Models of Electrodes and Contacts in Molecular Electronics

San-Huang Ke,Harold U. Baranger,Weitao Yang

DOI: https://doi.org/10.1063/1.1993558

2006-09-08

Abstract:Bridging the difference in atomic structure between experiments and theoretical calculations and exploring quantum confinement effects in thin electrodes (leads) are both important issues in molecular electronics. To address these issues, we report here, by using Au-benzenedithiol-Au as a model system, systematic investigations of different models for the leads and the lead-molecule contacts: leads with different cross-sections, leads consisting of infinite surfaces, and surface leads with a local nanowire or atomic chain of different lengths. The method adopted is a non-equilibrium Green function approach combined with density functional theory calculations for the electronic structure and transport, in which the leads and molecule are treated on the same footing. It is shown that leads with a small cross-section will lead to large oscillations in the transmission function, T(E), which depend significantly on the lead structure (orientation) because of quantum waveguide effects. This oscillation slowly decays as the lead width increases, with the average approaching the limit given by infinite surface leads. Local nanowire structures around the contacts induce moderate fluctuations in T(E), while a Au atomic chain (including a single Au apex atom) at each contact leads to a significant conductance resonance.

Mesoscale and Nanoscale Physics,Materials Science

What problem does this paper attempt to address?

This paper attempts to solve two key problems in molecular electronics:

1. **Atomic structure differences between experimental and theoretical calculations**: In molecular electronics, precisely controlling the atomic structure of a single - molecule device (including at least one molecule and two electrodes as current leads) is a major challenge. Experimentally, for the lead - molecule - lead (LML) system constructed by the break - junction technique, detailed atomic structure information cannot be obtained, so its influence on the device transport properties cannot be clearly determined.

2. **Quantum confinement effect**: Studying the quantum confinement effect in nanoscale leads (especially leads with a finite cross - section) is crucial for understanding relevant experimental phenomena and making a reasonable comparison between theoretical calculations and experimental results. Specifically, the paper explores the influence of different lead models (such as leads with different cross - sections, infinite - surface leads, and surface leads with local nanowires or atomic chains) on the electron transport properties.

### Research methods and findings

To explore these problems, the authors adopted the method of non - equilibrium Green's function (NEGF) combined with density functional theory (DFT) to perform first - principles calculations on the Au - benzenedithiol - Au system. The main findings are as follows:

1. **Waveguide effect of small - cross - section leads**: Small - cross - section leads will cause significant oscillations in the electron transmission function \(T(E)\), and these oscillations depend on the structure of the leads (such as different orientations). As the lead width increases, these oscillations gradually weaken, and the average value approaches the result of the infinite - surface lead, at which time the influence of different lead structures is significantly reduced.

2. **Influence of local nanowire structure**: The local nanowire structure introduced at the contact between the lead and the molecule will cause a moderate fluctuation in \(T(E)\), but will not change its main characteristics unless the nanowire is too long.

3. **Influence of single - atom - chain connection**: Introducing a gold - atom chain (including a single - vertex gold atom) at each contact point will cause obvious conductance resonances, especially a large resonance peak near the Fermi energy.

4. **Differences between theory and experiment**: Although the introduction of additional gold atoms can increase the conductance value predicted by theory, compared with the experimental results, the theoretical prediction is still one to two orders of magnitude higher. This indicates that there are still huge differences between theory and experiment in the field of molecular electronics that need to be urgently solved.

### Summary

By studying different lead and contact structure models, the paper reveals the quantum confinement effect in small - cross - section leads and its influence on electron transport properties, and attempts to bridge the gap between experiment and theory. However, the consistency problem between theory and experiment is still an important direction for future research.