Redox Potential Prediction

Redox potentials measure how easy or hard it is to add or remove electrons from an atom. To remember the difference between reduction and oxidation, you can use “OIL RIG”: Oxidation Is Loss (of electrons) and Reduction Is Gain (of electrons).

Rowan's redox potential prediction workflow predicts the reduction and/or oxidation potential of a molecule by computing the energy difference between the reduced/oxidized and starting state of the molecule.

How It Works

Rowan's redox potential workflow first runs an optimization on the input molecule, followed by a single-point energy at a higher level of theory. Then, depending on whether reduction or oxidation (or both) were requested, an electron is added or removed from the input structure. (Typically this results in an open-shell species with multiplicity 2.) An optimization and single-point energy are then run on the optimized or reduced molecule.

The exact levels of theory used for optimization and single-point-energy calculation depend on the mode selected by the end user.

The difference in energy between the neutral and reduced/oxidized molecule is converted from Hartree to electron-volts, and a constant correction of 4.422 V is subtracted to convert the values to the saturated calomel electrode (SCE) reference. (For GFN2-xTB single-point energies, an additional correction of 4.846 V is applied, as reported by Neugebauer et al.) To convert data to other reference electrodes, conversion factors from Pavlishchuk and Addison are employed.

Accuracy

On the OROP benchmark set, Rowan's redox potential prediction workflow (in rapid mode) displays a mean absolute error of 0.32 V (excluding a few values that don't start from a neutral species or aren't in acetonitrile).

Modes

Rowan's redox potential prediction workflow can be run in four modes: reckless, rapid, careful, or meticulous. Here's what selecting each mode tunes:

(ωB97X-3c was recently highlighted as a particularly effective DFT method for redox potential prediction.)