Electronic Flux Density Maps Reveal Unique Current Patterns in a Single-Molecule-Graphene-Nanoribbon Junction
To assist the design of novel, highly efficient molecular junctions, a deep understanding of the precise charge transport mechanisms through these devices is of prime importance. In the present contribution, we describe a procedure to investigate spatially-resolved electron transport through a nanojunction from first principles, at the example of a nitro-substituted oligo-(phenylene ethynylene) covalently bound to graphene nanoribbon leads. Recently, we demonstrated that the conductivity of this single-molecule-graphene-nanoribbon junction can be switched quantitatively and reversibly upon application of a static electric field in a top gate position, in the spirit of a traditional field effect transistor [J. Phys. Chem. C, 2016, 120, 28808-28819]. The propensity of the central oligomer unit to align with the external field was found to induce a damped rotational motion and to cause an interruption of the conjugated $π$-system, thereby drastically reducing the conductance through the nanojunction. In the current work, we use the driven Liouville-von-Neumann (DLvN) approach for time-dependent electronic transport calculations to simulate the electronic current dynamics under time-dependent potential biases for the two logical states of the nanojunction. Our quantum dynamical simulations rely on a novel localization procedure using an orthonormal set of molecular orbitals obtained from a standard density functional theory calculation to generate a localized representation for the different parts of the molecular junction. The transparent DLvN formalism allows us to directly access the density matrix and to reconstruct the time-dependent electronic current density, unraveling unique mechanistic details of the electron transport.