Chemistry and electronics of carbon nanotubes go together.

One of the critical issues for the application of single-wall carbon nanotubes (SWCNTs) in nanoelectronics is the control of their electronic properties, which can be either metallic or semiconducting in their pristine form, depending on their diameter and chirality. Since all known preparative methods yield mixtures of metallic and semiconducting nanotubes, extensive research has been devoted to the modification of the nanotube electronic structure and to the separation between metallic and semiconducting carbon nanotubes. Recently, metallic and semiconducting nanotubes have been physically separated by electric fields, based on their different dielectric constants. Recently, two groups simultaneously reported how chemical reactions involving covalent bonds can modify the electronic structure of single-wall carbon nanotubes, and have a high selectivity for metallic versus semiconducting carbon nanotubes, thus opening a potential avenue to their complete separation by chemical manipulation. Beyond the practical implication of this breakthrough for nanotechnology, the fundamental interplay between electronic structure and chemical reactivity in carbon nanotubes begins to be revealed. Previously, the electronic properties of carbon nanotubes were modified by ionic doping, and recent advances have enabled significant discrimination between metallic and semiconducting nanotubes to be made, based on differences in physical properties, such as their dielectric response to electric fields and electrical breakdown, 7] or their selective ability to adsorb surfactants and DNA. What is new in the work of Kamaras et al. and Strano et al. is that they directly exploit the connection between the chemical reactivity and the electronic structure of metallic and semiconducting carbon nanotubes, which are treated as molecular species. The effect of covalent modification on the electronic structure of single-wall carbon nanotubes is demonstrated by Kamaras et al. , who treated the nanotubes with dichlorocarbene (Figure 1a). This covalent sidewall functionalization is found to convert metallic nanotubes into semiconducting ones, as shown by a strong decrease in intensity in the farinfrared absorption spectrum, assigned to intraband transitions near the Fermi level, and a simultaneous increase in the intensity in the visible region. The effect is opposite to that of ionic doping, which turns semiconducting nanotubes into conducting (metallic) nanotubes, by injecting electrons or holes into the valence or conduction bands, respectively. Covalent chemistry, on the other hand, breaks up the all-conjugated system into a series of smaller condensed aromatic systems by introducing saturated sp carbon atoms in the nanotube backbone. The concomitant electronic localization and loss of translational symmetry open a gap at the Fermi level of the metallic carbon nanotubes, turning them semiconducting (the Fermi level is the chemical potential of the electrons, in metals it is the topmost filled level at zero temperature, whereas in semiconductors it lies within the bad gap, where no states are allowed). The effect of electronic structure on chemical reactivity, on the other hand, is demonstrated by Strano et al., who treated single-wall carbon nanotubes with diazonium reagents (Figure 1b), while monitoring spectral changes. UV/ Vis-NIR spectra indicates that 4-chlorobenzenediazonium reacts preferentially with metallic nanotubes, while Raman spectroscopy spectra provide details about the electronic structure of specific nanotubes under reaction. Low-wavenumber Raman spectra, associated with radial breathing modes, allow for the full Figure 1. Two covalent reactions of single-wall carbon nanotubes. a) Result of the reaction with dichlorocarbene, which turns a nominal double bond into a cyclopropane ring (circled). b) Reaction with 4-chlorobenzenediazonium tetrafluoroborate, which releases nitrogen and adds a 4-chlorophenyl group to the nanotube.