摘要
应用B3LYP/6-311G**水平计算得到的量子化学参数,在实验数据的基础上,建立了部分苯酚衍生物生物降解性能(二氧化碳生成量(PCD))的定量结构-生物降解性相关模型(2D-QSBR),该模型包括分子平均极化率(α)和熵(SΘ)两个参数,其中α对PCD的影响更为显著,模型的相关系数R2=0.933,交叉验证相关系数q2=0.894.该结果优于midix,6-31G*计算水平下得到的2D模型,此外,应用基于分子模拟技术的比较分子相似性指数(CoMSIA)方法,建立了3D-QSBR模型,模型的R2=0.964,q2=0.716,化合物的氢键供受体特性对降解能力的影响较为显著,其次是化合物的疏水性.结果表明,2D和3D-QSBR模型都具有良好的稳定性和预测能力,可以预测同类化合物的PCD值.
Based on the experimental data of production of carbon dioxide （PCD） of phenol derivatives, a 2D-QSBR model was estab- lished between PCD and quantum chemistry parameters on phenol derivatives which were calculated at B3LYP/6-311G＊＊ level. The 2D-QSBR model contains two variables： molecular average polarizability （a） and standard entropy （S~）. a has greater effect on PCD than S~＇. The correlation coefficients （R2） and cross-validation correlation coefficients （q2） of the 2D-QSBR model at B3LYP/6-311G＊＊ level are 0.933 and 0.894, respectively, showing the priority of this model over models based on midix and 6-31G＊ level. Correspondingly, a 3D-QSBR model was also established using comparative molecular similarity index analysis （CoMSIA）. The R2 and q2 of CoMSIA model are 0.964 and 0.716, respectively. The 3D contour maps suggested that the H-bonding donor and H-bonding acceptor factors have greater impact on PCD, and hydrophobic factors also have significant effect on PCD. The result indicates that the 2D-QSBR and 3D-QSBR models have strong stability and good predictability.
出处
《科学通报》
EI
CAS
CSCD
北大核心
2010年第16期1600-1607,共8页
Chinese Science Bulletin
基金
国家自然科学基金资助项目(20977046
20737001)
