摘要
用径向基函数神经网络(RBF-Network)构建了一个无机二元氢化物酸性强度pKa与结构参数的非线性关联模型。13个样本的pKa计算值与实验值相关系数R达到0·9998,平均偏差0·315,显著优于文献报道的结果。交互预测的结果也非常理想,Rcv达到0·9984。参数试验的结果显示,以键合氢原子数m表征的构型因素和以键长rA—H等表征的分子大小因素是影响无机二元氢化物酸性强弱的主要影响因素。
A RBF-Network was applied to study the relationships between the pKa of 13 kinds of inorganic binary hydride and the structural parameters. There are good relationship between the calculated and experimental pK, data with a fitting correlation coefficient 0.9998 and the ideal cross-validation Roy which is up to 0.9984. rA-H and m play an important role in determining the acidity of the inorganic binary hydride.
出处
《化学通报》
CAS
CSCD
北大核心
2006年第9期711-714,共4页
Chemistry
基金
湖南省高校科研资助项目(01C035)
