Abstract:
The properties of nanoscopic objects depend critically on the position of each atom, since finite-size and quantization effects play an important role. For carbon nanotubes, the electronic, mechanical, and vibrational properties vary significantly depending on their structure. For example, a carbon nanotube can be metallic or semiconducting with varying band-gaps depending on its lattice structure. Yet, most investigations on individual carbon nanotubes are carried out on objects with unknown structure. The analysis of individual single-walled nanotubes (SWNTs) is commonly done by Raman spectroscopy or electric transport measurements, providing information about the vibrational properties and electronic structure. In some Raman spectroscopic investigations, the nanotube structures could be deduced from the spectroscopic data by a modelization of electronic and vibrational properties. But no previous work exists where an unambiguous and independent identification of the SWNT lattice structure is available on the same nanotube as the Raman or transport measurement.
This problem was solved in this work by developing a versatile approach that facilitates an electron diffraction analysis in a transmission electron microscope (TEM) in combination with a wide range of experimental techniques on the precisely same nano-object. I present in this thesis the first analysis of Raman active vibrational modes of individual single-walled carbon nanotubes with an independent and unambiguous identification of the nanotube lattice structure by electron diffraction. This is important because as long as the structure is deduced via modelization, a verification of the model is not possible. Further, the index assignment only from Raman spectroscopic data is only possible for small diameter nanotubes. We have measured the radial breathing mode frequency dependence on the nanotube diameter for a large diameter range, and find that the previously established relationships can not be extrapolated to the larger diameter nanotubes. Also, our measurements are done on freely suspended nanotubes, so that an influence from the environment (substrate, suspension) can be excluded. Yet, we find a surprisingly good agreement with previous work with nanotubes on substrates or in suspension, which shows that the effect of the environment is much smaller than previously assumed.
I also present electronic transport measurements, partly in a transistor configuration, with an independent characterization of the same object by high-resolution transmission electron microscopy and electron diffraction. The metallic or semiconducting behaviour as predicted for the given indices is confirmed by the transport measurement. Further, in-situ transport measurements provide data on modifications of the metal-nanotube contact, and allow a visualization of nanotube breakdown under high currents.
Further, I show various novel devices based on individual single-walled carbon nanotubes. These include nanoelectromechanical devices comprising individual SWNTs as molecular-scale motion-enabling element. The nanoelectromechanical devices demonstrate the enormous potential for carbon nanotubes in future integrated devices that may be not only electronically but also mechanically active. The single-molecule based pendulum can be turned to any position between 0° and almost 180° with the use of a single electrostatic potential. Deflections and oscillations can be induced by extremely small forces, so that the device may serve as very sensitive nanoscale force sensor. We observe the thermally excited oscillations at room temperature. The torsional deformation of a chiral molecule lifts the mirror symmetry between the enantiomers, which allows a determination of its handedness.