When KC is cooled, its resistivity begins to drop sharply at about 18°K, indicating the onset of superconductivity (Hebard, 1991). Interestingly, as larger alkali-metal cations are incorporated into the lattice and the fcc lattice parameter (a) increases, the superconducting transition temperature, T, also increases (see Figure VIII.A). Hence, the T for Rb C rises to 28°K. This rise in T may be related to an increase in the density of states at the Fermi level with increasing lattice constant (Fleming, 1991).

Figure VIII.A: Plot of superconducting transition temperature T (°K) vs. lattice parameter a (angstroms) for various compositions of AC.

The correlation between T and lattice constant (a) suggests that even higher T's could be obtained by incorporating larger and larger cations, A. There are two potential problems with this strategy. 1) As the C ions move apart, electron flow may be shut down. 2) If the cations, A, become too large to be accommodated in the octahedral and tetrahedral holes of the fcc lattice, a major reorganization of the packing would be required, which might lead to a loss of superconductivity.

Although the detailed mechanism of superconductivity in AC remains to be established, the simplicity of the materials and the progress already made suggest a definitive resolution of this question may be achieved more quickly than in the case of high-T copper-oxide superconductors.

C and its derivatives, because of their large size, stability, and hydrophobic character, may prove to have value as diagnostic or therapeutic agents in medicine. For example, derivatives of C are currently being investigated as potential inhibitors of the protease enzyme specific to the human immunodeficiency virus 1 (HIVP) (Friedman, 1993). The active site of this enzyme can be roughly described as an open-ended cylinder which is lined almost exclusively by hydrophobic amino acids. Notable exceptions to this hydrophobic trend are two catalytic aspartic acids which catalyze the attack of water on a peptide bond of the substrate.

Because a C molecule has approximately the same radius as the cylinder that describes the active site of HIVP and since C and its derivatives are primarily hydrophobic, an opportunity exists for a strong hydrophobic van der Waals interaction between the nonpolar active-site surface and the C surface. In addition, however, there is an opportunity for increasing binding energy by the introduction of specific electrostatic interactions. One obvious possibility involves salt bridges between the catalytic aspartic acids on the floor of the HIVP active site and basic groups such as amines introduced on the C surface. The key to exploiting this promising system will be the development of organic synthetic methodology to derivatize the C surface in highly selective ways.

In 1991, scientists at NEC Corporation in Japan discovered that graphitic carbon needles grew on the negative carbon electrode of the arc-discharge apparatus used for the mass production of C (Iijima, 1991). The needles ranged up to 1 mm in length and consisted of nested tubes (concentric cylinders) of rolled graphite sheets (see Figure VIII.C). The smallest tube observed was 2.2 nm in diameter, which corresponds roughly to a ring of 30 carbon hexagons. Some of the needles consisted of only two nested tubes, while others contained as many as 50. The separation between the tubes was 0.34 nm (3.4 angstroms), which matches the separation of the sheets in bulk graphite. The tips of the needles were generally closed by caps that were curved or cone-shaped. Subsequent work at NEC optimized the synthetic procedure, allowing gram quantities of carbon needles (or "nanotubes") to be isolated (Ebbesen, 1992).

Figure VIII.C: Cross-sections of several typical nested carbon nanotubes. a) Five nested tubes with an overall diameter of 6.7 nm. b) Two nested tubes with an overall diameter of 5.5 nm. c) Seven nested tubes with an overall diameter of 6.5 nm. The smallest tube in c) has a diameter of 2.2 nm.

One potentially important application for these carbon nanotubes is in the area of composites (Calvert, 1992). Carbon fibers, made from organic polymers, are used to strengthen lightweight high-tech materials such as the carbon/epoxy resins used in golf clubs, tennis racquets, bicycle frames, and yachts. Each fiber is 6,000-10,000 nm in diameter or about five times thinner than a human hair. The carbon nanotubes are substantially thinner, 4-30 nm in diameter, and could be an even more effective strengthening agent than carbon fibers in carbon/epoxy resins. The reason for this is that the breaking strength of brittle materials decreases as the size of the largest internal flaw increases. A 30 nm diameter fiber cannot have a transverse flaw bigger than 30 nm.

A somewhat more speculative use for carbon nanotubes is as molecular wires (or "nanowires"). Theoretical studies have suggested that the extremely small diameters of the tubes could lead to high conductivity, comparable to that of metals at room temperature (Mintmire, 1992).

Also intriguing is the thought of opening nanotubes on the ends and then filling them with various reagents. Synthetic procedures for this kind of nanotube "loading" have only recently been developed (Tsang, 1994). Upon treatment with concentrated nitric acid, the nanotubes caps are oxidized, resulting in the formation of opened tubes. The preferential attack of the caps (as compared to the bodies of the tubes) is apparently due to the greater curvature and, therefore, greater strain in this region. The filling of the tubes can be accomplished by carrying out the nitric acid reaction in the presence of other reagents. For example, introduction of hydrated nickel nitrate leads, after annealing, to nanotubes filled with NiO. Treatment of the NiO filled tubes with H at 400°C leads to reduction and production of tubes filled with nickel metal.