Efficient Computational Methods for Accurately Predicting Reduction Potentials of Organic Molecules

Abstract

A simple computational approach for predicting ground-state reduction potentials based upon gas phase geometry optimizations at a moderate level of density functional theory followed by single-point energy calculations at higher levels of theory in the gas phase or with polarizable continuum solvent models is described. Energies of the gas phase optimized geometries of the S0 and one-electron-reduced D0 states of 35 planar aromatic organic molecules spanning three distinct families of organic photooxidants are computed in the gas phase as well as well in implicit solvent with IPCM and CPCM solvent models. Correlation of the D0 - S0 energy difference (essentially an electron affinity) with experimental reduction potentials from the literature (in acetonitrile vs SCE) within a single family, or across families when solvent models are used, yield correlations with r(2) values in excess of 0.97 and residuals of about 100 mV or less, without resorting to computationally expensive vibrational calculations or thermodynamic cycles.

Electron spin changes during general anesthesia in Drosophila

“…The calculated R value seems to give a good heuristic for anesthetic action and correctly predicts the high anesthetic potencies of Xe, barbital, etomidate, alfaxalone, and cyclopropane and the nonanesthetic nature of fluothyl and DMFPC. SF 6 is clearly anomalous, maybe because DFT methods, although generally accurate when calculating HOMO energies (87), are known to miscalculate SF 6 by several electronvolts (88). The other results are in general agreement with the Meyer-Overton relationship and its best-documented exceptions.…”

Building and testing correlations for the estimation of one‐electron reduction potentials of a diverse set of organic molecules

“…Given the computational cost savings of our present method only requiring a single closed‐shell calculation, the added cost of a single CPCM calculation can generally easily be absorbed and is advisable wherever feasible. It was also observed that the root mean square deviation for individual families was in general just slightly better than for the global correlations (Supporting Information), indicating that when a correlation is available for an exact family of molecules, the correlation is particularly good for that family; however, the almost equally good global correlations clearly indicate that good predictive ability should be able to be obtained even for molecules not necessarily well represented structurally in the initial correlation, just as the Gillmore group noted for their prior method …”

“…The correlations in Tables and were built with the experimental reduction potentials on the x ‐axis, because this was the control variable as the correlations were built and also to be consistent with our previous work the energy difference between the initial closed‐shell singlet and one‐electron reduced doublet states was used as an analog to an electron affinity in a dielectric continuum rather than the gas phase.…”

“…These open‐shell calculations can be computationally expensive, especially for large molecules. In previous work, the Gillmore group has explored the accuracy of this type of calculation using a simplified version of the Born–Haber cycle, where the geometry optimizations of the structures (closed‐shell singlets and one‐electron‐reduced doublets) were performed in the gas phase, followed by a single‐point energy calculation of each structure in the presence of a continuum solvent model . This simplified method did not specifically calculate a reduction potential but something more akin to an electron affinity (albeit in solution).…”

“…The advantage of this approach is that, in principle, only a single geometry optimization of the molecule of interest in the initial closed‐shell singlet (prior to single electron oxidation or reduction) is required to predict both the experimental first one‐electron reduction and oxidation potentials, and all calculations are on closed‐shell spin‐paired species. In our recent preliminary report, this strategy was tested with 51 polycyclic aromatic hydrocarbons (PAHs) with known oxidation and/or reduction potentials, measured in acetonitrile, and found to be both more accurate and more computationally efficient than the Gillmore group's methods.…”