Quantum Chemical Microsolvation by Automated Water Placement

Abstract

We developed a quantitative approach to quantum chemical microsolvation. Key in our methodology is the automatic placement of individual solvent molecules based on the free energy solvation thermodynamics derived from molecular dynamics (MD) simulations and grid inhomogeneous solvation theory (GIST). This protocol enabled us to rigorously define the number, position, and orientation of individual solvent molecules and to determine their interaction with the solute based on physical quantities. The generated solute–solvent clusters served as an input for subsequent quantum chemical investigations. We showcased the applicability, scope, and limitations of this computational approach for a number of small molecules, including urea, 2-aminobenzothiazole, (+)-syn-benzotriborneol, benzoic acid, and helicene. Our results show excellent agreement with the available ab initio molecular dynamics data and experimental results.

Trendbericht Theoretische Chemie 2/2: Mit dem Computer zu effizienteren Katalysatoren

“…Grid Inhomogeneous Solvation Theory (GIST) ermöglicht, die Wechselwirkungen zwischen gelöstem Molekül und Lö-sungsmittel zu quantifizieren, und erlaubt eine physikalisch basierte Antwort auf die Frage, wie viele Lösungsmittelmoleküle zu berücksichtigen sind und wo sie platziert werden müssen (Abbildung S. 61). 21) Unschuldige Gegenionen? Die Entstehung von Ionenpaaren aus katalytisch aktiver Spezies und Gegenion ist in der experimentellen Organometallchemie lange bekannt.…”

The Thermochemistry of Alkyne Insertion into a Palladacycle Outlines the Solvation Conundrum in DFT

“…The few solutions based on molecular dynamics that are emerging with the aim of minimizing user's bias in positioning explicit solvent molecules [9] are still hardly transposable to full scale reaction energy profile search. [10] Another reason for discrepancies, still not well recognized, is the dynamic nanostructuration of solvents into adaptive polar and apolar domains [11] that conventional static DFT-continuum solvation cannot address readily. However, we have already shown that in some cases static DFT-D (i.e dispersion corrected DFT) dubbed with a standard continuum screening solvation model (such as PCM [12] or COSMO [13] and COSMO-RS [14] ) can reproduce with moderate discrepancies [1a, 4a, 8b] experimental enthalpies of reactions obtained by ITC.…”

“…The few solutions based on molecular dynamics that are emerging with the aim of minimizing user's bias in positioning explicit solvent molecules [9] are still hardly transposable to full scale reaction energy profile search. [10] Another reason for discrepancies, still not well recognized, is the dynamic nanostructuration of solvents into adaptive polar and apolar domains [11] that conventional static DFT-continuum solvation cannot address readily.…”

Solvation Free Energies in Subsystem Density Functional Theory

“…16,17 In general, it is necessary to sample the configurations of the explicitly considered solvent molecules for large solvation shells because of the network of non-covalent interactions. Such sampling may rest on structures taken from molecular dynamics simulations [16][17][18] or by optimizing clusters starting from different initial configurations [19][20][21] and weighting the geometries by a Boltzmann distribution. 22 (iv) The fourth approach is based on machine learning models, which were shown to predict solvation free energies for a wide range of molecular systems 23,24 with a similar accuracy as continuum-solvation models since reference data for solvation free energies are readily available, e.g., through the FreeSolv data-base.…”

“…18 kJ/mol (± 1 kcal/mol) around the mean of ∆G solvation with 180 water molecules is indicated in the upper four rows of Fig.5to assess the convergence behaviour. This window of accuracy is reached by all combinations studied.…”