基于新型分子距边矢量ν与多元统计方法对于烷烃^(13)C NMR化学位移和(CSS)估计与预测的深入研究(英文)  

ON FURTHER INVESTIGATION INTO ESTIMATION AND PREDICTION OF CHEMICAL SHIFT SUM OF CARBON-13 NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY IN ALKANES BASED ON A NOVEL MOLECULAR DISTANCE-EDGE VECTOR (VMDE OR ν)

作　　者：袁晓燕[1] 余丹辉 刘树深[1] 曹晨忠[3] 谌其亭 李志良[1] 

机构地区：[1]重庆大学化学化工学院 [2]岳阳市卫生局汨罗药检所,岳阳411026 [3]湘潭师范学院化学系,湘潭411200 [4]长沙电力大学应用化学系,长沙410004

出　　处：《波谱学杂志》2001年第4期363-375,共13页Chinese Journal of Magnetic Resonance

基　　金：国家春晖计划项目教育部启动基金 ( 1 998);霍英东基金 ( 1 998)资助课题

摘　　要：基于本实验室提出一种新型以势能形式表达的分子距边矢量 ,深入地系统研究了核磁共振碳 -1 3谱化学位移和 (CSS)规律以及分子拓扑指数矢量在定量结构波谱关系(QSSR)中的应用 .借助多种计量化学方法包括多元线性回归、逐步多元回归、主成分回归、主筛选回归等进行分子拟模和定量相关研究 ,发现烷烃13CNMR化学位移和 (CSS)与其分子距边矢量及路径长度指数有良好线性相关性 ,回归方程及其统计参数为 :CSS=bν +cp3=∑mj=0bjνj+b11p3=b0 ν+b1ν1+b2 ν2 +b3ν3+b4ν4+b5 ν5 +b6 ν6 +b7ν7+b8ν8+b9ν9+b10 ν10 +b11p3=-1 3 .576+ 2 2 .1 79ν1+ 2 8.4 0 7ν2 + 2 5.950ν3+ 2 6.690ν4+ 1 4.4 98ν5 + 5.72 6ν6 -5.3 79ν7-3 .2 1 4ν8-1 5.0 2 1ν9-2 5.71 0ν10 + 1 2 .2 78p3　n=63 ,R =0 .997,EV =99.68% ,RMS =3 .73 48,SD =4 .1 1 8,F =773 .1 1 6,U =1 442 2 8.84 4 ,Q=864 .93 8;CV :R2 CV =0 .980 ,EV=98.83 % ,RMS=7.1 2 61 ,SDCV=7.63 4,FCV =2 2 1 .72 0 ,UCV =1 42 1 2 1 .891 ,QCV =2 971 .896.结果良好 .Based on a novel molecular distance-edge vector (VMDE,ν), expressed in the form of potential energy and developed in our laboratories, further systematic studies were made on the regularity of chemical shift sum ( CSS ) for carbon-13 nuclear magnetic resonance ( 13C NMR). With aid of several chemometrical techniques including multiple linear regression (MLR), stepwise multiple regression (SMR) and principal component regression (PCR), various molecular modelling equations were established to correlate chemical shift sum (CSS) of carbon-13 nuclear magnetic resonance ( 13C NMR) excellently to the molecular distance-edge vector (VMDE) together with the path count of three bonding segment (C-C-C) as follows, whose very good results were obtained for the modelled estimation and accurate prediction with cross validation (CV) of leave-one-out (LOO) procedure: CSS=bν+cp 3=∑mj=0b jν j+b 11p 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 9+b 10ν 10+b 11 p 3=-13.576+22.179ν 1+28.407ν 2+25.950ν 3+26.690ν 4+14.498ν 5+5.726ν 6-5.379ν 7-3.214ν 8-15.021ν 9 -25.710ν 10+12.278 p 3 n=63, R=0.997, EV=99.68%, RMS=3.7348, SD=4.118, F=773.116, U=144228.844, Q=864.938; CV: R 2 CV=0.980, EV=98.83%, RMS=7.1261, SD CV=7.634, F CV= 221.720, U CV=142121.891, Q CV=2971.896. Some reliable correlation models have been developed by using the adjoin structural descriptors through combination of the MDE vector ( ν vector) and molecular path counts of length three (p 3).

关 键 词：分子模拟 分子距边矢量ν 化学结构参数化 核磁共振碳谱 烷烃 化学位移和(CSS) 定量构谱关系 ^13C NMR 路径长度指数 多元统计 估计 预测 

分 类 号：O623.11[理学—有机化学] O621.15[理学—化学]