2011

Low-temperature scanning tunneling microscopy and spectroscopy are employed to investigate electron tunneling from a C₆₀-terminated tip into a Cu(111) surface. Tunneling between a C₆₀ orbital and the Shockley surface states of copper is shown to produce negative differential conductance (NDC) contrary to conventional expectations. NDC can be tuned through barrier thickness or C₆₀ orientation up to complete extinction. The orientation dependence of NDC is a result of a symmetry matching between the molecular tip and the surface states.

Recent low-temperature scanning-tunneling microscopy experiments [T. Kumagai et al., Phys. Rev. B 79, 035423 (2009)] observed the vibrationally induced flip motion of a hydroxyl dimer (OD)₂ on Cu(110). We propose a model to describe two-level fluctuations and current-voltage characteristics of nanoscale systems that undergo vibrationally induced switching. The parameters of the model are based on comprehensive density functional calculations of the system's vibrational properties. For the dimer (OD)₂, the calculated population of the high- and low-conductance states, the I-V, dI/dV, and d²I/dV² curves are in good agreement with the experimental results and underline the different roles played by the free and shared OD stretch modes of the dimer.

The role of the tip in inelastic electron tunneling spectroscopy (IETS) performed with scanning tunneling microscopes (STM) is theoretically addressed via first-principles simulations of vibrational spectra of single carbon monoxide (CO) molecules adsorbed on Cu(111). We show how chemically functionalized STM tips modify the IETS intensity corresponding to adsorbate modes on the sample side. The underlying propensity rules are explained using symmetry considerations for both the vibrational modes and the molecular orbitals of the tip and sample. This suggests that single-molecule IETS can be optimized by selecting the appropriate tip orbital symmetry.

The transport of charge through a conducting material depends on the intrinsic ability of the material to conduct current and on the charge injection efficiency at the contacts between the conductor and the electrodes carrying current to and from the material. According to theoretical considerations, this concept remains valid down to the limit of single-molecule junctions. Exploring this limit in experiments requires atomic-scale control of the junction geometry. Here we present a method for probing the current through a single C₆₀ molecule while changing, one by one, the number of atoms in the electrode that are in contact with the molecule. We show quantitatively that the contact geometry has a strong influence on the conductance. We also find a crossover from a regime in which the conductance is limited by charge injection at the contact to a regime in which the conductance is limited by scattering at the molecule. Thus, the concepts of 'good' and 'bad' contacts, commonly used in macro- and mesoscopic physics, can also be applied at the molecular scale.