Great advances have been achieved recently in producing carbon nanotubes (CNTs). By the same token, precisely controlling the growth of both multiwalled and single-walled carbon nanotubes (MWNTs and SWNTs) to generate desired structures remains a major challenge. The importance of this deficiency cannot be underestimated because some of the most promising applications of CNTs in next-generation electronics and optical devices depend strongly on nanotube diameter size and specific chirality (i.e., whether the tube winds left or right). For this reason, better understanding of the initial stages and mechanisms of CNT growth could be key to developing practical, large-scale synthesis of CNTs.
Generating nanotubes involves reacting a gaseous carbon source on metal catalyst particles in a process known as chemical vapor deposition (CVD). Observation of the gas–solid interactions is possible using an ‘environmental’ transmission electron microscope (TEM) fitted out with gas injectors and cells. Indeed, several groups have carried out in situ observations of carbon nanofiber and nanotube growth.1,2 But nucleation—the first and most important step of nanotube production—is still a puzzle, mainly because the environmental TEM images are blurred by vibrations caused by the chemical reaction or electrical currents. In order to get around this problem, we observed the formation of CNTs in an entirely condensed phase (solid) process. Moreover, we were able to monitor, at atomic resolution, the birth of both SWNTs and MWNTs in a standard TEM.3
To grow a CNT inside the TEM, we subject an MWNT containing a catalytically active metal particle core to electron irradiation. This causes carbon from the graphitic shells of the MWNT to be injected into the body of the metal core and subsequently to emerge as an SWNT or MWNT inside the host nanotube (see Figure 1). This process differs from previous in situ studies because it does not involve a gaseous source of carbon, and electrical currents do not pass through the nanotubes. Instead, it is based on the dynamics of electron-irradiated carbon atoms in heated carbon nanostructures.4,5
Figure 1. The sequence of images shows the growth of an MWNT from a metal catalyst particle. The process is driven by the injection of carbon atoms, produced by electron irradiation, into the metal particle.
Host MWNTs containing iron, cobalt, or iron-cobalt nanowires were produced in a CVD reaction chamber by pyrolysis of the vapor derived from a hydrocarbon-organometallic solution. Electron irradiation and imaging of the CNT was carried out in a TEM (FEI Tecnai F30) with an accelerating voltage of 300kV. All TEM experiments were performed at a specimen temperature of 600°C using a specially equipped heating stage.
Our observations indicate that there is direct bonding between the CNTs and the metal surface from which the tubes sprout. The results are readily explained by bulk (subsurface) diffusion of carbon through the body of the catalytic particle. In addition, we were able to derive a possible mechanism of CNT growth (see Figure 2).
Figure 2. Model of CNT growth. The initial metal-filled MWNT is irradiated by the electron beam, and the carbon atoms injected into the metal catalyst particle are expelled as a newly born CNTs via bulk diffusion.
In this model, carbon atoms from the host MWNT are knocked into the metal particle by the electron beam. The flat end of the metal particle appears to deform, creating a convex dome. Apparently simultaneously, a surface-covering carbon cap appears, which almost certainly is a hemispherical fullerene structure. Around the edge, at the base of the dome, metallic steps develop from which the walls of new CNT segments sprout. Finally, carbon atoms are fed continuously into the metal particle, through which they migrate rapidly to supply the root regions of the growing CNTs.
Metal-filled MWNTs could be used as solid-state reaction cells for monitoring nanotube growth in situ at very high resolution. CNT growth has been observed for a variety of encapsulated metals, including iron, cobalt, and nickel. The most compelling explanation of these results is that carbon atoms liberated from host MWNTs by irradiation are injected into the crystalline metallic wires and trigger extrusion of capped nanoubes via a bulk diffusion process. This mechanism could also serve as a new method of making CNTs.
Mauricio Terrones, Julio A. Rodríguez-Manzo, Humberto Terrones
Advanced Materials Department
Instituto Potosino de Investigación Científica y Tecnológica (IPICyT)
San Luis Potosí, Mexico
Mauricio Terrones has coauthored more than 195 publications in international journals, and he counts more than 4250 independent citations to his work. He has published papers in Nature, Science, Physical Review Letters, Nano Letters, and Applied Physics Letters. He has received several awards, including an Alexander von Humboldt fellowship, the Javed Husain Prize (UNESCO), and the TWAS Prize in Engineering Physics.
Julio A. Rodríguez-Manzo received his PhD in nanosciences and nanotechnology from IPICyT in 2007. He has published more than 14 papers in journals such as Science, Nature Nanotechnology, Small, and Nano Letters.
Humberto Terrones heads the Advanced Materials Division at IPICyT. He obtained his PhD at Birkbeck College, University of London, in 1992, and has received international and national recognition. His research interests include theoretical and experimental aspects of fullerenes and curved structures, mathematics of surfaces, flexicrystallography, and synthesis of complex atomic structures. He has published more than 130 papers in major journals.
Harold W. Kroto
Department of Chemistry and Biochemistry
Florida State University
Harold Kroto received a BSc (chemistry, 1961) and a PhD (molecular spectroscopy, 1964) from the University of Sheffield. He started his academic career at the University of Sussex, Brighton, in 1967. He became a professor in 1985 and a Royal Society Research Professor in 1991. In 1996 he was knighted for his contributions to chemistry. Later that year, he was awarded the Nobel Prize in Chemistry for the discovery of fullerenes.
Institut für Physikalische Chemie
Litao Sun is currently working as posdoctoral fellow at the University of Mainz. He is investigating irradiation effects on carbon nanostructures.
Institut de Physique et Chimie des Matériaux
Université de Strasbourg
Florian Banhart is professor at the University of Strasbourg. He has coauthored more than 100 publications in prestigious journals and in recent years has concentrated on the irradiation effects of carbon nanostructures. He has published key papers in major journals, including Science, Nature, Nature Nanotechnology, Nature Materials, Physical Review Letters, and Nano Letters.
1. S. Helveg, C. López-Cartes, J. Sehested, P. L. Hansen, B. S. Clausen, J. R. Rostrup-Nielsen, F. Abild-Pedersen, F. Abild-Pedersen, J. K. Nørskov, Atomic-scale imaging of carbon nanofibre growth, Nature 427, no. 6973, pp. 426-429, 2004. doi:10.1038/nature02278
3. J. A. Rodríguez-Manzo, M. Terrones, H. Terrones, H. W. Kroto, L. Sun, F. Banhart, In situ nucleation of carbon nanotubes by the injection of carbon atoms into metal particles, Nat. Nanotechonol. 2, no. 5, pp. 307-311, 2007.