Comparison of Implicit and Explicit Solvent Models for the Calculation of Solvation Free Energy in Organic Solvents

Abstract

2015, 55, 1192-1201) and experimental data. The comparison indicates that both PB and GB give poor agreement with explicit solvent calculations and even worse agreement with experiments (root-mean-square deviation ≈ 15 kJ/mol). The main problem seems to be the prediction of the apolar contribution, which should include the solvent entropy. The quantum mechanical-based SMD model gives significantly better agreement with experimental data than do PB or GB, but it is not as good as explicit solvent calculation results. The dielectric constant ε of the solvent is found to be a powerful predictor for the polar contribution to the free energy in implicit models; however, the Onsager relation may not hold for realistic solvent, as suggested by explicit solvent and SMD calculations. From the comparison, we also find that with an optimization of the apolar contribution, the PB model gives slightly better agreement with experiments than the SMD model, whereas the correlation between the optimized GB models and experiments remains poor. Further optimization of the apolar contribution is needed for GB models to be able to treat solvents other than water.

Statistical efficiency of methods for computing free energy of hydration

“…[16][17][18] Faster simulation methods using implicit solvent, such as the generalized Born (GB) 19 and Poisson-Boltzmann (PB) 20,21 methods, are less accurate. [22][23][24][25] Gibbs free energies of solvation in other solvents than water are effectively uncorrelated to experimental data with correlation coefficients R from 0.25 (GB) to 0.5 (PB) compared to 0.85 for explicit models, 25 suggesting that implicit solvent models should be used with caution. In two studies, the reference interaction site models (RISM [26][27][28] and 3D-RISM [29][30][31] ) have been used for HFE calculations of drug-like molecules.…”

Solvent effect on stabilization energy: An approach based on density functional reactivity theory

“…It should be mentioned in this context that the IEF‐PCM solvation model, as implemented in Gaussian 09, considers only the electrostatic interaction. Hence, the desolvation energy values of the individual species as well as of the adduct (AB) generated by this model depend on the electrostatic part of the interaction, and so correlate well with the Onsager polarity function (which is also based on electrostatic interaction only). As SMD solvation model takes care of nonelectrostatic interactions also (apart from electrostatic ones), through various empirical parameters), the values of desolvation energies (and so, ) generated through this model correlate poorly with Onsager polarity function.…”

“…It should be mentioned in this context that the IEF-PCM solvation model, as implemented in Gaussian 09 [26] , considers only the electrostatic interaction. Hence, the desolvation energy values of the individual species as well as of the adduct (AB) generated by this model depend on the electrostatic part of the interaction, and so correlate well with the Onsager polarity function (which is also based on electrostatic interaction only [39] ). As SMD solvation model takes care of nonelectrostatic interactions also (apart from electrostatic ones), through various empirical param- Moreover, the generated ΔE solvent SE AB ð Þ values for the adduct formation process between (azidomethyl)benzene and methylpropiolate at M062X/6-31G(d,p) level of theory are coming to be positive, apparently giving an impression that the adduct formation is a thermodynamically unfavorable process (which, in reality, is not true).…”

“…When (Tables 1,2, 5,6) are plotted against the Onsager function, (ε − 1)/(2ε + 1), reasonably reliable linear fitting is obtained (as evidenced from high adjusted R 2 values, Figures ). In all these plots values becoming numerically less negative with the increase in the value of (ε − 1)/(2ε + 1).…”

“…When ΔE solvent SE AB ð Þ (Tables 1,2, 5,6) are plotted against the Onsager function, (ε − 1)/(2ε + 1) [38,39] , reasonably reliable linear fitting is obtained (as evidenced from high adjusted R 2 values, Figures 3-6). In all these plots ΔE solvent…”