Quantifying Uncertainties in Solvation Procedures for Modeling Aqueous Phase Reaction Mechanisms

Area of Science:

• Computational quantum chemistry
• Renewable energy catalysis
• Solvation models

Background:

• Accurate modeling of solvated reaction mechanisms is crucial for renewable energy catalysis.
• Explicit solvent interactions and counterions significantly impact reaction pathways but are computationally expensive.
• The performance of continuum solvent models for reaction mechanisms remains less understood.

Purpose of the Study:

• To evaluate the performance of different solvent models in aqueous phase charge migrations relevant to renewable energy catalysis.
• To assess the impact of explicit solvent shells and counterions on reaction energy profiles.
• To identify and mitigate errors in modeling charge-separated states.

Main Methods:

• Quantum mechanics/molecular mechanics (QM/MM) molecular dynamics simulations.
• Evaluation of various continuum solvent models: PCM, CANDLE, COSMO-RS, ESM-RISM.
• Analysis of static calculations with and without explicit solvent shells and counterions.
• Utilized hybrid functionals to address self-interaction errors.

Main Results:

• QM/MM simulations closely matched QM energy profiles for the CO2 reduction by NaBH4 reaction.
• The Na+ counterion had a negligible effect on ensemble-averaged reaction pathways.
• Static calculations with continuum solvent models showed significant variability, dependent on explicit solvent and counterion inclusion.
• Self-interaction errors in gas-phase descriptions of charge-separated states were identified as a key source of variability.
• Employing hybrid functionals and explicit solvent shells reduced these errors.

Conclusions:

• Explicit solvent shells and advanced hybrid functionals are recommended for accurately modeling charge-separated states in solvation studies.
• Careful consideration of solvation models is essential for reliable computational predictions in renewable energy catalysis.
• The study provides guidance for future computational efforts in this field.