电性拓扑态指数预测PCDFs光解半衰期

以电性拓扑态指数(E-state indices)为参数, 应用偏最小二乘法(PLS)对48种多氯代二苯并呋喃(PCDFs)在云杉(Picea abies)针叶和飞灰表面的光解半衰期(t_1/2)进行模拟分析, 分别获得5变量和4变量定量结构-性质相关(QSPR)模型. 模型交叉验证相关系数(Q_cum²)分别为0.758和0.701, 标准偏差(SD)分别为0.047和0.049, 模型具有较高的预测能力和可靠性. 随着氯原子取代数目的增多, PCDFs与活性氢供体接触的“位阻效应”增大, PCDFs分子中苯环的静电荷密度随之降低, 以致高氯代PCDFs较难发生光解作用.

Prediction of Photolysis Half-lives of PCDFs with the Electrotopological State Indices

Using partial least-squares (PLS) regression and electrotopological state indices (ESI), two quan- titative structure-property relationship models on photolysis half-lives (t1/2) of polychlorinated dibenzofurans (PCDFs) on spruce (Picea abies) needles and fly ash surfaces were developed. For the optimal quantitative structure-property relationship (QSPR) models, the cross validated coefficients Qc2u m were 0.758 and 0.701, the standard deviation values were 0.047 and 0.049 respectively. The QSPR models have great robustness and good predictive abilities. The results show that the stability of the PCDF molecules increases with the increasing of chlorine atoms in the parent molecules, which may be a result of steric hindrance and low elec- tron density on phenyl rings of highly chlorinated PCDFs.