|Titel:||Toward a First-Principles Evaluation of Transport Mechanisms in Molecular Wires||Sprache:||Englisch||Autor*in:||Kröncke, Susanne||Schlagwörter:||Molecular Electronics; Organic Mixed-Valence Systems; Charge Transport; Biomolecules; Hopping; Tunneling; DNA; Density Functional Theory||Erscheinungsdatum:||2021-01||Tag der mündlichen Prüfung:||2021-03-12||Zusammenfassung:||
The quest for a deep understanding of charge transport through molecular wires is not only driven by its vital importance for biochemical processes, but also by the perspective of establishing novel functionalities beyond conventional silicon-based electronics. A central question in this context is whether the charge transport is governed by a coherent tunneling or by an incoherent hopping mechanism, which are the two main transport regimes observed in molecular conductance experiments. The predominant charge transport mechanism is strongly determined by the molecular length: a transition from the tunneling to the hopping regime is commonly indicated by a change of the length and temperature dependence of the conductance when going from shorter to longer wires around 3 to 4 nm. Since the theoretical descriptions of these two transport regimes are fundamentally different, this work aims at a predictive approach that allows for identifying the crossover length and, thus, the transport mechanism in any kind of molecular wires in an easily applicable and computationally efficient way.
The core idea behind the strategy presented in this work relies on the association of the predominating transport mechanism with charge localization properties, where transport by tunneling and hopping would be connected to delocalization and localization of an excess charge, respectively. Building on this concept, a variety of computational protocols based on Kohn–Sham density functional theory was validated with respect to their performance of describing length-dependent charge localization in organic mixed-valence systems as observed in experiments. These donor–bridge–acceptor complexes function as model systems for molecular junctions in this work due to their close relation to charge transport mechanisms based on equally length-dependent electron-transfer properties, as characterized by their Robin–Day classes.
Here, it is shown that a protocol based on the non-standard BLYP35 hybrid functional combined with a polarizable continuum model, suggested earlier by Renz and Kaupp, performs well in characterizing length-dependent charge localization in mixed-valence compounds in agreement with the experiments, even when they are located on the borderline between charge delocalization and localization. In contrast, caution needs to be exercised when applying long-range corrected functionals such as ω-B97X-D or ω-PBE with high amounts of exact exchange, as they have a tendency of overlocalization. Moreover, the computational protocol was used to demonstrate how charge delocalization can be tuned toward longer bridge lengths in para-phenylene-based wires by exploiting a captodative substitution approach that was proposed earlier in the context of molecular conductance by Stuyver and coworkers.
In the second part of this work, the DFT protocol was combined with a new measure for charge delocalization tailored for molecular wires to predict the charge transport mechanism indicated from experiments on molecular junctions at room temperature. On the example of five different sets of conjugated organic molecular wires, it was demonstrated that the tunneling-to-hopping transition length can be determined with a maximum error of one subunit. Based on the results obtained with the BLYP35 protocol in this work, it can be anticipated that it performs just as well for estimating the extent of hopping sites that are involved in the charge transport in, for example, biomolecules such as proteins or DNA, where not only the molecular length, but also the chemical structure or base sequence determines the transport mechanism.
Preliminary results were obtained on two variants of guanine-based DNA strands in this work, for which the presence of an intermediate tunneling-hopping regime, as indicated from experiments by Tao and coworkers, was supported by calculations with the BLYP35 protocol. Moreover, promising results obtained for radical anionic mixed-valence systems may pave the way for the extension of the BLYP35 protocol to the prediction of electron rather than hole transport.
Altogether, the findings resulting from this work represent a step toward the prediction of length-dependent charge transport mechanisms in molecular wires based on a first-principles protocol that is easily applicable to a variety of species. Moreover, the approach presented here offers the possibility of gaining deeper
insights into the extent of hopping sites involved in the transport process. With a view to screening larger data sets of molecules and building on preliminary results from this work, semi-empirical methods may be evaluated in more detail as a less computationally expensive alternative to DFT calculations in the future, thus contributing to an efficient and predictive theoretical framework
for molecular electronics.
|Enthalten in den Sammlungen:||Elektronische Dissertationen und Habilitationen|
geprüft am 23.09.2021
geprüft am 23.09.2021