应用线性溶剂化能参数预测多氯有机物的分配性质.pdfVIP

中国环境科学 2002,22(4)：329~333 China Environmental Science 应用线性溶剂化能参数预测多氯有机物的分配性质 * 戴家银,陈 玲,赵建夫 (同济大学污染控制与资源化研究国家重点实验室,上海 200092) 摘要：应用线性溶剂化能方法构建定量结构-性质相关(QSPR)模型.该模型能较好地预测多氯有机物的正辛醇/水分配系数(Kow)和沉积物吸 附系数(Koc). Kow和Koc平均误差分别为0.43和0.37个对数单位.模型的稳健性检验结果表明,模型中自变量不存在相关关系,残差呈正态分布; 对Koc来说,PCB和PCDD/PCDF在构建模型中起核心作用,剔除该类化合物导致模型相关系数明显下降,农药类和氯硝基苯类化合物的存在 降低了模型的精度. 关键词：正辛醇/水分配系数；沉积物吸附系数；定量结构-性质相关；稳健性 中图分类号：X131 文献标识码：A 文章编号：1000-6923(2002)04-0329-05 Prediction of partition property of polychlorinated organic compounds using linear solvation energy parameter. DAI Jia-yin, CHEN Ling, ZHAO Jian-fu (State Key Laboratory of Pollution Control and Resource Reuse, Tongji University, Shanghai 200092, China). China Environmental Science. 2002,22(4)：329~333 Abstract：Linear solvation energy technique was used to develop quantitative structure-property relationships (QSPR) model. The model could predict more effectively the octanol/water partition coefficient (Kow) and the sediment sorption coefficient (Koc) of the polychlorinated organic compounds (PCOCs). The mean errors of Kow and Koc of the compounds were 0.43 and 0.37 log units respectively. Results of robustness test of the model demonstrated that the independent variables in the model did not interrelate and the residuals presented normal distribution. For Koc, PCB and PCDD/PCDF are important in developing the model, the interrelation coefficient of the model decreased markedly by deleting these compounds; the accuracy of the model decreased due to existence of pesticides and chlorinated nitrobenzene compounds. Key words ：octanol/water partition coefficient；sediment sorption coefficient；quantitative structure-property relationships ； robustness 目前我国每年排放干污泥约为500万t,部分 既困难又费时.线性溶剂化能相关(LSER)被公认 地区已在尝试污泥的土地利用,让污泥回归于自 是一种十分有效的定量结构-性质相关(QSPR) [1] 然 .美国环境保护局的统计表明,污泥的土地利 模型.通过LSER方程式可精确预测化学品的 用已经成为污