF and molecular weight by multi-linear regression （MLR） with the correlation coefficient （R） of 0.991. Then, the predicted and observed gas chromatographic retention durations are confirmed to be in conformity with cross-validation（CY）, gained by using a leave-one-out （LOO） procedure with the help of correlation coefficients（ Rcv ） of 0.970 and 0.978 for the models, respec- tively. Our experimental results prove that the above-said models can satisfy the estimation stability and favorable predictable abilities. However, since the retention time of a compound in a chromatographic column depends on the interactions of the solution in the stationary phase, such interactions should be founded on the structural, chemical and electronic properties of the compound. Therefore, from our research, it can be concluded that the 3D-HoVAIF helps to successfuly describe the structures as well as the chemical and electronic properties of the compounds from the environment, including in this given situation alkane, alkene, arene and so on so as to extend the span of their molecular structures quite large and, in turn, be characteristic of chemical compositions from the environment. Thus, compared with some other reported researches, this study, since breaking through 3D-HoVAIF, can be applied to the QSRR research conditi

关 键 词：环境分析 挥发性有机物 三维原子场全息作用矢量 定量结构色谱保留相关 

分 类 号：O641[理学—物理化学]