NANOTUBE ELECTROMECHANICAL OSCILLATOR
Carbon nanotubes are often presented as the ultimate material for mechanical resonators and other nanoelectromechanical systems (NEMS). Nanotubes devices have cross-sections of the order of the nanometer, which is smaller than the resolution limits of common lithographic techniques. Contrary to silicon often used in NEMS, carbon nanotubes are chemically inert and do not suffer from the surface roughness inherent to lithographically patterned NEMS. This is very important in order to obtain high Q resonance quality factors. Furthermore, they are the stiffest material known and have low density, so that frequencies are expected to be very high, above 1GHz. Our goal is to fabricate nanotube resonators with the resonance frequency and the Q factor that are as high as possible. This may enable the probe of ultra-low forces or ultra-low masses with sensitivities that have never been detected. On a basic research point of view, NEMS oscillators have been proposed for the exploration of quantum phenomena in macroscopic systems such as the detection of the zero-point motion. Such an observation in the vibrations of a nanotube would show that the quantum rules of the microscopic world can still be applied to systems that are much larger and that are composed of several thousands of atoms. The observation of the zero-point motion for a macroscopic NEMS oscillator has not been achieved yet, but different groups are racing for such a demonstration.
NANOTUBES AS MOLECULAR CONNECTORS
Building electronic devices at the molecular scale has motivated intense research for the last years. Although the concept was proposed 30 years ago, only recently experiments aimed at contacting individual organic molecules are being reported. High expectations have been raised with the first results, but problems have quickly appeared due to the interface between the molecule and metal electrodes. The organic molecules are usually anchored to gold electrodes by thiol groups. However, this is not enough for the control of the interface at the atomic scale. In view of the difficulties related with the reproducibility of the molecule/electrode interface, we use nanotubes as molecular connectors in order to contact smaller molecules. This is motivated because nanotubes are chemically inert, so that the surface stays quite clean. Moreover, nanotubes have a well defined geometry with a diameter comparable to the size of the studied molecule. Nanotubes are thus a system of choice to electrically connect small molecules to the outside world. We develop devices with new layouts using nanotubes are molecular connectors. This may allow a better understanding of such small molecules. For example the molecules can then be manipulated using electric field or light in order to characterise the modifications brought to the molecule.
LUTTINGER LIQUID IN CARBON NANOTUBES
Coulomb interaction effects have pronounced consequences in carbon nanotubes due to their one-dimensional nature. In particular, electrons are expected to form a Luttinger liquid (LL) rather than a conventional Fermi liquid phase. The electron is an unstable particle and spontaneously decays into collective plasmon modes. The velocity vg of these modes differs from the Fermi velocity vf. In a Luttinger liquid, spin and charge plasmons decouple and moreover propagate with different velocities - the spin mode propagates at vf. This phenomenon is called spin-charge separation and implies that the spin and charge degrees of freedom of an electron brought into a LL will spatially separate. In a Fermi liquid vg=vf and therefore this characteristic feature will not show up. Another interesting issue concerns the fractionalized stable excitations of the LL. For a spinless system, one can theoretically establish that quasiparticles scattered by a weak impurity potential have fractional charge and obey fractional statistics. This has a lot in common with the famous Laughlin quasiparticles in the fractional quantum Hall effect. In view of those characteristics, it is understandable that, for many decades, experimentalists have attempted to find LL behaviours. Recently, some experimental evidences for LL behaviours in individual SWNTs have been reported. A lot of very important predictions of the Luttinger liquid theory however remain to be tested, such as the spin-charge separation or the existence of fractionalized stable excitations.