Carbon Nanotube Electronics and Optoelectronics
- PDF / 476,697 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 85 Downloads / 237 Views
Carbon Nanotube
Electronics and Optoelectronics Phaedon Avouris
Abstract Carbon nanotubes (CNTs) are one-dimensional nanostructures with unique properties. This article discusses why CNTs provide an ideal basis for a future carbonbased nanoelectronic technology, focusing specifically on single-carbon-nanotube fieldeffect transistors (CNT-FETs). Results of transport experiments and theoretical modeling will be used to address such issues as the nature of the switching mechanism, the role of the metal contacts, the role of the environment, the FET scaling properties, and the use of these findings to produce high-performance p-type, n-type, and ambipolar CNT-FETs and simple intra-nanotube circuits. CNTs are also direct-gap nanostructures that show promise in the field of optoelectronics. This article briefly reviews their optical behavior and presents results that show that ambipolar CNT-FETs can be used to produce electrically controlled light sources based on radiative electron–hole recombination. The reverse process—that is, the generation of photocurrents by the irradiation of single CNT-FETs—and photoconductivity spectra of individual CNTs are also demonstrated. Keywords: carbon nanotubes, excitons, field-effect transistors, molecular transport junctions, nanoelectronics, nanophotonics, photoconductivity.
C na1 + ma2 , where a1 and a2 are the unit vectors of the hexagonal honeycomb lattice. Any CNT can be described by a pair of integers (n,m) that define its chiral vector. The unit cell of the CNT is defined as the rectangle formed by C and the 1D translational vector T, identified in Figure 1a. The interesting electrical properties of CNTs are due in large part to the peculiar electronic structure of graphene (Figure 1b). In going from graphene to a CNT by folding, one has to account for the additional quantization arising from electron confinement around the CNT circumference. This circumferential component of the wave vector kC can only take values fulfilling the condition kC C 2j, where j is an integer. As a result, each graphene band splits into a number of 1D subbands labeled by j. These allowed energy states are cuts of the graphene band structure. When these cuts pass through a K point (Fermi point) of the graphene Brillouin zone, the tube is metallic (Figure 1b). Otherwise the tube is semiconducting (Figures 1c and 1d). It can be shown that an (n,m) CNT is metallic when n m [see the (10,10) tube in Figures 1c and 1d]. It has a small gap, due to curvature-induced – mixing, when n – m 3i, where i is an integer; meanwhile, CNTs with n – m 3i are truly semiconducting [see the (20,0) tube in Figures 1c and 1d].2,3,4 Semiconducting CNTs have a diameter-dependent bandgap; a tight binding description of the electronic structure is given by Eg 2a/3 dCNT,
Introduction Over the past several decades, we have witnessed extraordinary advances in siliconbased electronic technologies. This progress has been achieved primarily through the downscaling of the metal oxide semiconductor field-effect
Data Loading...
