Bent and Twisted Nanotubes

Odom et al. [29] recently reported a kink in an atomically-resolved SWNT rope arising from a mechanical distortion (Fig. 7). The bend angle defined by this kink is approximately 60°. Tunneling spectroscopy was used to characterize the electronic properties of the uppermost nanotube in the bent rope. I-V measurements were performed at the positions indicated by the symbols in Fig. 7a, and their corresponding dl/dV are displayed in Fig. 7b.

The positions of the van Hove peaks indicate that the tube is metallic. Significantly, the data also show new features at low bias voltages on either side of the bend. These peaks are likely due to the presence of the bend, since five nanometers away (+) from the kink, the sharpness and prominence of these features have greatly diminished. In addition, Avouris et al. [40] reported that STS spectra taken near a kink in a semiconducting nanotube also showed an increased DOS at low energies. Notably, recent calculations on bends in armchair tubes show similar low energy features in the DOS for similar bend angles [39] observed by [29] and [40].

The bend region observed by Odom and co-workers was further investigated using bias-dependent STM imaging. On the right side of the bend, a superstructure on the tube is observed at the biases of the localized peaks (Fig. 7c,d ). Figure 7c shows stripes parallel to the zigzag direction of the tube and Fig. 7d displays a triangular ring structure, where the spacing between nearest-neighbor rings is ca. 0.42 nm (the zigzag spacing). These new electronic features could be due to electron scattering and interference at the defect site [41]. Although the bias voltage, 0.45 V, at which Fig. 7e was imaged is not at a prominent peak in the dI/dV, some electronic structure can be seen extending ~ 1.5 nm to the right of the bend. However, this additional structure diminishes and an unperturbed atomic lattice is observed, consistent with the spectroscopic measurements. Further experimental and

Voltage (V)

Fig. 7. STM image and spec-troscopy of a bend in a rope of SWNTs. (a) Image of ~ 60° bend. The symbols correspond to locations where I-V were measured. The scale bar is 1 nm. (b) Differential conductance calculated from the locations indicated in (a). The upper portion of the graph is spectroscopy performed on the left side of the bend over 5 nm. The lower portion of the graph is spectroscopy performed on the right side of the bend over 2 nm. The dashed lines highlight the low-energy features. (c—e) STM images recorded at bias voltages of —0.15, 0.15, and 0.45 V, respectively [29]

computational work is needed to elucidate clearly such interesting observations.

Besides bends in nanotubes, twisting of individual tubes within ropes has been reported by Clauss et al. [33]. Large-scale nanotube twists may result from mechanical interactions during deposition upon surfaces, be introduced during the growth process and frozen by shear forces, or result from different helicity tubes attempting to align their hexagonal lattices within ropes. Clauss et al. [33] observed anomalous lattice orientations upon careful inspection of many tubes, namely, that the armchair direction is on average perpendicular to the tube axis, and that the average angle between the zigzag and armchair direction is greater than 90°. This apparent distortion from an equilibrium conformation can be explained if the imaged nanotubes are of the armchair-type with a twist distortion of several degrees, or from distortions contributed by the finite size and asymmetry of the STM tip.

0 0

Post a comment