Instituto de Química , Universidade Estadual de Campinas , Barão Geraldo , 13083-970 Campinas - São Paulo , Brazil.
Instituto de Ciências Exatas e Tecnologia , Universidade Federal do Amazonas , Campus de Itacoatiara , 69100-021 Itacoatiara - Amazonas , Brazil.
J Phys Chem A. 2019 Sep 26;123(38):8314-8320. doi: 10.1021/acs.jpca.9b05380. Epub 2019 Sep 13.
For 22 monoprotic acids, p values were calculated using the G4CEP composite theory. The solvation effect was included considering the continuous SMD solvation model; SMD and one explicit water molecule; and SMD, one water molecule, and linear correction with respect to the experimental p values. The three tests provided mean absolute errors equal to 0.83, 0.51, and 0.30 p units, respectively, indicating excellent performance of the G4CEP method. Comparison with density functional theory at the B3LYP and BMK levels showed that these results are quickly obtained but with a significant error. The best performance of the functionals was obtained from the combination of SMD, one explicit water molecule, linear regression correction, and basis set including diffuse functions. However, the dispersion of the results with DFT can lead to deviations of up to two p units, whereas for G4CEP the largest deviations seldom exceed one p unit.
对于 22 种一元酸,使用 G4CEP 组合理论计算了 p 值。考虑连续 SMD 溶剂化模型、SMD 和一个显式水分子、以及 SMD、一个水分子和与实验 p 值的线性校正,包括溶剂化效应。这三种测试的平均绝对误差分别为 0.83、0.51 和 0.30 p 单位,表明 G4CEP 方法具有出色的性能。与 B3LYP 和 BMK 水平的密度泛函理论的比较表明,这些结果虽然很快得到,但存在显著误差。最好的泛函性能来自 SMD、一个显式水分子、线性回归校正以及包含弥散函数的基组的组合。然而,DFT 的结果分散可能导致偏差高达两个 p 单位,而对于 G4CEP,最大偏差很少超过一个 p 单位。