Molecular Electronics


Molecular electronics aims to generate devices in which the electrical characteristics are determined by the chemical and supramolecular properties of the molecules. By far, most technologies are based on inorganic semiconductors and metals, but organic based electronics are potentially much cheaper and enable technologies that are difficult to achieve with conventional technologies, e.g., flexible, stretchable, and transparent electronics. To exploit the potential of organic- and molecular electronics, the fundamentals of charge transport across organic materials need to be well-understood. This goal is difficult to achieve as uncertainties in the structure of each component of the device, i.e., the electrodes, the interfaces of the electrodes with the organic part, and the organic part itself, caused during the fabrication process can be a source of artifacts hampering the interpretation of the data.
The mechanisms of charge transport across metals and semi-conductors are reasonably well understood, but not across insulators – (bio) molecules, or self-assembled monolayers (SAMs). We study the mechanisms of charge transport across junctions of the form electrode-SAM-electrode fabricated using the “EGaIn technique” as function of the chemical structure of the molecules, the supramolecular structure of the SAMs, the properties of the electrode materials, and the molecule-electrode interface energetics. This knowledge is then used to develop/improve molecular diodes. Recently, we have also begun to incorporate biomolecules in our junctions and to study plasmon excitation.

Fig 1a. Schematic of the junctions with SAMs of HSCnFc, n = 10, 11.
Fig 1b. Energy level diagram of the junctions of AgTS–SCnFc//Ga2O3/EGaIn with n = 6–15.

The Topography of the Bottom Electrode

The most commonly used approach to model charge transport across these SAM-based tunneling junctions is with a simplified form of the Simmons equation (J = J0e-βd), where β (nC-1 or Å-1) is the tunneling decay constant, d (nC or Å) is the width of tunneling barrier, and J0 (A/cm2) is a constant that depends on the system and includes, for instance, contact resistance. A controversy in molecular electronics is that across test-beds different values of β and J0 have reported for a given series of molecular structures. We aim to unravel the factors that affect the values of β and J0. For example, we found that the topography of the bottom-electrode plays a major role and showed that values of β as low as 0.5 nC-1 for junctions of SAMs of n-alkanethiolates formed on rough bottom-electrodes, while the value of β was 1.0 n-1 (which is the consensus value) for those junctions with smooth bottom-electrodes. For all types of junctions, the yield in working devices was high and independent of the value of β (see Angew. Chem. Int. Ed. 2014, 53, 3377-3381) and, counterintuitively, the yield is not a good indicator of junction quality. Likewise, the performance of molecular diodes also depends heavily on the quality of the bottom-electrode (see J. Am. Chem. Soc. 2014, 136, 6554-6557).

Fig 1: Dependence of tunneling decay coefficient (β) on surface roughness expressed in terms of bearing volume.

Equivalent Circuits determined by impedance spectroscopy

Usually the electrical characteristics of molecular tunnel junctions are determined by DC methods. It is difficult to disentangle the contribution of each component of the junction, e.g., the molecule−electrode contacts, protective layer (if present), or the SAM, to the electrical characteristics of the junctions using the DC methods. We showed that frequency dependent AC measurements (impedance spectroscopy) make it possible to separate the contribution of each component in the molecular junction (see J. Am. Chem. Soc. 2014136, 11134-11144) and on top of that also allows to determine the capacitance of the SAMs inside the junctions. We found that the contact resistance is dominated by the SAM−top contact resistance (and not by the conductive layer of GaOx) and is independent of molecular chain length while the oxide layer becomes important and dominates the electrical characteristics when its thickness was intentionally increased electrochemically to roughly 5 nm prior to junction fabrication (see J. Am. Chem. Soc. 2014136, 11134-11144). The estimated resistance per molecule is found to be similar to values obtained by single molecule experiments. In addition, we found that the molecule-electrode contact resistance are independent of the applied bias and temperature indicating that the contacts are well-behaved and the thin oxide layer can be ignored (because its resistance is far smaller than that of the SAM).

Fig 2 : A schematic illustration of a SAM-based tunnel junction and equivalent circuits.

We used this technique to demonstrate a 4-fold increase in the dielectric constant (see Adv. Mater. 2015, 27, 6689–6695) by altering a single polarizable atom.

Fig 3 : A schematic illustration of SC11X based junctions and the dependence of tunneling current and dielectric constant on the terminal group.

Molecule-Electrode Interface Engineering

We showed that the coupling of a molecule with the electrodes can be controlled by spatially varying a ferrocene moiety along an alkylthiolate backbone. The junctions of the form AgTS-SCnFcC13-n//GaOx/EGaIn show how rectification varies with n as seen in the following figure. (see Nat. Commun. 2015, 6, 6324)

Biomolecular electronics

Investigation of the mechanisms of efficient long-range charge transport in biological systems is the key to understanding various biological processes in nature. The mechanism of charge transport across such systems will lead to interesting technological applications in biomolecule-based sensors or biomolecular spin-electronics. We fabricated and studied temperature and iron loading dependent electron transport through ferritin monolayer assembled on AuTS substrate. We found that three distinct mechanisms of electron transport:  coherent tunneling for biomolecular tunnel junction with small tunneling distance d < 7 nm, sequential tunneling electron transport process operative when d > 8 nm and hopping mechanism dominating for apoferritin sample without metal ion. We showed that electrons can tunnel through ~12 nm in across ferritin (obtained from our collaborator Sierin Lim). The mechanism of electron transport and interfacial magnetic interactions are investigated by temperature-dependent measurements, variation in particle size and XMCD at the SINS beamline at SSLS. (see Adv. Mater. 2016, 28, 1824–1830)

Fig 4: A schematic illustration of a ferritin-based tunnel junction GaOx/EGaIn stabilised in PDMS through-hole device. This work was highlighted by Advanced Materials as a cover.