摘要
基于化合物分子的拓扑图结构建立的分子电负性距边连接性指数 (anovelscalarindexofmolecularelectroneg ativitydistance -edgeconnectivity ,Imedc ,Λ) ,在对有机物的定量构效关系 (QSAR QSPR)研究中 ,具有高度选择性、相关性、稳定性和良好的预测能力 ;本文尝试将其扩展到无机物 :以分子电负性距边连接性指数 (Imedc ,Λ)实现对金属卤代化合物的结构性质的表征及相关研究。运用多元线性回归 (MLR)方法 ,以此分子电负性距边连接性指数建立 2 0个碱金属卤代化合物的晶格能 (U)和 34个过渡族金属卤代化合物的标准生成焓 (ΔfHθ)结构性质相关模型 ,对于碱金属卤代化合物的晶格能相关模型的相关系数为RMX =0 .9732 (n =2 0 ) ,对于过渡族金属卤代化合物的标准生成焓RMX=0 .890 7(n =34) 。
Chemical structures of various inorganic compounds are expressed by a novel scalar index of molecular distance edge connectivity (Imedc, Λ in short), developed in our laboratory, which consists of the modified MEDC descriptors based on the identical/ pseudo atom. Then quantitative structure property relationships (QSPRs) between the new I MEDC index and the formation heat ( Δ f H θ )of 33 transitional element halides or the crystal energy (lg U ) of 20 alkali metal halides were generated by multiple linear regression method (MLR) through either the modified atomic characteristic delta value δ c i defined physically and the molecular connectivity indexes ( Λ ) expressed mathematically respectively. All (QSPR) models possess high correlation: the correlation coefficients with R =0.956～0.994 respectively. Their modeling correlation capabilities are more advantageous than those obtained by some other methods reported in different literatures. It has been demonstrated that the method possesses the advantages of easy computation, clear significance and convenient application. Then cross validation with leave one out of procedure (CV) are performed and satisfactory results have been obtained with high correlation. The obtained results show that the new index is adapt to characterize chemical structure and to model the physicochemical properties of various inorganic species.
出处
《重庆师范学院学报（自然科学版）》
2002年第4期54-58,共5页
Journal of Chongqing Normal University(Natural Science Edition)
基金
国家"春晖计划"教育部启动基金(99)
原机械工业部跨世纪学术带头人专项基金(97)
重庆应用基础研究基金 (0 1)资助
