Side-wall opening of single-walled carbon nanotubes (SWCNTs) by chemical modification: a critical theoretical study.

Abstract

Single-walled carbon nanotubes (SWCNTs) have unique electronic, mechanical, and structural characteristics; consequently, promising applications derived from these materials, such as chemical sensors or nanometer-scale electronic devices, can be expected. Structurally altered nanotubes with appropriate addends should facilitate their use by improving solubility, processability, and ease of dispersion, as well as by providing sites for chemical attachment to surfaces and polymer matrices. Avexing problem is ascertaining the detailed structures of nanotube derivatives after their preparation. The characterization of functionalized SWCNTs is difficult; all experimental attempts to determine the precise location and mode of addition of newly attached groups have failed. SWCNT adducts with possible three-membered rings (3MRs) that result from oxygen, methylene, and NH additions are simple but very important side-wall functionalized derivatives. Oxidation reactions are used widely to purify nanotubes, and the electrical properties of carbon nanotubes are extremely sensitive to oxygen exposure. Methylene and NH adducts are the prototypes of the recently synthesized covalently bonded dichlorocarbene and nitrene adducts. The available theoretical studies on the structures of nanotube oxide and dichlorocarbene adducts that involve either armchair or zigzag tube models, employed rather unsatisfactory methodology (see below). Owing to the large size of nanotubes, carefully chosen truncated models, appropriate for the problem being investigated, are required. One approach uses small nanotube fragments to simulate a full nanotube, but carries out computations at a relatively high level. The other approach uses the ONIOM technique, which treats part of the system at a high theoretical level but the rest of the system at a lower level. This strategy allows larger systems to be simulated at a practical computational cost. Thus, a recent ONIOM(B3LYP/6-31G*:AM1) study employed a C16 fragment (Figure 1a) for the high-level computation to