In 1990, almost exactly five years after C60 Buckminsterfullerene (see Figure 1) was discovered serendipitously during a series of graphite laser vaporization experiments designed to simulate the chemistry in a red giant carbon star, the molecule was isolated in macroscopic amounts. This breakthrough has triggered an explosion of research, and the molecule has already exhibited a wide range of novel phenomena which promise exciting applications. In addition to generating intense experimental interest in its physicochemical properties, the molecule holds a particular fascination for theoreticians; indeed the existence of the molecule was predicted in 1970, some 15 years before its discovery in 1985. It is interesting to consider the intrinsic aesthetic aspects of this molecule which are responsible for the extraordinary charisma it has for scientists and nonscientists alike.

Although C60 and other such molecules had been predicted earlier on the basis of theoretical considerations, the identification of these molecules by mass spectrometry in the year 1985 is an important landmark. The discovery of a ready method of preparation by arc evaporation of graphite in 1990 gave a big boost to this area of research. Today, research on carbon has undergone a renaissance and the new carbon molecules and structures probably constitute the second greatest sensation in physical science since the 1986 discovery of high-temperature super-conductivity in cuprates.

These fascinating cage molecules of carbon, containing 6- and 5-membered rings, with considerable pi electron density, exhibit interesting chemical as well as materials properties. The phase transitionh of fullerenes and the electronic properties of the derivatives of C60 are indeed interesting. And, as if this were not enough, many other novel carbon structures have recently been discovered. Nanotubes are a major discovery. Closed on either end, nanotubes can be opened by mild oxidation at the reactive site at the closed end and can be filled with metals. Carbon onions made of concentric giant fullerenes constitute another interesting form of carbon. The onions have also been stuffed with metals and metal carbides. Clearly, carbon research has many surprises in store, with promising applications. It is only appropriate that the MRS Bulletin has dedicated a special issue to the new carbon structures and we are happy to serve as guest editors, presenting a wide variety of topics that convey the flavor of this fascinating field. We begin with a personal account of Buckminsterfullerene by co-guest editor Harry Kroto.

By virtue of their unique structures, fullerenes exhibit novel chemical transformations. Particularly pertinent to this article are the interesting properties exhibited by fullerenes in the solid state. These molecules are spherical or near-spherical in shape. Molecules with high point-group symmetry, which are not bound strongly in the solid state, tend to crystallize into structures with long-range periodicity of the molecular centers of mass, but the molecular orientations are random or even dynamically disordered. When dynamically disordered, the

molecules rotate about some preferred axis. C60 and C70 satisfy the criteria for such orientationally disordered solids and exhibit rich phase behavior in the solid state. Since C60 has high electron affinity, it forms anion salts with alkali and alkaline-earth metals as well as with strong organic donor molecules. With tetrakis dimethylaminoethylene (TDAE), which is a very powerful electron donor, C60 forms a 1:1 solid that is ferromagnetic. C60-TDAE is the molecular organic ferromagnet with the highest Tc (of 16 K) known to date. Some of the alkali and alkaline-earth fullerides, on the other hand, show superconductivity, with transition temperatures going up to 33K. We shall briefly examine some of these solid-state properties.

One of the many wonders that fullerenes have brought to us during the past few years is the variety of their shapes. When the elusive C60 finally showed up in 1990, the perfect symmetry and astounding beauty of its molecular structure touched the hearts of scientists before they could consider the molecule's vast technical possibilities. Already much has been said about the unique shape of C60 and its potentialities. C70 and higher fullerenes have simultaneously been found in the same soot that produced C60 and were quickly revealed to be shaped like rugby balls or oblong eggs. Hence we were aware that there had to be an extensive series of roundish polyhedral clusters of carbon atoms.

Then, in the following year, multilayered tubular fullerenes (Figures 1a and 1b) were discovered by Iijima and were named buckytubes (see the article by Iijima in this issue). Iijima also observed similarly huge and multilayered carbon balls, before C60 was discovered. Soon after, buckyonions were recognized as an important class of fullerene (Figure 1c, see article by Ugarte in this issue). So, in the early days of fullerene research, we already knew three forms of fullerene: sphere, tube, and particle. At that time, however, nobody anticipated that this was only the beginning of a big show of stunning variations in the shapes of fullerenes. This article introduces current developments in the study of these fullerene styles.

Pure carbon materials, graphite and diamond, possess a wide array of interesting physical properties, and so attract a large spectra of interests and applications. Carbon microparticles (carbon black) and carbon fibers are widely used in practical applications including common materials (paints, inks, polymers, etc.) and high-performance composite materials.

Carbon displays a remarkably rich and complex chemical behavior (three different possible hybridizations: sp1, sp2, and sp3). In particular, the covalent carboncarbon bond is one of the strongest in nature, and induces a high melting temperature (> 4000°C). The phase changes associated with unusually high temperatures and pressures as revealed in the carbon phase diagram, and the fact that the solid sublimates at low pressures before melting, lead to many experimental difficulties in the study of high-temperature properties of carbon materials. Experiments must therefore rely on transient melting, for example, laser vaporization or arc-discharge heating. This explains why fullerenes and related graphitic structures have only recently been discovered.

From a fundamental point of view, the discovery of fullerenes has introduced new ideas about how carbon atoms bond. The curvature and closure of graphitic surfaces has become a standard concept in carbon chemistry, and recently a wide range of structures formed by curved graphitic networks has been observed. A surprising aspect of fullerene research is that these novel graphitic structures were found in well-known experiments, and that they had been overlooked for so many years.

This article will describe recent progress in the generation and physical characterization of graphitic nanoparticles, or multishell fullerenes. The lack of an efficient method for producing, as well as a method for purifying these particles makes it difficult to characterize them and to develop possible applications.

Nanostructured materials have recently attracted the attention of some materials scientists. Because of their unique properties occurring in low-dimensional structures, nanostructured materials are sought for their possible industrial applications. This article introduces a specific nanostructured material, the carbon nanorube—an extremely thin filaments of graphite considered to be a quasi one-dimensional structure, with a simple well-understood atomic structure. Because of these qualities, the carbon nanorube has elicited great interest from diverse fields of basic and technological research. My discovery of carbon nanotubes was inspired by the discovery of C60 and its family and their mass production. The carbon nanotubes were serendipitously found during the examination of fullerene materials by a high-resolution transmission electron microscope (HRTEM). Since introducing this technique in 1971, I have been employing HRTEM to characterize the microscopic structural details of a variety of materials, including carbonaceous materials. So far, only nanotubes have been revealed with HRTEM.

Interest in the carbon nanorube is multifold. Academically the nanotube is an ideal model structure for a quasi one-dimensional structure since its known atomic structure makes computer simulations more reliable. It is worthwhile to study both rare structures of graphite—cylindrical forms with a helical arrangement of carbon atom hexagons and flexible graphitic sheets containing topological surface defects. These materials may find practical uses as tough graphite fibers, molecular wires, catalyst supports, molecular adsorbers, and so on.

The discovery that C60 Buckminsterfullerene may be created spontaneously in high yield when carbon vapor condenses indicates that graphene sheet curvature and closure is a common occurrence during carbon nucleation to form extended networks. As a consequence, a net microscopic perspective on graphitelike carbonaceous materials has evolved. This perspective is summarized here because the net observations relate to various types of nonplanar graphitic structures that promise to be useful as viable nanoscale engineering materials.

C60 Buckminsterfullerene was discovered in 1985 among the products of chemical nucleation of carbon in the gas phase. The faint whisper, embedded in the helium wind that blew a laser-initiated carbon plasma into a mass spectrometer, was interpreted as the possible signature of the elegant soccer-ball-shaped molecule. That noticed signal, which indicated that a pure 60-carbon-atom molecule might be very stable, was quite unexpected and thus a truly serendipitous discovery. The original aim of the experiments was to simulate the conditions in a red giant carbon star and particularly to probe an earlier proposal that long carbon-chain molecules might have originated in such celestial objects and might have been subsequently ejected into interstellar space. The carbon chains themselves had been discovered to be abundant in certain regions of the interstellar medium by a radioastronomy search program in the late 1970s. The molecular rotational frequencies, which formed the basis for the radio search, resulted from a spectroscopic study of cyanopolyynes that had been synthesized in the laboratory.