'Seed' molecules have been made that enable synthesis of just one kind of single-walled carbon nanotube, rather than a mixture of species. This paves the way for the preparation of pure samples of any nanotube species. See Letter p.61
The wonderful thing about single-walled carbon nanotubes (SWCNTs) is that more than 100 species can be produced by various growth methods. But this is also the most frustrating thing about them. It is expected that different nanotube species will have different applications, but the nanotubes generally form as mixtures of around 5–50 species from any preparation method1,2. Moreover, separation methods are cumbersome because of the many species that can form. After many attempts over the past two decades to grow single species of SWCNTs, Sanchez-Valencia et al.3 present a route to success on page 61 of this issue.
Each SWCNT species can be defined by a pair of integers (n,m) called chirality indices, which describe how a graphene sheet (a single layer of carbon atoms in graphite) would hypothetically be rolled up to generate a tubular structure4. Chirality indices can be used to determine the two unique, fundamental parameters of each rolled-up graphene structure — the tube diameter and the angle with respect to a plane perpendicular to the tube's long axis at which the graphene would be rolled into a tube. The term 'chiral' is sometimes a misnomer, because chirality is a property associated with asymmetry, but some SWCNTs are not asymmetric.
Although there are many species of SWCNT (Fig. 1), there are only two main types: metallic nanotubes, which conduct electricity in the same way as gold or aluminium; and semiconducting nanotubes, whose electrical conductivities are tunable, as in the semiconductors silicon and gallium arsenide. Conductivities are determined by a property called the bandgap — the smaller the bandgap, the larger the room-temperature conductivity. Metallic SWCNTs have a bandgap of 0 electronvolts (eV), whereas semiconducting nanotubes have a bandgap that can vary from approximately 1 meV to 1.5 eV (ref. 5). Specific bandgaps are required for particular applications. For example, a bandgap of 0 eV is desirable for electrical wire and cable applications, whereas a larger bandgap is preferred for transistors. For photonic applications, different bandgaps are required to generate or detect different colours6.
Sanchez-Valencia and co-workers prepared exclusively (6,6) species of tubes starting from predefined 'seeds' — organic molecules prepared by multi-step synthesis. They grew SWCNTs from each of these seeds on a platinum surface at 500 °C, using ethanol as a source of carbon atoms. The idea of using molecules to control the chirality of nanotubes is not new7, but the authors have taken the concept of seed design for specific nanotube growth to an extraordinary level: the precise arrangement of the atoms in the seed predefines the species of tube grown. Their work suggests that it should be possible to design and synthesize seeds for any desired SWCNT species.
Impressively, the researchers used scanning tunnelling microscopy to image the orientation of the seeds on the platinum surface and to take snapshots during the main phases of the nanotube-forming process — that is, the formation of a bowl-shaped cap from the seed and the subsequent 'base-growth' stages (in which the catalytic platinum atoms stay at the substrate surface and the top of the nanotube is catalyst-free). They also studied their nanotubes using Raman spectroscopy, observing a single peak in a band of the Raman spectrum that is diagnostic of the species of nanotube being analysed. This provided beautifully simple confirmation that only one species of nanotube forms from the seeds, and unambiguously pinpointed the nanotube structure. Furthermore, the authors performed extensive computational modelling to understand the different phases of the nanotube-forming process.
Sanchez-Valencia and co-workers' method is currently the only one that allows predictable control of the chirality of SWCNTs. In another approach reported8 this year, (12,6) SWCNTs were prepared as 92% of the nanotube mixture using a solid alloy catalyst, but the tube species that grew could not be predetermined. The relatively low temperature (500 °C) used in Sanchez-Valencia and colleagues' procedure probably helps to maintain species specificity because, at higher temperatures, small variations in the growth temperature can cause changes in the chirality index along a tube9.
Some might view the need for a 10-step organic synthesis of the seeds as an overburdening limitation of the new approach. It is not. Consider that 1 mole of seeds is 6 × 1023 molecules, equating to 1.2 kilograms of material — a quantity that could easily be prepared by a chemical company. If, as Sanchez-Valencia et al. show, 50% of those seeds adopt the necessary conformation for growth at the platinum surface, then more than 5 tonnes of 10-micrometre-long SWCNTs could be obtained from just 1 mole of seeds.
However, further challenges remain. The new method produces nanotubes that stand perpendicular to the growth surface, like the bristles of a carpet. This minimizes entanglement of the nanotubes, but they will still form bundles when they reach a certain length. Many applications need SWCNTs to be unbundled, and so the nanotubes will need to be subsequently treated with solvent or wrapped with polymers. Furthermore, the surface area covered by SWCNTs using typical growth methods is of the order of 1% (ref. 10); using Sanchez-Valencia and colleagues' approach, about 30 km2 of platinum would be needed to accommodate 1 kg of seeds at this surface density, assuming that half of them grow. Placing nanotubes in arrays precisely where they are needed has also been a persistent obstacle to the development of many devices. Lastly, it remains to be seen whether molecular seeds can be made that selectively control the formation of other nanotube chiralities.
Sanchez-Valencia and colleagues' work represents a stellar breakthrough in the synthesis of SWCNTs. To those who have worked in this field for the past two decades, it is humbling to think that the selective growth of these diminutive structures has taken so long. But it is comforting to see it done so definitively.
Kitiyanan, B., Alvarez, W. E., Harwell, J. H. & Resasco, D. E. Chem. Phys. Lett. 317, 497–503 (2000).
Bachillo, S. M. et al. Science 298, 2361–2366 (2002).
Sanchez-Valencia, J. R. et al. Nature 512, 61–64 (2014).
Dresselhaus, M. S., Dresselhaus, G. & Eklund, P. C. in Science of Fullerenes and Carbon Nanotubes (Academic, 1996).
Matsuda, Y., Tahir-Kheli, J. & Goddard, W. A. III J. Phys. Chem. Lett. 1, 2946–2950 (2010).
Hároz, E. H. et al. Nanoscale 5, 1411–1439 (2013).
Fort, E. H. & Scott, L. T. J. Mater. Chem. 21, 1373–1381 (2011).
Yang, F. et al. Nature 510, 522–524 (2014).
Doorn, S. K. et al. Phys. Rev. Lett. 94, 016802 (2005).
Pint, C. L., Xu, Y.-Q., Pasquali, M. & Hauge, R. H. ACS Nano 2, 1871–1878 (2008).
About this article
Journal of Materials Chemistry A (2016)
Enhanced Air Stability in REPb3(RE = Rare Earths) by Dimensional Reduction Mediated Valence Transition
Inorganic Chemistry (2016)
Selective Surface Charge Sign Reversal on Metallic Carbon Nanotubes for Facile Ultrahigh Purity Nanotube Sorting
ACS Nano (2016)
Scientific Reports (2016)