摘要
系统研究了核磁共振碳谱与化学位移和规律 ,以及分子拓扑指数在定量 [结 ]构 [波 ]谱关系 (QSSR)中的应用 .本文基于矢量路径长度矢量 p =(P1,P2 ,P3 ,… ,Pm)与分子中原子相互作用 ,提出了一种新型分子距边矢量并发现它与烷烃13 CNMR化学位移和有良好线性相关性 ,回归方程及其统计参数为 :CSS =bν +p3 =Σmj=0 bjνj+b10 p3 =b0 ν +b1ν1+b2 ν2 +b3 ν3 +b4 ν4 +b5ν5+b6ν6+b7ν7+b8ν8+b9ν9ν +b10 P3 =- 13.6 0 11+2 2 .2 133ν1+2 8.412 1ν2 +2 5 .9416ν3 +2 6 .6 70 9ν4 +14.4976ν5+5 .72 40ν6- 5 .3830ν7- 3.2 15 2ν8- 15 .0 2 13ν9- 2 5 .70 99ν10 +12 .2 786P3 (n =6 3,R =0 .9970 ,EV =99 .6 8% ,RMS =3.734 8,F =2 418.2 ;交互校验CV为 :R =0 .9893,EV =98.83% ,RMS =7.12 6 1,F =6 6 4.0 46 ) ;结果良好 .
Systematic studies were made on carbon 13 nuclear magnetic resonance ( 13 C NMR) and its regularity of chemical shift sum (CSS) . In the present paper, a novel molecular distance edge vector (VMDE) or ν vector expressed in the form of energy was developed and found to be correlated excellently with the chemical shift sum (CSS) for carbon 13 nuclear magnetic resonance ( 13 C NMR). The multiple linear regression (MLR) equation of molecular modelling was established by the molecular distance edge vector together with the path count of three bonding segment (C C C). It gives very good results for the modelled estimation and prediction with cross validation (CV) of leave one out (LOO) procedure: CSS=bν+p\-3=Σ\+m\-\{j=0\}b\-jν\-j +b\-\{10\}p\-3=b\-0ν+b\-1ν\-1+b\-2ν\-2+b\-3ν\-3+b\-4ν\-4+b\-5ν\-5+b\-6ν\-6+b\-7ν\-7+b\-8ν\-8+b\-9ν\-9ν+b\-\{10\}P 3 =-13.6011+22.2133 ν\-1+28.4121 ν\-2+\{25.9416\} ν\-3+26.6709 ν\-4+14.4976 ν\-5+5.7240ν\-6-5.3830 ν\-7-3.2152 ν\-8-15.0213 ν\-9-\{25.7099\} ν\-\{10\}+12.2786P 3 (n=63, R=0.9970, EV=99.68%, RMS=3.7348, F=2418.2, CV: R=0.9893, EV =98.83%, RMS=7.1261, F=664.046) . A reliable correlation model has been developed by using the combination of MDE vector ( ν vector) and molecular path counts of length three (P 3) , as adjoin structural descriptors.
出处
《波谱学杂志》
CAS
CSCD
北大核心
2001年第1期51-60,共10页
Chinese Journal of Magnetic Resonance
基金
国家春晖计划项目教育部启动基金及霍英东基金资助项目
关键词
分子距边矢量
化学结构参数比
核磁共振碳谱
化学位移
烷烃
分子建模
定量构谱关系
分子拓扑
Modified molecular distance edge vector ν, strcuctural parameterization
carbon-13 nuclear magnetic resonance spectroscopy, cross validation, chemical shift sum, alkanes, molecular modelling
