论文部分内容阅读
多氯代二苯并呋喃(PCDFs)是环境持久性污染物,光化学降解是其主要降解途径。计算了PCDFs的原子特征值δ_i,利用分子图形技术获得连接性修饰指数~mX_L,以该指数作为分子结构描述符,采用最小二乘法用逐步多元线性回归分析48种PCDFs的光解半衰期,建立了2个能预测吸附到云杉针叶表面和飞灰表面的PCDFs光解半衰期的定量结构-性质关系模型,估算的平均误差分别为0.047和0.062。采用留一法检验模型稳健性,得预测模型,预测另外一些PCDFs分子的光解半衰期,预测值和实验值几乎一致。这些模型能解释PCDFs光解半衰期的递变规律,而且稳定性较好,预测能力较强。
Polychlorinated dibenzofurans (PCDFs) are persistent pollutants and photochemical degradation is the main degradation pathway. The atomic eigenvalues δ_i of PCDFs were calculated, and the connectivity modification index ~ mX_L was obtained by using molecular graphic technique. Using this index as the descriptor of molecular structure, the photolysis half-lives of 48 PCDFs were analyzed by using the least square method and stepwise multiple linear regression Two quantitative structure-property relationship models of photodegradation half-lives of PCDFs adsorbed onto spruce needle needles and fly ash surfaces were estimated. The estimated average errors were 0.047 and 0.062, respectively. The retention method was used to test the robustness of the model to obtain a predictive model to predict the photolysis half-life of other PCDFs. The predicted and experimental values were almost the same. These models can explain the gradual change law of photodecomposition half life of PCDFs, and they have better stability and stronger prediction ability.