Nanotubes from Inorganic Materials

Reshef Tenne1 and Alex K. Zettl2

1 Weizmann Institute of Science, Department of Materials and Interfaces Rehovot 76100, Israel

[email protected]

2 Department of Physics, University of California Berkeley Berkeley, CA 94720, USA [email protected]

Abstract. The inorganic analogs of carbon fullerenes and nanotubes, like MoS2 and BN, are reviewed. It is argued that nanoparticles of 2D layered compounds are inherently unstable in the planar configuration and prefer to form closed cage structures. The progress in the synthesis of these nanomaterials, and, in particular, the large-scale synthesis of BN, WS2 and V2O5 nanotubes, are described. Some of the electronic, optical and mechanical properties of these nanostructures are reviewed. The red-shift of the energy gap with shrinking nanotube diameter is discussed as well as the suggestion that zigzag nanotubes exhibit a direct gap rather than an indirect gap, as is prevalent in many of the bulk 2D materials. Some potential applications of these nanomaterials are presented as well, most importantly the superior tribological properties of WS2 and MoS2 nested fullerene-like structures (onions).

Following the discovery of carbon fullerenes [1] and later on carbon nanotubes [2], it was recognized that polyhedral structures are the thermodynam-ically stable form of carbon under the constraint that the number of atoms is not allowed to grow beyond a certain limit. However, if one considers the stimulus for the formation of such nanostructures, it is realized that these kinds of perfectly organized nanostructures should not be limited to carbon, only. As shown in Fig. 1, the propensity of nanoparticles of graphite (Fig. 1a) to form hollow closed structures stems from the high energy of the dangling bonds at the periphery of the nanoparticles, a property which is also common to materials like MoS2 Fig. (1b). It was therefore hypothesized [3,4] that the formation of closed polyhedra and nanotubes is a generic property of materials with anisotropic 2D layered structures. These kind of structures are often called Inorganic Fullerene-like (IF) structures. Thus Fig. 2 shows a multi-wall nanotube of MoS2, which was observed in ultra-thin films of such compounds grown on quartz substrates. Numerous examples of the validity of this concept have been provided over the last few years. Phrased in different terms, a significant body of evidence suggests that the phase diagram of elements, which form layered compounds, include the new phase of hollow and closed nano-materials (nanostructures) within the phase diagram of the layered compound itself. Provided that the crystallites cannot grow beyond a certain size (less than say 0.2^m), this nanostructured phase would be the

M. S. Dresselhaus, G. Dresselhaus, Ph. Avouris (Eds.): Carbon Nanotubes, Topics Appl. Phys. 80, 81-112 (2001) © Springer-Verlag Berlin Heidelberg 2001

Fig. 1. Schematic drawings of graphite (upper figure) and WS2 nanoclusters (lower figure) in a 2H polytype structure. Note that in both cases the surface energy, which destabilizes the planar topology of the nanocluster, is concentrated in the prismatic edges parallel to the c-axis (|| c) [3]

thermodynamically preferred phase. Nanotubular structures were produced also from 3D compounds, like TiO2 [5], and there seems to be no limit to the kind of compound that can serve as a precursor for the formation of nanotubular structures. However, as will become clear from the discussion below, a clear distinction holds between nanotubular structures obtained from 3D and layered 2D compounds. Intuitively, a 3D compound cannot form a perfectly ordered, flawless nanotubular or polyhedral structure, since some of the bonds, particularly on the surface of the nanotube remain unsatisfied. On the other hand, 2D (layered) compounds form perfectly crystalline closed cage structures, by introducing elements of lower symmetry into the generically hexagonal tiling of the crystalline planes. For a recent review of the subject, the reader is referred to [6].

A number of different synthetic strategies have been developed, which yield large amounts of nanotubes of V2O5 [7], WS2 [8,9], BN [10] and related

Fig. 2. TEM image of an edge of a multi-wall nano-tube (MWNT) of MoS2, 7 molecular layers thick. The distance between adjacent layers is 6.15 A. The c-axis is always normal to the surface of the nanotube [4]

structures. Although the detailed growth mechanism of the inorganic nanotubes is not fully understood, some progress has been realized in unraveling the growth mechanism.

Structural aspects of these nanoparticles are also discussed in this review. It was found that different growth strategies lead to inorganic nanotubes with quite distinct structures, as discussed below.

In some cases (such as for BN [10]) methods have been found to fine-tune the synthesis procedure and, as for carbon-based systems, it is possible to generate inorganic nanoparticles and nanotubes of a specified and monodisperse size and with a uniform number of layers. However, in general such size and product selection has not yet been demonstrated for the majority of the inorganic fullerene and nanotube analogs which we here describe. It is believed that this reflects more an "early development stage" phenomenon rather than intrinsic fundamental synthesis barriers.

So far, the properties of inorganic nanotubes have been studied rather scantily. Optical measurements in the UV-vis range and Raman scattering have provided important clues regarding the electronic structure of these nanoparticles. Generically, semiconducting nanoparticles of 3D compounds exhibit a blue shift in the absorption and luminescence spectrum due to quantum size effects. In contrast, the bandgap of semiconducting nanotubes, like BN [11], shrinks with decreasing nanotube diameter, which is attributed to the strain in the folded structure.

The mechanical properties of inorganic fullerenes and nanotubes are expected to be unique and in some cases truly exceptional. The strong covalent sp2 bonds of BN-based systems, for example, yield nanotubes with the highest Young's modulus of any known insulating fiber [12]. Very good tips for scanning probe microscopes have been prepared from WS2 nanotubes [13]. The mechanical and chemical stability of these structures is attributed to their structural perfection and rigidity. The potential applications of inorganic nanotubes as conducting or non-conducting structural reinforcements, or tips for scanning probe microscopy for the study of soft tissues, rough surfaces, and for nano-lithography are further discussed in this review. Most importantly, these kinds of nanoparticles exhibit interesting tribological properties, also briefly discussed.

0 0

Post a comment