采用密度泛函理论方法,在B3LYP/6-31G(d)水平下,计算了23种磺酰类化合物的分子极化率及分子骨架中各原子的Milliken电荷.提出了一种新的QSAR建模方法,并据此对其中18种化合物进行多元线性回归分析,建立了除草活性的预测模型(R=0.96,R^2=0.92,r^2adj=0.88,F=26.26,q^2=0.71,P〈0.01,SE=0.36),对剩余五种化合物进行预测,结果吻合.该模型从化合物的亲水性、分子几何特征的角度对如何提高磺酰脲类化合物的除草活性进行了分析,并对提高化合物除草活性的方法做出预测:提高苯环和嘧啶环取代基的亲水性,增加N13周围的电子云密度,为苯环接入较小的取代基团,在嘧啶环上接入较大取代基团都可提高化合物的除草活性.预测结果与3D-QSAR方法的预测结果一致.
Using the density functional theory method, the theoretical quantum chemical descriptors of twentythree sulfonylurea compounds were calculated at B3LYP/6-31G(d) theoretical level. Described by the descriptors from eighteen compounds and optimized by multiple regression, a new QSAR model was established 2 to correlate the herbicidal activity(R =0. 96, r^2adj =0. 88, F =26.26, q^2 =0. 71, p 〈0.01, SE =0. 36). To obtain new products with a higher herbicidal activity, it is suggested that it is feasible to increase the electron density of the 13th nitrogen atom and to increase the hydrophilicity of the substituents in benzene and pyrimidine. Those predictions are in good agreement with the results obtained from 3D-QSAR method. The present study may be helpful for probing the mechanism of herbicidal activities and understanding the herbicidal chemistry in-depth study in the future.