A Computationally Efficient and Accurate Method for Predicting Conductance of Single-Molecule Junctions

Despite significant progress in the field of molecular electronics over the last two decades, the quantitative prediction of metal-molecule-metal junction conductance remains a challenge. The standard computational framework combines density functional theory (DFT) with nonequilibrium Green's functions (NEGF) using low-rung exchange-correlation functionals such as PBE, which overestimate the conductances. More advanced correction methods exist but require complex workflows and high computational cost, limiting their accessibility. Here, we introduce a physically motivated approach that approximates results obtained with high-rung functionals. Our method fits the PBE-calculated transmission to a Breit-Wigner form and subsequently refines the fit parameters using molecular orbital energies and metal densities of states computed for the isolated subsystems with high-rung functionals. This approach is applicable to a broad range of molecular junctions yielding conductance values in quantitative agreement with experiments. Our approach is simple, low-cost, and accurate, making it well-suited for routine and large-scale prediction of single-molecule junction conductance.