Proteins are one of the essential building blocks of cells, and of life. Their functions are defined by unique but often complicated three-dimensional structures of folded polypeptides and cofactors; some effect DNA replication (e.g., DNA polymerase), some transport molecules (e.g., membrane transport proteins), some sustain cell structures (e.g., cytoskeleton), and some catalyze metabolic reactions (e.g., photosystems). Given the extraordinary functionality and the abundance in nature, proteins have inspired researchers from a broad range of scientific disciplines as a promising class of materials. In molecular electronics, a handful of proteins have demonstrated the capability to mediate efficient long-range (e.g., > 10 nm) charge transport and/or rectify electrical currents, creating resistors and diodes in reproducible fashion. When compared to small molecules, electrical junctions comprising ensembles (e.g., self-assembled monolayers, or SAMs) of proteins are less prone to short circuits as a result of their inherent mechanical robustness. Proteins that bear all of these useful features may potentially be the solution to transform test-bed molecular junctions into integrated circuits based on conventional fabrication and without the need of external electrical components. Now the question is, which one?
It did not take us long before for us to become interested in photosystem I (PSI), a trimeric protein complex that converts photon energy into separated electron/hole pairs to generate the redox potential imperative to photosynthesis. The disk-shape PSI trimer accommodates three electron transport chains that facilitate unidirectional flows of electrons from the P700 reaction center to the FB iron-sulfur subunit and a built-in dipole, while its unparalleled stability as membrane protein benefits device lifetime tremendously. Researchers, including us, have demonstrated the usefulness of PSI, in the form of SAMs, as the active material in bulk heterojunction and dye-sensitized solar cells, and as molecular resistors in ensemble junctions. Everything shows that we are on the right track, but we also realized that a better-controlled orientation of PSI and lower series resistance are critical to practical applications – in other words, we need a better linker.
Fullerene might just be the answer. In our recent study, C60 fullerene and its derivatives (in particular, a fulleroids decorated with glycol ethers) exhibited the unprecedented capability to self-assemble on coinage metals via non-covalent interactions (e.g., van der Waals interactions), providing a competent alternative to the established thiol chemistry in molecular electronics with improved robustness. In these ensembles, fullerenes that serve as anchoring groups are hybridized with the Fermi level of the substrate, instituting an efficient charge-injection barrier. Our intuition was to use a carboxylic acid-functionalized C60 fullerene (phenyl-C61-butyric acid, or PCBA) to guide the self-assembly of PSI, as the functional group has been known to form strong hydrogen bonds with the hydrophilic luminal or stromal surface of PSI trimers, and the densely packed fullerene cages separate the complexes from the substrate (in this work, atomically flat Au epitaxially grown on mica) to prevent undesired deformation upon physisorption.
The initial observation was already a surprise. The PSI trimers formed a dense monolayer on the PCBA linkers, and we measured a rectification ratio two orders of magnitude higher than the SAMs of PCBA alone in both EGaIn junctions (i.e., a conical tip made of eutectic gallium-indium alloy is used as top electrode to contact ~500 µm2 of the SAMs) and conductive probe atomic force microscopy (i.e., a Au-coated AFM tip is used as top electrode to contact individual complex). As control, replacing the carboxylic acid in the linker with an ester (PCBM) resulted in symmetric charge-transport across the SAMs of PSI, suggesting a unidirectional orientation of PSI on the PCBA linker; denaturing the PSI trimer or replacing it with bovine serum albumin, which does not possess an electron transport chain, significantly reduced rectification, which emphasized that the electron transport chain (or rather, its built-in dipole) is crucial to facilitate asymmetric charge-transport across the junctions. The successful identification of the role of PSI in the junctions also established a method for synthesizing resistors and diodes from the complex on demand (Figure 1b).
The asymmetric charge-transport across PSI ensembles is often considered to operate in the quantum tunneling regime, in which tunneling probability is modified by the alignment between the built-in dipole and the external electric field. The temperature dependence of current distinguishes activationless tunneling from other thermally activated processes, e.g., hopping, particularly in past works where junctions yielded smaller rectification ratios. In this work, we observed that i) charge-transport across the SAMs of PSI is temperature-independent, and ii) the rectification is dependent on the roughness of the substrate, i.e., the SAMs of PSI grown on rough substrates exhibited an order of magnitude lower rectification ratio as a result of the disrupted anisotropy of the monolayer, supporting the hypothesis that rectification originates from the collective action of the aligned dipoles. The SAMs of PSI simultaneously serve as an efficient medium for charge-transport over a distance of ~10 nm, comparable to the SAMs of small, conjugated molecules. We identified that long-range charge-transport is a property of folded polypeptides, but the mechanism of this peculiar phenomenon (e.g., coherent tunneling or activationless Marcus-Landauer hopping) requires further investigation.
The mechanical robustness of the SAMs of PSI makes it compatible with junctions fabricated from the direct injection of EGaIn top electrodes, which can be further converted into functional printed circuits. For example, two diodes comprising the SAMs of PSI on PCBA and a resistor comprising the SAM of PSI on PCBM are wired together to form an AND logic gate without external electrical components, where top electrodes and interconnects are fabricated from printed EGaIn (Figure 1a). We further modified these logic circuits to showcase their capability to modulate pulse at a frequency around 3k Hz in a reproducible fashion (Figure 1c).
The successful demonstration of functional protein ensembles that enable printable circuits immediately opens up new avenues for the development of (bio)molecular electronics. In the future, we hope to unravel the mystery of the efficient long-range charge-transport across the SAMs of PSI and apply the fabrication technique over an extended library of biocomplexes.
Author Information
Xinkai Qiu1,3 and Ryan C. Chiechi1,2
1 Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 4, 9747 AG Groningen, the Netherlands
2 Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204, United States
3 Current address: Optoelectronics Group, Cavendish Laboratory, University of Cambridge, JJ Thomson Avenue, Cambridge CB3 0HE, United Kingdom
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in