(64b) Structure-Conductance Relationships in Atomic-Scale Junctions: Insights from Reactive Molecular Simulation

Molecular electronics, i.e., self-assembled molecular-based electronic systems, have been proposed as a possible alternative to the traditional Si-based components used in the nano electronics industry. While several promising examples have been created, much is still not well understood in these systems, often as a result of the disconnect between experiment and computation. Experimentally, the mechanically controllable break-junction technique has served as a platform for studying the electron transport properties of low-dimensional metallic structures and individual molecules. Computational studies of these junctions have provided a wealth of atomic-level insight, enabling the identification of important structural mechanisms that influence the transport properties. However, the utility of these computational studies relies not only on the ability to accurately model the atomic interactions, but also on the ability to appropriately capture the non-ideality present in the experiments. Much of this non-ideality arises from thermal motion and the mechanical deformation process by which the junctions are formed, neither of which are typically considered in first-principles-based calculations of the molecular junctions.

In order to model the formation of junctions under conditions more representative of experiment, we use reactive molecular simulations, parameterized from first principles calculations, to model the spontaneous formation of single-molecule junctions of benzene-1,4-dithiolate (BDT) bridged between Au tips[1-3]. These simulations are used as input to high-fidelity first-principles conductance calculations to establish structure-conductance relationships of the more realistic molecular junctions[2,3]. This combined approach allows for significantly reduced computational cost, as compared to first principles calculations alone, enabling the accurate study of large system sizes, long time-scales, and many independent simulations, needed to fully understand the behavior of these junctions. Results detailing the role that structure, thermal motion, and stochasticity play on the conductance of break-junctions is presented, with an emphasis on the structural origin of "anomalous" conductance increases seen in experiment for Au-BDT-Au junctions[4].

[1] French WR, Iacovella CR, Cummings PT, ACS Nano, 6, pp 2779–2789, 2012
[2] French WR, Iacovella CR, Rungger I, Souza A, Sanvito S, Cummings PT, Nanoscale, 5 pp 3654-3659, 2013
[3] French WR, Iacovella CR, Rungger I, Souza A, Sanvito S, Cummings PT, J. Phys. Chem. Lett., 4 pp 887–891, 2013
[4] Bruot C, Hihath J, Tao, NJ, Nature Nano. 7, 35-40, 2012