CARBON nanotubes1 are expected to have a wide variety of interesting properties. Capillarity in open tubes has already been demonstrated2–5, while predictions regarding their electronic structure6–8 and mechanical strength9 remain to be tested. To examine the properties of these structures, one needs tubes with well defined morphologies, length, thickness and a number of concentric shells; but the normal carbon-arc synthesis10,11 yields a range of tube types. In particular, most calculations have been concerned with single-shell tubes, whereas the carbon-arc synthesis produces almost entirely multi-shell tubes. Here we report the synthesis of abundant single-shell tubes with diameters of about one nanometre. Whereas the multi-shell nanotubes are formed on the carbon cathode, these single-shell tubes grow in the gas phase. Electron diffraction from a single tube allows us to confirm the helical arrangement of carbon hexagons deduced previously for multi-shell tubes1.

Single-shell carbon nanotubes of 1-nm diameter

The use of individual molecules as functional electronic devices was first proposed in the 1970s (ref 1) Since then, molecular electronics2,3 has attracted much interest, particularly because it could lead to conceptually new miniaturization strategies in the electronics and computer industry The realization of single-molecule devices has remained challenging, largely owing to difficulties in achieving electrical contact to individual molecules Recent advances in nanotechnology, however, have resulted in electrical measurements on single molecules4,5,6,7 Here we report the fabrication of a field-effect transistor—a three-terminal switching device—that consists of one semiconducting8,9,10 single-wall carbon nanotube11,12 connected to two metal electrodes By applying a voltage to a gate electrode, the nanotube can be switched from a conducting to an insulating state We have previously reported5 similar behaviour for a metallic single-wall carbon nanotube operated at extremely low temperatures The present device, in contrast, operates at room temperature, thereby meeting an important requirement for potential practical applications Electrical measurements on the nanotube transistor indicate that its operation characteristics can be qualitatively described by the semiclassical band-bending models currently used for traditional semiconductor devices The fabrication of the three-terminal switching device at the level of a single molecule represents an important step towards molecular electronics

Room-temperature transistor based on a single carbon nanotube

CARBON exhibits a unique ability to form a wide range of structures. In an inert atmosphere it condenses to form hollow, spheroidal fullerenes. Carbon deposited on the hot tip of the cathode of the arc-discharge apparatus used for bulk fullerene synthesis will form nested graphitic tubes and polyhedral particles. Electron irradiation of these nanotubes and polyhedra transforms them into nearly spherical carbon 'onions'. We now report that covaporizing carbon and cobalt in an arc generator leads to the formation of carbon nanotubes which all have very small diameters (about 1.2 nm) and walls only a single atomic layer thick. The tubes form a web-like deposit woven through the fullerene-containing soot, giving it a rubbery texture. The uniformity and single-layer structure of these nanotubes should make it possible to test their properties against theoretical predictions.

Cobalt-catalysed growth of carbon nanotubes with single-atomic-layer walls

Dimensionality plays a critical role in determining the properties of materials due to, for example, the different ways that electrons interact in three-dimensional, twodimensional (2D), and one-dimensional (1D) structures.1-5 The study of dimensionality has a long history in chemistry and physics, although this has been primarily with the prefix “quasi” added to the description of materials; that is, quasi-1D solids, including square-planar platinum chain and metal trichalcogenide compounds,2,6 and quasi2D layered solids, such as metal dichalcogenides and copper oxide superconductors.3-5,7,8 The anisotropy inherent in quasi-1D and -2D systems is central to the unique properties and phases that these materials exhibit, although the small but finite interactions between 1D chains or 2D layers in bulk materials have made it difficult to address the interesting properties expected for the pure low-dimensional systems. Are pure low-dimensional systems interesting and worth pursuing? We believe that the answer to this question is an unqualified yes from the standpoints of both fundamental science and technology. One needs to look no further than past studies of the 2D electron gas in semiconductor heterostructures, which have produced remarkably rich and often unexpected results,9,10 and electron tunneling through 0D quantum dots, which have led to the concepts of the artificial atom and the creation of single electron transistors.11-15 In these cases, lowdimensional systems were realized by creating discrete 2D and 0D nanostructures. 1D nanostructures, such as nanowires and nanotubes, are expected to be at least as interesting and important as 2D and 0D systems.16,17 1D systems are the smallest dimension structures that can be used for efficient transport of electrons and optical excitations, and are thus expected to be critical to the function and integration of nanoscale devices. However, little is known about the nature of, for example, localization that could preclude transport through 1D systems. In addition, 1D systems should exhibit density of states singularities, can have energetically discrete molecularlike states extending over large linear distances, and may show more exotic phenomena, such as the spin-charge separation predicted for a Luttinger liquid.1,2 There are also many applications where 1D nanostructures could be exploited, including nanoelectronics, superstrong and tough composites, functional nanostructured materials, and novel probe microscopy tips.16-29 To address these fascinating fundamental scientific issues and potential applications requires answers to two questions at the heart of condensed matter chemistry and physics research: (1) How can atoms or other building blocks be rationally assembled into structures with nanometer-sized diameters but much longer lengths? (2) What are the intrinsic properties of these quantum wires and how do these properties depend, for example, on diameter and structure? Below we describe investigations from our laboratory directed toward these two general questions. The organization of this Account is as follows. In section II, we discuss the development of a general approach to the rational synthesis of crystalline nanowires of arbitrary composition. In section III, we outline key challenges to probing the intrinsic properties of 1D systems and illustrate solutions to these challenges with measurements of the atomic structure and electronic properties of carbon nanotubes. Last, we discuss future directions and challenges in section IV.

http://cmliris.harvard.edu/assets/AccChemRes_32_435.pdf

Chemistry and Physics in One Dimension: Synthesis and Properties of Nanowires and Nanotubes

Nanotubes from Carbon.

On the basis of realistic tight-binding band-structure calculations, we predict that carbon microtubules exhibit striking variations in electronic transport, from metallic to semiconducting with narrow and moderate band gaps, depending on the diameter of the tubule and on the degree of helical arrangement of the carbon hexagons. The origin of this drastic variation in the band structure is explained in terms of the two-dimensional band structure of graphite.

New one-dimensional conductors: Graphitic microtubules.

Energetics of carbon nano-tubes

Common electronic structure and pentagon pairing in extractable fullerenes

We propose a model for porous silicon, an irregular structure obtained by removing some silicon atoms randomly from a perfect silicon crystal. It is shown by using two-dimensional clusters with a tight-binding model that this model exhibits the energy-gap widening and the nonexponential decay of the photoluminescence (PL), which are in good agreement with the observed properties of porous silicon. In this model, the energy-gap widening is due to the localization of eigenstates caused by the randomness of the structure. The distributions in both size and position of the localized eigenstates, i.e., a statistical effect, yield nonexponential PL decay which is describable by the stretched exponential function. A dynamical effect due to electron and hole hopping between the eigenstates explains the observed strong energy dependence of the PL lifetimes.

Mechanisms of visible photoluminescence in porous silicon.

Photoluminescence (PL) of porous silicon in the redband is studied at temperatures from 15 to 293 K. With increasing temperature, the PL spectrum peak redshifts and the width narrows, while the PL intensity increases until \ensuremath{\sim}70 K and then decreases slightly. At 293 K, the PL at higher emission energy decays faster with a broader rate distribution. The PL decay gets slower upon cooling, and becomes independent of emission energy below \ensuremath{\sim}70 K. These results are qualitatively interpreted using a porous-cluster model that assumes random vacancy sites in a Si lattice. The low-temperature PL can be explained by carrier recombination through localized states which are distributed in energy, and dimensionally disordered. The changes in the PL with temperature, on the other hand, are understood by thermal activation of recombination processes, including hopping of carriers among distributed localized states.

Shin-ichi Sawada

Papers

New one-dimensional conductors: Graphitic microtubules.

Energetics of carbon nano-tubes

Common electronic structure and pentagon pairing in extractable fullerenes

Mechanisms of visible photoluminescence in porous silicon.

Temperature dependence of photoluminescence in porous silicon and its interpretation using the porous-cluster model.