We measure the temperature dependence of the Raman spectra of metallic and semiconducting nanotubes. We show that the different trend in metallic tubes is due to phonon re-normalization induced by the variation in electronic temperature, which is mod
eled including non-adiabatic contributions to account for the dynamic, time dependent nature
Graphene nanoribbons are the counterpart of carbon nanotubes in graphene-based nanoelectronics. We investigate the electronic properties of chemically modified ribbons by means of density functional theory. We observe that chemical modifications of z
igzag ribbons can break the spin degeneracy. This promotes the onset of a semiconducting-metal transition, or of an half-semiconducting state, with the two spin channels having a different bandgap, or of a spin-polarized half-semiconducting state -where the spins in the valence and conduction bands are oppositely polarized. Edge functionalization of armchair ribbons gives electronic states a few eV away from the Fermi level, and does not significantly affect their bandgap. N and B produce different effects, depending on the position of the substitutional site. In particular, edge substitutions at low density do not significantly alter the bandgap, while bulk substitution promotes the onset of semiconducting-metal transitions. Pyridine-like defects induce a semiconducting-metal transition.
We demonstrate electrochemical top gating of graphene by using a solid polymer electrolyte. This allows to reach much higher electron and hole doping than standard back gating. In-situ Raman measurements monitor the doping. The G peak stiffens and sh
arpens for both electron and hole doping, while the 2D peak shows a different response to holes and electrons. Its position increases for hole doping, while it softens for high electron doping. The variation of G peak position is a signature of the non-adiabatic Kohn anomaly at $Gamma$. On the other hand, for visible excitation, the variation of the 2D peak position is ruled by charge transfer. The intensity ratio of G and 2D peaks shows a strong dependence on doping, making it a sensitive parameter to monitor charges.
