Solid–liquid–solid growth mechanism of single-wall carbon nanotubes
Introduction
Just after their discovery in 1993, a number of physical and chemical techniques for production of macroscopic quantities of single-wall carbon nanotubes (SWNTs) was successfully demonstrated [1], [2]. However, in spite of the enormous progress in the synthesis, the understanding of the catalytic growth of SWNTs lags behind. Although a number of suggestions have been proposed to date [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], no consensus exists on the primary mechanisms of SWNT nucleation and growth. In the present paper, we attempt to reconsider the known experimental facts and to argue that the mechanism of the SWNT synthesis is similar to that of carbon nanofibers, with a difference, however, that the catalytic nanoparticles involved in the process are molten.
Section snippets
Summary of relevant experimental facts
In spite of different realizations, the physical SWNT synthesis techniques are based on the same principle: the solid carbon-containing source material is evaporated by an external energy source, the hot evaporation products react to form SWNTs, which are collected somewhere outside the synthesis zone of the reactor. The addition of a transition metal catalyst (usually Fe, Co, Ni, or their mixtures) in the source material is a characteristic feature, which is common for all these techniques.
The
Discussion
The principal inference that follows immediately from the relationship tSWNT>>tcond is that the carbon exists in the form of clusters and large aggregates during most of the SWNT growth time in the laser-furnace technique. The experiments [21] and Fig. 2 demonstrate that the existence of carbon vapor is not a necessary condition for the SWNT synthesis. Therefore, the mechanisms of the SWNT formation where the growth of the SWNTs is believed to take place completely in a vapor phase through
Summary
The existing experimental data point out to the fact that the liquefaction of the catalyst particles seems to be a decisive condition for the SWNT formation in the physical synthesis techniques. The presented arguments allow it to be considered as a liquid phase graphitization of non-graphitic forms of carbon catalyzed by molten metal–carbon alloy nanoparticles. The melting of the catalytic nanoparticles leads to a pronounced increment of carbon solubility, which alters the character of the
Acknowledgements
The authors would like to thank all the reviewers of this paper for the constructive criticism and useful suggestions. The work was supported by the European Commission (Project IST-10593 SATURN), BMBF (Project 13N7575), DFG (Project PO 392/10-1), SMWK (Project 4-7531.50-03-823-98).
References (31)
- et al.
Diameter control of the single-walled carbon nanotubes
Carbon
(2000) - et al.
Extrusion of single-wall carbon nanotubes via formation of small particles near an arc evaporation source
Chem Phys Lett
(1995) - et al.
Nanotube growth by surface diffusion
Phys Lett A
(1999) - et al.
Graphite formation by the dissolution–precipitation of carbon in cobalt, nickel and iron
Carbon
(1975) - et al.
Formation of filamentous carbon from iron, cobalt and chromium catalyzed decomposition of acetylene
J Catal
(1973) Why are carbon filaments tubular?
J Cryst Growth
(1984)- et al.
Production of carbon nanotubes
Appl Phys A
(1998) - et al.
Single-walled carbon nanotubes growing radially from YC2 particles
Appl Phys Lett
(1994) - et al.
Mechanism of carbon nanotube formation in the arc discharge
Phys Rev B
(1995)