为预测60种烯烃类单体（M1）与苯乙烯（M2）的自由基共聚合竞聚率lgr1S值,采用密度泛函理论（DFT）B3LYP方法在6-31G（d）基组水平上对烯烃类单体（C1H2=C2XY）进行了计算。用于构建支持向量机（SVM）模型的最佳参数子集包括：原子R3的Mulliken电荷qMR3,C1的Mulliken电荷QMC-1（H原子电荷全部合并到与之相连的重原子上）,参数QMC-1与qMC-2之比RQq（=QMC-1/qMC-2）,最低未占分子轨道（LUMO）能级（ELUMO）和最高占据分子轨道（HOMO）能级（EHOMO）之差ΔEg。最佳SVM模型为高斯径向基核函数（C=1000,ε=0.0001及γ=0.2）。该模型训练集、验证集及测试集的均方根（rms）误差分别为0.043,0.157及0.192,与现有竞聚率lgr模型相比,本文所得SVM模型具有更好的统计品质。
To predict monomer reactivity ratios（lgr1S） in radical copolymerization between 60 vinyl monomers（M1） and styrene（M2）,density functional theory（DFT） calculations were carried out for monomers M1（C1H2=C2XY）,at the B3LYP level of theory with 6-31G（d） basis set.The optimum subset of descriptors selected for the support vector machine（SVM） model comprises of Mulliken charges of R3（qMR3）,Mulliken charges of C1 with hydrogens summed into heavy atoms（QMC-1）,the ratio of parameters QMC-1 and qMR-2（RQq） and the LUMO and HOMO orbital energy difference（ΔEg）.The optimal SVM model was obtained with the Gaussian radical basis kernel（C=1000,ε =0.0001 and γ = 0.2）.The root-mean-square（rms） errors for training set,validation set and test set are 0.043,0.157 and 0.192,respectively.Comparison to existing models,the SVM model shows better statistical characteristics.
参考文献
| [1] | XINLIANG YU;BING YI;XUEYE WANG .Prediction of Refractive Index of Vinyl Polymers by Using Density Functional Theory[J].Journal of Computational Chemistry: Organic, Inorganic, Physical, Biological,2007(14):2336-2341. |
| [2] | 刘万强,王学业,李新芳,龙清平,文小红,李建军.聚甲基丙烯酸酯类玻璃化温度的量子化学-神经网络方法研究[J].高分子材料科学与工程,2006(01):170-173. |
| [3] | Yu XL;Yi B;Wang XY .Quantitative structure-property relationships for the reactivity parameters of acrylate monomers[J].European Polymer Journal,2008(12):3997-4001. |
| [4] | Xinliang Yu;Wenhao Yu;Bing YI .PREDICTION OF MONOMER REACTIVITY RATIOS IN RADICAL COPOLYMERIZATION OF VINYL MONOMERS[J].Collection of Czechoslovak Chemical Communications,2009(9):1279-1294. |
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%