A quantum chemical study from a molecular perspective: ionization and electron attachment energies for species often used to fabricate single-molecule junctions

TL;DR

This study uses advanced quantum chemistry methods to accurately calculate ionization and electron attachment energies of molecules used in single-molecule junctions, highlighting the importance of basis set quality and orbital distribution in transport properties.

Contribution

It provides benchmark calculations with EOM-CCSD for key molecules, emphasizing the need for better basis sets and detailed orbital analysis in molecular transport modeling.

Findings

Significant differences from current transport approaches in $EA$ and $IP$ values.

Highlighting the necessity of diffuse basis functions for accurate $EA$ calculations.

Challenging the point-like molecular orbital assumption in electrode interactions.

Abstract

The accurate determination of the lowest electron attachment ($EA$) and ionization ($IP$) energies for molecules embedded in molecular junctions is important for correctly estimating, \emph{e.g.}, the magnitude of the currents ($I$) or the biases ($V$) where an $I-V$-curve exhibits a significant non-Ohmic behavior. Benchmark calculations for the lowest electron attachment and ionization energies of several typical molecules utilized to fabricate single-molecule junctions characterized by n-type conduction (4,4'-bipyridine, 1,4-dicyanobenzene, and 4,4'-dicyano-1,1'-biphenyl) and p-type conduction (benzenedithiol, biphenyldithiol, hexanemonothiol, and hexanedithiol] based on the EOM-CCSD (equation-of-motion coupled-cluster singles and doubles) state-of-the-art method of quantum chemistry are presented. They indicate significant differences from the results obtained within current approaches to molecular transport. The present study emphasizes that, in addition to a reliable quantum chemical method, basis sets much better than the ubiquitous double-zeta set employed for transport calculations are needed. The latter is a particularly critical issue for correctly determining $EA$'s, which is impossible without including sufficient diffuse basis functions. The spatial distribution of the dominant molecular orbitals (MO's) is another important issue, on which the present study draws attention, because it sensitively affects the MO-energy shifts $\Phi$ due to image charges formed in electrodes. The present results cannot substantiate the common assumption of a point-like MO midway between electrodes, which substantially affects the actual $\Phi$-values.