论文部分内容阅读
用DFT-B3LYP方法,在较高基组6-311G**水平上对28种取代苯甲醛类化合物进行全优化计算获得了相应量子化学参数,利用线性逐步回归法建立取代苯甲醛溶解度的定量结构-活性相关性(QSPR)模型.采用内部及外部双重验证的办法深入分析和检验模型的稳健性,选出了最佳模型,其复相关系数(R2),留一法(LOO)交互检验复相关系数(Rc2v),外部预测样本复相关系数(Re2xt)分别为0.935,0.933和0.842,表明所建立的QSPR模型的稳定性和预测能力良好.结果表明:取代苯甲醛的溶解度lgSW与分子总能量,分子体积,分子最低空轨道能和最负原子的静电荷相关性较好.
Quantitative chemical parameters of 28 substituted benzaldehydes were obtained by DFT-B3LYP method at the level of 6-311G **. The quantitative structure of substituted benzaldehyde was established by linear stepwise regression - Activity Dependency (QSPR) model.Using internal and external double verification method to deeply analyze and test the robustness of the model, the best model was selected, and its correlation coefficient (R2), LOO The correlation coefficient (Rc2v) and the external correlation coefficient (Re2xt) were 0.935, 0.933 and 0.842, respectively, indicating that the established QSPR model has good stability and prediction ability.The results show that the solubility of substituted benzaldehyde lgSW and the total molecular energy The correlation between the molecular volume, the lowest empty orbital energy of the molecule and the electrostatic charge of the most negative atom is better.