Assessing Electronic-Structure Methods for Redox Potentials of an Iron-Sulfur Cluster

Abstract

Iron-sulfur (FeS) clusters play a crucial role in biological redox processes. In this study, we evaluate the accuracy of various electronic-structure methods for calculating the redox potentials of the synthetic [Fe₄S₄(SC(CH₃)₃)₄] cluster by comparing them to experimental data. Our assessment includes a range of density functionals within broken-symmetry density functional theory (BS-DFT), the most commonly used approach for this purpose, though it has not yet been systematically compared to other methods. We also explore correlated methods such as the random phase approximation (RPA) and auxiliary-field quantum Monte Carlo (AFQMC), which are rarely applied to FeS clusters, as well as complete active space (CAS) methods combined with density matrix renormalization group (DMRG) theory and various active space constructions. Among these, BS-DFT with the hybrid functionals B3LYP, PBE0, and TPSSh showed the highest accuracy, together with RPA in combination with the approximate exchange kernel (AXK). While AFQMC demonstrated some promise, DMRG-CAS methods were significantly less accurate, likely due to inconsistencies between the active spaces within a redox pair.