The current huge interest in carbon nanotubes is a direct consequence of the synthesis of buckminsterfullerene, C60 , and other fullerenes, in 1985. The discovery that carbon could form stable, ordered structures other than graphite and diamond stimulated researchers worldwide to search for other new forms of carbon. The search was given new impetus when it was shown in 1990 that C60 could be produced in a simple arc-evaporation apparatus readily available in all laboratories. It was using such an evaporator that the Japanese scientist Sumio Iijima discovered fullerene-related carbon nanotubes in 1991. The tubes contained at least two layers, often many more, and ranged in outer diameter from about 3 nm to 30 nm. They were invariably closed at both ends.

A transmission electron micrograph of some multiwalled nanotubes is shown in the figure (left). In 1993, a new class of carbon nanotube was discovered, with just a single layer. These single-walled nanotubes are generally narrower than the multiwalled tubes, with diameters typically in the range 1-2 nm, and tend to be curved rather than straight. The image on the right shows some typical single-walled tubes It was soon established that these new fibres had a range of exceptional properties (see below), and this sparked off an explosion of research into carbon nanotubes. It is important to note, however, that nanoscale tubes of carbon, produced catalytically, had been known for many years before Iijima’s discovery. The main reason why these early tubes did not excite wide interest is that they were structurally rather imperfect, so did not have particularly interesting properties. Recent research has focused on improving the quality of catalytically-produced nanotubes.

The bonding in carbon nanotubes is sp², with each atom joined to three neighbours, as in graphite. The tubes can therefore be considered as rolled-up graphene sheets (graphene is an individual graphite layer). There are three distinct ways in which a graphene sheet can be rolled into a tube, as shown in the diagram below.

The first two of these, known as “armchair” (top left) and “zig-zag” (middle left) have a high degree of symmetry. The terms "armchair" and "zig-zag" refer to the arrangement of hexagons around the circumference. The third class of tube, which in practice is the most common, is known as chiral, meaning that it can exist in two mirror-related forms. An example of a chiral nanotube is shown at the bottom left.

The structure of a nanotube can be specified by a vector, (n,m), which defines how the graphene sheet is rolled up. This can be understood with reference to figure on the right. To produce a nanotube with the indices (6,3), say, the sheet is rolled up so that the atom labelled (0,0) is superimposed on the one labelled (6,3). It can be seen from the figure that m = 0 for all zig-zag tubes, while n = m for all armchair tubes.

The arc-evaporation method, which produces the best quality nanotubes, involves passing a current of about 50 amps between two graphite electrodes in an atmosphere of helium. This causes the graphite to vaporise, some of it condensing on the walls of the reaction vessel and some of it on the cathode. It is the deposit on the cathode which contains the carbon nanotubes. Single-walled nanotubes are produced when Co and Ni or some other metal is added to the anode. It has been known since the 1950s, if not earlier, that carbon nanotubes can also be made by passing a carbon-containing gas, such as a hydrocarbon, over a catalyst. The catalyst consists of nano-sized particles of metal, usually Fe, Co or Ni. These particles catalyse the breakdown of the gaseous molecules into carbon, and a tube then begins to grow with a metal particle at the tip. It was shown in 1996 that single-walled nanotubes can also be produced catalytically. The perfection of carbon nanotubes produced in this way has generally been poorer than those made by arc-evaporation, but great improvements in the technique have been made in recent years. The big advantage of catalytic synthesis over arc-evaporation is that it can be scaled up for volume production. The third important method for making carbon nanotubes involves using a powerful laser to vaporise a metal-graphite target. This can be used to produce single-walled tubes with high yield.

The strength of the sp² carbon-carbon bonds gives carbon nanotubes amazing mechanical properties. The stiffness of a material is measured in terms of its Young's modulus, the rate of change of stress with applied strain. The Young's modulus of the best nanotubes can be as high as 1000 GPa which is approximately 5x higher than steel. The tensile strength, or breaking strain of nanotubes can be up to 63 GPa, around 50x higher than steel. These properties, coupled with the lightness of carbon nanotubes, gives them great potential in applications such as aerospace. It has even been suggested that nanotubes could be used in the “space elevator”, an Earth-to-space cable first proposed by Arthur C. Clarke. The electronic properties of carbon nanotubes are also extraordinary. Especially notable is the fact that nanotubes can be metallic or semiconducting depending on their structure. Thus, some nanotubes have conductivities higher than that of copper, while others behave more like silicon. There is great interest in the possibility of constructing nanoscale electronic devices from nanotubes, and some progress is being made in this area. However, in order to construct a useful device we would need to arrange many thousands of nanotubes in a defined pattern, and we do not yet have the degree of control necessary to achieve this. There are several areas of technology where carbon nanotubes are already being used. These include flat-panel displays, scanning probe microscopes and sensing devices. The unique properties of carbon nanotubes will undoubtedly lead to many more applications. Nanohorns

Single-walled carbon cones with morphologies similar to those of nanotube caps were first prepared by Peter Harris, Edman Tsang and colleagues in 1994