Quantifying Image Charge Effects in Molecular Tunnel Junctions Based on Self-Assembled Monolayers of Substituted Oligophenylene Ethynylene Dithiols

Abstract

A number of factors contribute to orbital energy alignment with respect to the Fermi level in molecular tunnel junctions. Here, we report a combined experimental and theoretical effort to quantify the effect of metal image potentials on the highest occupied molecular orbital to Fermi level offset, εh, for molecular junctions based on self-assembled monolayers (SAMs) of oligophenylene ethynylene dithiols (OPX) on Au. Our experimental approach involves the use of both transport and photoelectron spectroscopy to extract the offsets, εh trans and εh UPS, respectively. We take the difference in these quantities to be the image potential energy eV image. In the theoretical approach, we use density functional theory (DFT) to calculate directly eV image between positive charge on an OPX molecule and the negative image charge in the Au. Both approaches yield eV image ∼ −0.1 eV per metal contact, meaning that the total image potential energy is ∼−0.2 eV for an assembled junction with two Au contacts. Thus, we find that the total image potential energy is 25–30% of the total offset εh, which means that image charge effects are significant in OPX junctions. Our methods should be generally applicable to understanding image charge effects as a function of molecular size, for example, in a variety of SAM-based junctions.

Quantifying Molecular Structure–Conductance Relationship in Nonlinear π-Conjugated versus Linear π-Conjugated Wire for Application in Molecular Electronics

“…The result was expected, but it is important that quantitative measures for both ε h and Γ are easily obtained with the analytical SLM model and that the relative contributions of ε h and Γ can be resolved. We note here that the nonlinear conjugation could cause a quantum interference effect on the transport, which led to a decrease in conductance. ,,− In our MJs based on CP-AFM, the transport properties are however controlled by the interfacial state, ,,− metal–S bonding; as a result, we could exclude the effect of quantum interference on the transport properties. Quantum chemical calculations are necessary for further insight into the great difference in the electronic coupling, Γ, and quantum interference effect.…”

“…During the measurement, the gold substrate was grounded, while voltage (V) was swept to the AFM tip. The low-bias resistance ( R ) or conductance (1/ R ) was acquired in ±0.3 V, while the I – V characteristic at the high window was recorded at ±1.0 V for the junctions. ,,,, During the measurement, the gold-coated AFM tips are usually tested on the SAMs formed by 1-decanethiol.…”

“…It was first theoretically proposed by Ratner and Aviram in 1974 and then recently demonstrated as molecular rectifiers, − conductance switches, − and light emission . One of the central objectives of molecular electronics is to understand chemical structures and conductance relationship properties ,− in metal–molecule–metal junctions, Scheme . For direct single-step tunneling junctions, the analytical single-level model (SLM) derived from the Landauer picture has proven to be a convenient method to quantify key transport parameters from current–voltage ( I – V ) characteristics acquired from transport measurement. ,,,,− The transport metrics consist of the HOMO (or LUMO) energy offset, ε h (or ε l ), and the electronic coupling Γ, as illustrated in Scheme B.…”

“…One of the central objectives of molecular electronics is to understand chemical structures and conductance relationship properties ,− in metal–molecule–metal junctions, Scheme . For direct single-step tunneling junctions, the analytical single-level model (SLM) derived from the Landauer picture has proven to be a convenient method to quantify key transport parameters from current–voltage ( I – V ) characteristics acquired from transport measurement. ,,,,− The transport metrics consist of the HOMO (or LUMO) energy offset, ε h (or ε l ), and the electronic coupling Γ, as illustrated in Scheme B. Recent work by Bâldea et al and others has demonstrated that Γ and ε h (ε l ) are easily quantified from experimental I – V data using the SLM, and the resulting I – V curve simulated by the SLM matches well the experimental I – V characteristics. ,,,− The utility of the SLM in a tunneling junction was also verified by the independent determination of ε h experimentally determined by ultraviolet photoelectron spectroscopy (UPS) for oligophenylene thiol, oligophenylene dimethanethiol, and alkanethiol systems and by its prediction of transport behavior for single-step direct tunneling that matches experimental I – V results. ,,, …”

“…We briefly summarize the relevant analytical SLM formulas here. ,,,, The SLM stimulates the I – V characteristics for a single-step tunneling junction using eq . I = G V ε normalh 2 ε normalh 2 − ( e V / 2 ) 2 where the γ factor is the orbital shift (Stark) factor and zero-bias conductance G = 1/ R of junction can be expressed as below: G = N G 0 normalΓ 2 ε normalh 2 where Γ = Γ s Γ t is the average interface coupling, and Γ s and Γ t are determined by the molecular coupling to the substrate (s) and the tip (t), for junctions formed by CP-AFM. G 0 is the quantum conductance.…”