【摘要】 目的:通过建立多元线性回归(MLR)模型研究多环芳烃(PAHs)一类难降解污染物。方法:使用Randic指数(X0,X1),对90种PAHs的表面积(A),体积(V)和气相色谱保留时间(I)相关系数的关联,进行多元线性回归(MLR)分析。结果:2个描述子和3个理化性质系数(V,A和I)的相关系数(r)都超过0.99。Randic指数与3种物化性质的关联系数分别达到了r=0.9959,r=0.9973,r=0.9912。结论:结果表明,模型有良好的线性相关性,由方程得出的预测值与实验值之间能很好地吻合,为预测和估算多环芳烃的表面积、体积、气相色谱保留指数提供了一个简便的方法并能扩展现有的实验数据。更多还原

【Abstract】 Objecttive: To establish the multiple linear regression(MLR) models for studying on polyclic aromatic hydrocarbons(PAHs),a sort of refractory organics.Methods:The data set was composed of 90 PAHs compounds,including subsituents with different steric and electronic characteristics.Multiple linear regression was employed to create the models best suited for the analysis of additive property of PAHs.A correlation equation including Polyclic aromatic hydrocarbons(PAHs) have received more and more concerns as a group of ubiquitous potential persistent organic pollutants(POPs).By using Randic index,multiple linear regression(MLR) models were developed for area(A),volume(V) and(I) of 90 PAHs.Results:The correlation coefficients of Area,Volume,I with 2 Randic index of PAHs were 0.9959,0.9973 and 0.9912,respectively.Their correlation coefficients were larger than 0.99.Conclusion:The Randic topological index has highest structural selectivity,good property relativity and easy calculation.The quantitative structure-property/retention relationship(QSPR/QSAR) models have high relative coefficients and good stability.These models can be used to predict the property/retention of PAHs and the model predictions can hence extend the current database of experimental values.更多还原