利用新的价连接性指数估算醇类化合物Kováts指数

Evaluation of the Kováts Index of Alcohols by a New Connectivity Index

以连接性指数为结构描述符,用多元回归技术建立了醇类化合物的结构与色谱保留值的相关性模型,各样本总体所建模型的相关系数均在0.98以上,这些模型较好地解释了醇类化合物保留指数的递变规律.为了检验模型的稳定性和预测能力,进行了留一法交互校验,相关系数均在0.98以上.该模型相关系数高,稳定性好,预测能力强.

A model of multiple linear regression method about the relationship between the structure of alcohols and their chromatographic retention indices has been set by means of connectivity indices. Six models,each of which is constructed by using all sample sets,with high correlation coefficient,r>0.98,are developed for six columns (SE-30,OV-3,OV-7,OV-11,OV-17,OV-25).These models can better elucidate the change rule of retention indices for the alcohols. In order to test the equation stability and prediction ability of each model,it is essential to perform a cross validation (CV) procedure. Satisfactory CV results have been obtained by using one external predicted sample every time with high correlation coefficients,r>0.98.And the QSRR models have high relative coefficients,good stability and good predictability.