In this fourth lecture, we look at carbon nanotubes, analyzing their structures, scientific history, production methods, associated chemistry, purification and separation processes, and physical properties. Carbon nanotubes (CNTs) are nanomaterials that can be described as resulting from the winding of a graphitic sheet (graphene-like plane) onto itself. This winding gives rise to three main types of carbon nanotubes (chair, zigzag, helical-chiral), whose electronic properties are respectively metallic (chair CNTs) or semiconducting (chiral CNTs, zigzag CNTs). CNTs are often mixtures, but depending on the synthesis method, single-walled carbon nanotubes (SWCNT) or multi-walled carbon nanotubes (MWCNT) may be more or less preferred. Single-walled carbon nanotubes (SWCNT) are hollow carbon cylinders with diameters in the nanometer range, lengths from tens of nanometers to centimeters, and walls that are one atomic layer thick. Multiwall carbon nanotubes (MWCNTs) are made up of a few dozen concentric cylinders with regular periodic interlayer spacing of the order of 0.36 nm. Depending on the number of layers, the internal diameter of MWCNTs varies from 0.4 nm to a few nanometers, and the external diameter from 2 nm to 30 nm. The two ends of the CNTs are generally dome-shaped, as they are closed and capped by half-molecules of fullerenes (another carbon allotrope with a football-like structure of hexagonal and pentagonal panels). The existence of CNTs in very ancient materials has recently been demonstrated in swords made from Damascus steel dating back to ancient India. This discovery supports the hypothesis that these renowned blades owe their exceptional characteristics to the presence of carbon nanotubes. The scientific history of CNTs (MWCNTs) is more recent. As early as the early 1950s, numerous authors mentioned the presence of carbon threads or tubes observed by microscopy. These observations were solidly confirmed between 1973 and 1976. However, this field of research did not make its quantitative debut until 1991, with the publication in Nature by Sumio Iijima of Japan. It was undoubtedly the combined effect of the publication of a good article in a major multidisciplinary journal, a subject that resonated with the discovery of fullerenes and a certain maturity, or interest, in nanotechnology on the part of society that generated the scientific and technological tidal wave associated with CNT. However, the case of CNTs is more complicated than it seems, because they are not well-defined molecules. CNTs have different structures, masses and dimensions, and consequently different properties. Their polydispersity also leads to non-uniform properties that are difficult to predict. The manufacture of homogeneous materials requires control not only of the individual building blocks, but also of the higher-level architecture at which the elementary patterns fit together. Because of these complications, the fabrication of functional macroscopic structures that can fully utilize the exceptional properties of individual CNTs has been difficult. Indeed, most technological developments require predictable and uniform performance, and consequently many of the research strategies have focused on chemistry to prepare, in particular, SWCNTs with perfectly defined diameters, lengths, chiralities and electronic properties. In order to study their properties and incorporate them into devices or materials, it is therefore very important to separate, select and classify the different types of CNTs. We began by exploring the two main routes to CNT synthesis, starting with high-temperature routes such as graphite evaporation (T> 3,200°C), electric arc synthesis (T> 1,700°C) or laser ablation (T> 1,200°C), and ending with intermediate-temperature methods (T = 600-900°C) such as molten-salt electrolysis and plasma-assisted chemical vapor deposition (PA-CVD).
The latter method, in the presence of selected catalysts, gives SWCNT yields of 96%, over 90% of which are semiconducting. These methods are very promising, but they are still costly, relatively slow and, despite everything, do not yet enable us to obtain a totally pure system. Since analysis of the resulting systems is an essential step in any development, we have now briefly analyzed the various methods available for determining the structure and degree of dispersion of CNTs. These include scanning tunneling and electron microscopy, spectroscopy (e.g. optical absorbance, photoluminescence and Raman) and electrical measurements. Among these, Raman spectroscopy provides a rapid and reliable first analysis of CNT-based systems. We then describe the different ways in which CNTs can be functionalized. Chemical post-modification of carbon nanotubes can take place either externally or internally, according to covalent or non-covalent functionalization modes. They fall into three categories:
- covalent grafting onto sidewalls, defective sites or open ends;
- adsorption of surfactants, encapsulation or non-covalent coating of the CNT with a water-soluble polymer, using non-covalent interactions for example;
- insertion of molecular motifs inside SWNTs.
These exo- or endohedral functionalizations or the grafting of external functions also make it possible to modulate the physical and chemical properties of CNTs. In recent years, significant progress has been made towards large-scale, economically acceptable production of monodisperse SWNTs. In the absence of a truly revolutionary breakthrough, it seems that the optimal solution today is to combine the selective growth techniques mentioned above with clever post-synthetic sorting methods. Purification and sorting methods are essentially based on dissolving CNTs in the presence of adsorbed additives (mainly surface-active molecules or polymers), followed by washing and filtration. Separation of the different types of CNT, in particular SWCNT, is achieved either by chromatographic techniques (size exclusion, ion exchange, gel permeation) or electrophoresis (gel, capillary) under direct or alternating current, or by ultracentrifugation using a density gradient or weak field. In the last part of this lesson, we presented a selection of examples illustrating the many remarkable properties of SWNTs. CNTs have very high tensile strength, flexibility and elasticity, are lightweight (densities: CNT 1.35; steel 7.8; aluminum 2.7; Kevlar 1.45), have a low coefficient of thermal expansion and high electronic and thermal conductivities, and are characterized by very high anisotropy. These properties have inspired a wide range of applications such as field-effect transistors, logic gates, electrical nano-interconnections, memories, conductive films, field emission sources, infrared emitters, sensors, scanning microscopy probes, nanomechanical devices, mechanical reinforcements, hydrogen storage elements and catalytic supports. In particular, CNTs have been used in flat-panel display technologies, as a source of electron emission that strikes three-color phosphors to produce a color image. They are also used "pure" or in the form of polymer-based nanocomposites. CNTs can be found in a wide range of sporting goods, as well as in boat hulls, electromagnetic shielding, batteries, solar cells, supercapacitors, water filters and more.
For the record, the JUNO spacecraft uses CNT shielding against electrostatic discharge. An application that could be implemented in the short term and on a larger scale is the use of carbon nanotubes as fillers to control the electrical properties of polymer systems. Indeed, in certain applications, CNTs could replace conventional conductive fillers in the formulation of conductive plastics. Despite these initial successes, mastering the chemistry of carbon nanotubes remains the key to widespread application of this astonishing material. Reliable, high-volume production capacity - in 2016, annual CNT production was in the region of 10,000 tonnes - and low-cost, more selectively synthesized nanotubes are the three factors that still need to be optimized to enable large-scale commercialization of carbon nanotube-based technologies.