Achieving Predictive Description of Molecular Conductance by Using a Range-Separated Hybrid Functional.

The conductance of molecular bridges tends to be overestimated by computational studies in comparison to measured values. While this well-established trend may be related to difficulties for achieving robust bridges, the employed computational scheme can also contribute to this tendency. In particular, caveats of the traditional functionals employed in first-principles-based calculations can lead to discrepancies reflected in exaggerated conductance. Here, we show that by employing a range-separated hybrid functional the calculated values are within the same order as the measured conductance for all four considered cases. On the other hand, with B3LYP, which is a widely used functional, the calculated values greatly overestimate the conductance (by about 1-2 orders of magnitude). The improved description of the conductance with a RSH functional builds on achieving a physically meaningful treatment of the quasi particles associated with the frontier orbitals.