化学学报 ›› 2006, Vol. 64 ›› Issue (7): 629-636. 上一篇    下一篇

研究论文

用量子化学参数研究烯烃聚合物定量构效关系

禹新良,王学业*,高进伟,李小兵,王寒露   

  1. (湘潭大学化学学院 湘潭 411105)
  • 投稿日期:2005-01-17 修回日期:2005-12-19 发布日期:2006-04-15
  • 通讯作者: 王学业

QSPR Studies on Vinyl Polymers Based on Quantum Chemical Descriptors

YU Xin-Liang, WANG Xue-Ye*, GAO Jin-Wei, LI Xiao-Bing, WANG Han-Lu   

  1. (College of Chemistry, Xiangtan University, Xiangtan 411105)
  • Received:2005-01-17 Revised:2005-12-19 Published:2006-04-15
  • Contact: WANG Xue-Ye

以密度泛函理论(DFT)方法所得的烯烃聚合物结构单元的物性参数如总能量Et、内能Ein、等容比热CV、熵S、四极矩Qii、偶极矩 µ、平均极化率α及原子最大负电荷q等8个量子化学参数, 用逐步回归法分别建立了这些参数与摩尔体积V298 K, 摩尔等张体积Ps、摩尔吸收常数色散分量Fd、摩尔折射率RLL、摩尔抗磁磁化率χ、摩尔粘度温度函数Hvsum、摩尔Rao函数UR及摩尔Hartmann函数UH的结构-性能定量关系 (QSPR) 模型, 其测试集的决定系数R2分别是: V298 K 为0.989, Ps为0.982, Fd为0.975, RLL为0.997, χ为0.988, Hvsum为0.914, UR为0.988及UH为0.972. 结果表明, 用这些量子化学参数所建立的聚合物QSPR模型能用于聚合物性能的预测.

关键词: 密度泛函理论, 量子化学参数, 结构-性能定量关系, 聚合物

Density functional theory calculations were carried out for repeated units of vinyl polymers, such as polyvinyls, polyacrylates, polymethylacrylates, polystyrenes and polyalkenes. The calculated results of the total energy Et, the internal energy Ein, the heat capacity at constant volume CV, the entropy S, the quadrupole moment Qii, the dipole moment µ, the average polarizability of the molecule α and the most negative net charge of atom q were used to predict the molar volume at room temperature V298 K, the molar parachor of Sugden Ps, the dispersion component Fd of the molar attractive force constant, the molar refraction of Lorentz and Lorenz RLL, the molar diamagnetic susceptibility χ, the molar viscosity-temperature function estimated as a sum of the structural units in the repeated unit Hvsum, the molar Rao ultrasonic velocity function UR and the molar Hartmann ultrasonic velocity function UH. Eight quantitative structure-property relationship (QSPR) models obtained from the training sets were evaluated externally using the test sets. Data of correlation coefficient R2 between the predicted values and experiment values are: 0.989 for V298 K, 0.982 for Ps, 0.975 for Fd , 0.997 for RLL, 0.988 for χ, 0.914 for Hvsum, 0.988 for UR and 0.972 for UH. The results indicate that the QSPR models constructed by such quantum-chemical descriptors can be well used to predict the properties of polymers.

Key words: density functional theory, quantum-chemical descriptor, quantitative structure-property relationship, polymer