Refined SMD Parameters for Bromine and Iodine Accurately Model Halogen‐Bonding Interactions in Solution

Abstract

Motivated by the need to calculate liquid-phase free energies of species and equilibria involving halogen bonding, recent experimental data were used to optimize new Coulomb radii for Br and I for use in the SMD universal solvation model for calculating free energies of solvation. The use of the SMD model with these parameters for Br and I and the SMD values of all other parameters is called SMD18. After parametrization, the SMD18 model was tested for data not used in the parametrization. These data are standard-state free energies (equivalent to equilibrium constants) for 18 ionic equilibria involving Cl, Br, and I halogen bonding in acetonitrile, and the agreement of theory and experiment is satisfactory. The SMD18 model is then used to compare hydrogen bonding to halogen bonding and to reassess the interpretation of recent experiments.

Is There a Single Ideal Parameter for Halogen‐Bonding‐Based Lewis Acidity?

“…Triple‐ζ basis sets were mostly applied with one or two polarisation functions. In addition, the halogens chlorine to iodine were augmented with a diffuse function derived by Rappoport and Furche (this basis set is denoted as def2‐TZVP(D)) based on the well‐known issues of DFT with anions and our own experience of solution‐phase calculations …”

“…[33] Triple-z basis sets were mostly applied with one or two polarisationf unctions. In addition, the halogensc hlorine to iodine werea ugmented with ad iffuse function derived by Rappoport and Furche [34] (this basis set is denoted as def2-TZVP(D) [22,35] )b ased on the well-known issues of DFT with anionsa nd our own experience of solution-phase calculations. [22,36] In our previous study, [18d] DCCSD(T) def2-QZVPPD data were obtained, with two-point complete basis set limit extrapolated MP2 energies, [37] for the complexes of halomethanes 2-4 ( Figure 2) with chloride and ammonia.…”

“…In addition, the halogensc hlorine to iodine werea ugmented with ad iffuse function derived by Rappoport and Furche [34] (this basis set is denoted as def2-TZVP(D) [22,35] )b ased on the well-known issues of DFT with anionsa nd our own experience of solution-phase calculations. [22,36] In our previous study, [18d] DCCSD(T) def2-QZVPPD data were obtained, with two-point complete basis set limit extrapolated MP2 energies, [37] for the complexes of halomethanes 2-4 ( Figure 2) with chloride and ammonia. Herein, we also included difluoroiodomethane( 1)a nd applied the extrapolation method of Feller as given by Vasilyev [38] on the aug-cc-pVnZ (n = 2-4) basis set and the procedure of Halkier et al for the extrapolation of the MP2 energies on the def2-nZVPPD and aug-cc-pVnZ-(PP) (n = 3-4) basis sets.…”

“…To obtain ab etter indicator for the prediction of Lewis acidity,l inear combinations of individual parameters were then tested in ac ompletely empiricala pproach. The linear combination of the s-hole depth with the energy of the s*o rbital, suitably optimised against at est set of halogen-bond donors (1-18), was found to be superior in the prediction of halogenbondings trength,e specially for the stronger complexes.T his parameter W s* provided as trong correlation when validated against as econd set of diversen eutral halogen-bond donors (19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35). Cationic halogen-based Lewis acids were described with ahigh accuracyfor chloride association and avery low accuracy fora mmonia by this parameter,b ut ar e-optimisation led to the derivative W s* ',w hich once againp rovided as trong correlation with the gas-phase binding energies.…”

“…Since solvente ffects would very likely complicate the analysis dramatically,w ew ill focus solely on gas-phase interactions. [22] The reliability and quality of the benchmarking results will obviously strongly depend on the quality of the interaction energies, against which all parameters will be tested. Thus, an important first step will be the generation of high-qualityc oupled-cluster energies and av alidation of DFT energies against this standard.…”