Composite Nanotubes

A composite nanotube refers to a multiwalled nanotube (MWNT) that is composed of shells with more than one kind of chemical composition. There is a big difference between a composite nanotube and compound nanotubes. A composite nanotube has distinguishable phase separation between cylindrical shells that constitute the tube, while all the shells in a compound nanotube consist of a single compound phase. Examples of compound nanotubes are nanotubes of W2S [15], Mo2S [15, 16], BN [17-20]. Composite nanotubes synthesized today are normally a mixture of carbon shells and shells of compound materials such as BN. B-C-N nanotubes are widely studied because theoretical calculation predicts that their electrical properties are controlled by their chemical composition instead of the subtle geometrical parameters [21, 22]. Similar to their pure carbon and compound counterpart, composite nanotubes are normally produced by arc-discharge and laser ablation method.

The main tool used for characterizing composite nano-tube is a high-resolution scanning transmission electron microscope (STEM) equipped with a parallel electron energy-loss spectrometer (PEELS) [23, 24]. The STEM-PEELS system has a spatial resolution in subnanometer scale which other analytical microscopes cannot compete with and thus can be used to map the spatial distribution of elements in nanostructures. A linear scan of focused electron beam across an object and measurement of the elemental concentration at each step during scanning is called elemental profiling, which is a powerful technique to determine the structure of one-dimensional nanostructures.

Earlier arc-discharge experiments for synthesizing B-C-N nanotubes used anodes made of a mixture of B, N, and C [25-27]. The obtained products contained a majority of carbon and only a very low concentration of boron and nitrogen. Elemental analysis of the nanotubes in the products showed inhomogeneous distribution of B, N, and C in the radial direction with normally B and N rich at outer surface layer [25, 27]. The B:N ratio was close to 1:1. The very low concentration of B and N in the sample suggests these elements are doped into the lattice of carbon nanotubes. There are several possibilities for how these elements are distributed in the tubes. One is that they form homogeneous BxCyNz compound shells that wrap the carbon nanotube cores. Another scheme is that BN exists in domains in the outer carbon shells. The weak signal of the B and N in STEM-PEELS, however, failed to give a conclusive evidence of intershell phase separation.

B-C-N nanotubes produced by laser ablation using a target made of a compressed mixture of BN and C powders gave a similar result as arc-discharge [24]. The results obtained by STEM-PEELS show all nanotubes are C rich although the BN and C ratio in the target is 1:1. A lot of BN crystallites were found in co-products. The tubes with a few walls are normally composed of pure carbon. Those with more walls are normally not uniform in diameter along their axis as well as the B and N distribution. The distribution of B and N is rich at the outer surface layers similar to the B-C-N nanotubes produced by earlier arc-discharge method [25, 27]. Again, it is unable to be determined whether the B and N are homogeneously doped into graphite lattice (true ternary Bx CyNz phase) or they are mixed binary phases such as BN or BCx embedded in the graphite outer layers (subcomposite tubes).

The laser ablated B-C-N nanotubes not only have similar morphologies and element distribution as those in earlier arc-discharged products, they should also share a similar growth mechanism since both methods employ a local high-temperature process to vaporize graphite and BN. From the feature that all obtained composite nanotubes have carbon cores, a two-step growth model has been proposed and is shown in Figure 1 [24]. The first step is the nucleation and growth of pure multiwalled carbon nanotubes. This step occurs in the high-temperature plasma region where B and N can easily diffuse out of the graphite lattice and form more favorable BN crystallites of planar structure. Carbon nanotubes formed in the process are all multiwalled because the presence of B and N atoms in this step could introduce defects on the graphitic wall that are preferable nucleation

Figure 1. Growth model of heterogeneous B-N-C composite nanotubes. Reprinted with permission from [24], Y. Zhang et al., Chem. Phys. Lett. 279, 264 (1997). © 1997, Elsevier Science.

sites for additional layers and the foreign atoms could also serve as the bridging atoms in lip-lip growth model [28]. The second step is partial coating of B-C-N layers. The growth of additional walls could also initiate from the defects on the carbon nanotube wall. The B and N diffusion process is reduced by lower temperature when the nanotube moves out of the plasma center. The inert gas used in the arc-discharge and laser ablation processes provides an efficient heat exchange medium for cooling down the species and works as a barrier to confine the expansion of the plasma and sustain a high density of multielement vapor for doping BN in nanotubes.

The two-step model, however, is not adequate to explain the formation of all B-C-N nanotube structures. One exception is the sandwiched C-BN-C nanotubes, where carbon is rich both in core shells and in outer shells [23]. The growth mechanism for such sandwiched nanotubes is still not clear. One possible model is a simultaneous growth of multiphase shells through a lip-lip-interaction [28]. Another possibility is that the tube flies through three zones in the discharge chamber during its layer-by-layer growth, and the three zones have different concentrations of active elements and thus form sandwiched structure.

The sandwiched C-BN-C nanotubes were produced by arc-discharge in nitrogen atmosphere using a HfB anode and a carbon cathode [23]. The use of non-carbon anode increases the B and N concentration in the product because the carbon cathode is only slightly evaporated in the arc-discharge. The high concentration of B and N enables more precise determination of local chemical composition using a STEM-PEELS. Figure 2 shows a high-resolution electron

0 0

Post a comment