论文部分内容阅读
分子内部电荷分布不均使分子具有极性,分子的极性又影响了分子间的氢键、范德华力等相互作用力,进而影响宏观物质的化学特性,如化合物的溶解性、正辛醇,水分配系数和液相色谱保留行为等。化学特性的本质来源于其分子微观电荷状态,而偶极矩正是反映分子内电荷分布情况的重要物理量,因此估测分子的偶极矩十分重要。文中选取8种结构类型共107个分子,用Dragon软件计算出每个分子的929个分子参数,用多元线性回归法和主成分分析法分析参数对偶极矩值的影响。分析结果表明,相比主成分分析法,多元线性回归的stepwise法为最优建模方法。该模型统计结果良好(R~2=0.878,SEC=0.330,F=65.697),预测结果良好(R~2=0.792,SEP=0.665)。结果表明分子的平均原子电负性、距离矩阵参数、分子的组成、相邻原子的对称性、价键连接指数和分子的电性拓扑指数是影响偶极矩的主要因素。建立的最优回归模型对有机小分子的偶极矩具有很好的预测能力,对研究物质的宏观化学特性有一定的帮助作用。
Unbalanced distribution of charge inside the molecule causes the molecule to have polarity. The polarity of the molecule also affects the intermolecular hydrogen bonds, van der Waals forces and other interaction forces, thereby affecting the chemical properties of the macroscopic substance, such as the solubility of the compound, n-octanol, Water partition coefficient and liquid chromatographic retention behavior. The nature of the chemical properties comes from the molecular charge state of the molecule, and the dipole moment is an important physical quantity that reflects the charge distribution in the molecule. Therefore, it is very important to estimate the molecular dipole moment. In this paper, a total of 107 molecules were selected from 8 structures. The molecular parameters of 929 molecules were calculated by using Dragon software. The influence of the parameters on the dipole moment was analyzed by multivariate linear regression and principal component analysis. The analysis results show that compared with the principal component analysis, the stepwise method of multiple linear regression is the optimal modeling method. The model has good statistical results (R ~ 2 = 0.878, SEC = 0.330, F = 65.697), and the prediction results are good (R ~ 2 = 0.792, SEP = 0.665). The results show that the average atomic electronegativity, the distance matrix parameter, the composition of the molecule, the symmetry of the adjacent atom, the valence bond index and the electrical topological index of the molecule are the main factors affecting the dipole moment. The established optimal regression model has good predictive ability for the dipole moment of small organic molecules, which will help to study the macroscopic chemical properties of the material.