First-principles many-body calculations of electronic conduction in thiol- and amine-linked molecules

TL;DR

This study uses advanced many-body calculations to accurately predict electronic conductance in benzene molecules linked to gold electrodes, revealing significant differences from standard DFT predictions and aligning well with experimental data.

Contribution

It demonstrates the importance of self-consistent GW calculations for realistic conductance predictions in molecular electronics, improving upon traditional DFT methods.

Findings

GW calculations agree with experiments for benzene conductance

Standard DFT overestimates conductance by a factor of five

Complex gold/thiolate structures influence conductance significantly

Abstract

The electronic conductance of a benzene molecule connected to gold electrodes via thiol, thiolate, and amino anchoring groups is calculated using nonequilibrium Green functions in combination with the fully selfconsistent GW approximation. The calculated conductance of benzenedithiol and benzenediamine is five times lower than predicted by standard density functional theory (DFT) in very good agreement with experiments. In contrast, the widely studied benzenedithiolate structure is found to have a significantly higher conductance due to the unsaturated sulfur bonds. These findings suggest that more complex gold/thiolate structures where the thiolate anchors are chemically passivated by Au adatoms are responsible for the measured conductance. Analysis of the energy level alignment obtained with DFT, Hartree-Fock and GW reveals the importance of self-interaction corrections (exchange) on the molecule and dynamical screening at the metal-molecule interface. The main effect of the GW self-energy is to renormalize the level positions, however, its influence on the shape of molecular resonances also affects the conductance. Non-selfconsistent G0W0 calculations, starting from either DFT or Hartree-Fock, yield conductance values within 50% of the selfconsistent GW results.