School of Chemistry and Biological Engineering, Hunan Provincial Key Laboratory of Materials Protection for Electric Power and Transportation, Changsha University of Science and Technology, Changsha 410004, China.
J Mol Graph Model. 2013 Jul;44:113-9. doi: 10.1016/j.jmgm.2013.04.007. Epub 2013 May 4.
The novel electro-negativity topological descriptors of YC, WC were derived from molecular structure by equilibrium electro-negativity of atom and relative bond length of molecule. The quantitative structure-property relationships (QSPR) between descriptors of YC, WC as well as path number parameter P3 and the normal boiling points of 80 alkanes, 65 unsaturated hydrocarbons and 70 alcohols were obtained separately. The high-quality prediction models were evidenced by coefficient of determination (R(2)), the standard error (S), average absolute errors (AAE) and predictive parameters (Qext(2),RCV(2),Rm(2)). According to the regression equations, the influences of the length of carbon backbone, the size, the degree of branching of a molecule and the role of functional groups on the normal boiling point were analyzed. Comparison results with reference models demonstrated that novel topological descriptors based on the equilibrium electro-negativity of atom and the relative bond length were useful molecular descriptors for predicting the normal boiling points of organic compounds.
从分子结构出发,通过原子平衡电负性和分子相对键长,导出了 YC、WC 的新型电负性拓扑描述符。分别获得了 YC、WC 的描述符以及路径数参数 P3 与 80 种烷烃、65 种不饱和烃和 70 种醇的沸点之间的定量结构-性质关系(QSPR)。通过决定系数(R²)、标准误差(S)、平均绝对误差(AAE)和预测参数(Qext²、RCV²、Rm²)来证明高质量的预测模型。根据回归方程,分析了碳原子主链长度、分子大小、分支度以及官能团的作用对沸点的影响。与参考模型的比较结果表明,基于原子平衡电负性和相对键长的新型拓扑描述符是预测有机化合物沸点的有用分子描述符。