Physical Society Colloquium

I will discuss some of my work on nano-elecronics using carbon nanotubes (CNTs) and silicon nanoscale field-effect transistors (Si-NFETs) as examples. Carbon nanotubes can be either metals or semiconductors depending on their structure. Their electrical properties can, in principle, be predicted from the electronic structure of graphite. I will first review these predictions, and then show the results of actual measurements. CNTs are also very polarizable objects that interact strongly via van der Waals forces with their environment and other nanotubes. The strong interaction with the substrate allows the manipulation of both position and shape of the CNTs. AFM manipulation allows the "fabrication" of simple electronic devices. The operation of a field-effect transistor employing a single 1.5 nm diameter CNT as its channel will be demonstrated. I will then show that the strong van der Waals attraction between two parts of the same tube can be used to form nanotube rings from straight CNTs. These rings when placed on appropriate electrodes have the right topology to allow the observation of electron interference effects. By measuring the resistance of the CNT-rings in the presence of a strong perpendicular magnetic field we can learn about the collisions the carriers undergo inside the CNTs and determine their coherence lengths. At very low temperatures the dominant decoherence mechanism is found to involve weakly inelastic electron-electron scattering. This is in accord with our view of these quasi-1D systems as highly correlated. Strong correlation is expected to lead to the appearance of a Fermi level singularity, and indeed such a singularity is observed at low temperature. Further discussion will involve the observation of weak and strong localization transitions, spin-flip scattering, and how these observations fit in the picture of the CNTs as prototype 1D systems.

Monday, December 20th 1999, 15:30
Ernest Rutherford Physics Building, Key Auditorium (room 112)