Benchmarking Semiempirical Methods To Compute Electrochemical Formal Potentials.

Area of Science:

• Computational chemistry
• Electrochemistry
• Materials science

Background:

• Accurate prediction of electrochemical redox potentials is vital for advancing energy technologies.
• Semiempirical models offer a computationally efficient alternative to traditional methods like density functional theory (DFT).

Purpose of the Study:

• To benchmark the performance of various semiempirical models in calculating reduction potentials for organic molecules.
• To compare two distinct computational approaches: isolated molecule energy differences and electrode-molecule complex interactions.
• To evaluate the accuracy of modern semiempirical models against older ones and DFT.

Main Methods:

• Benchmarking PM7 and PM6 semiempirical models against DFT for reduction potential calculations.
• Comparing two methods: S0/D0 energy differences for isolated molecules and orbital energy shifts in Ag20-molecule complexes.
• Assessing model robustness using varied geometries and implicit solvent models.

Main Results:

• PM6 and PM7 demonstrate good accuracy in predicting formal potentials, comparable to DFT, with robust results across different settings.
• Both PM6 and PM7 show significant improvement over older semiempirical methods (MNDO, AM1, PM3, INDO/S).
• A recently developed INDO parameter set more accurately models the Ag20 electrode surface than PM7, highlighting limitations in current surface modeling.

Conclusions:

• PM6 and PM7 are reliable for calculating solution-phase redox potentials of organic molecules.
• Existing semiempirical models require further refinement for accurate simulation of molecules interacting with electrode surfaces.
• The development of specialized parameters is necessary for improving the modeling of surface-molecule electronic interactions.