Accelerating computations of organometallic reaction energies through hybrid basis sets

Abstract

The quality of DFT calculations largely depends on the choice of the atomic orbital basis sets. When selecting a basis set for a project, computational chemists and users alike often seek a cost-accuracy compromise. We report a clear-cut hybrid basis set approach to significantly reduce computational times by placing the larger basis set in a reduced selection of atoms at the reaction center. Through a systematic evaluation, we identify efficient hybrid basis sets (e.g., def2-SVP/def2-TZVPP) responsible for slashing computational cost by 3–10-fold, depending on the density functional and size of the system, while maintaining the accuracy of reaction energies and barrier heights for a series of organometallic transformations (mean absolute deviation, MAD ≤ 0.1 kcal mol⁻¹).