Ball-and-stick representation of a 1.4-nm diamond with 275 atoms.

Representation of a classic 60-atom buckyball.

Are Diamonds Really Forever?

Nanodiamonds—particles of diamond dust measured in nanometers—have been found in meteorites, protoplanetary nebulae, and interplanetary dust. On Earth, nanodiamonds can be produced by high-explosive detonation and chemical vapor deposition. Interest in laboratory-produced nanodiamonds has grown recently in proportion to the number of technological applications that can be envisioned for them. As a light-emitting semiconductor, nanodiamonds may be used in optoelectronic components that process both light and electricity. Diamond's hardness and fracture strength make it attractive for miniaturized mechanical devices such as cantilevers and gears. Diamond "nanodots" and "nanofilms," composed solely of carbon atoms, would be compatible with biological materials and may be used for neuronal imaging, integrating biological systems with electronic devices, or detecting biological or chemical warfare agents. All of these applications, however, depend on scientists' understanding of diamond's behavior at the nanoscale. In this research, Raty et al. report that, in fact, diamonds don't last forever: they begin to morph into buckyballs (at least on the surface) if broken up into small enough pieces. Nevertheless, one thing remains true: diamond is a remarkable material whose properties continue to surprise us.

Diamond, like silicon and germanium, is a semiconductor whose behavior depends on the size of its optical gap—the energy difference between its valence and conduction bands. At the nanometer scale, quantum confinement effects on electrons and holes typically result in a widening of the optical gap, which causes a blueshift in the material's absorption and emission peaks. Thus, a potentially useful property of nanosemiconductor particles is their ability to emit light at different wavelengths depending on particle size. Previous studies have demonstrated this effect for silicon and germanium particles as large as 7 nm in diameter. The Livermore group extended these studies to nanodiamonds and found them to be unusual in both their electronic and geometric structure.

Using density functional theory and quantum Monte Carlo calculations, the researchers modeled the electronic structure of nanodiamond clusters. Their results indicated that there is no appreciable quantum confinement effect on the optical gap of nanodiamonds until they get down to about 2 nm in diameter, contrary to what is found for silicon and germanium using the same theoretical tools. Geometrically, the researchers found that as a nanodiamond gets smaller, its structure becomes unstable. Quantum simulations revealed graphitization of the first atomic layer, followed by the formation of pentagons linking the graphene fragments with the atoms underneath. This provided further curvature to the surface, which eventually adopted a geometry similar to that of half a buckyball.

Ball-and-stick representation of bucky diamond cluster with 275 atoms, 1.4 nm in diameter, showing diamond core (yellow) and a fullerenelike reconstructed surface (red).

The results of the calculations and simulations were consistent with x-ray emission and absorption spectra taken at ALS Beamline 8.0.1 and Stanford Synchrotron Radiation Laboratory. Emission and absorption spectroscopy together reveal the optical gap in semiconductors, with emission revealing the valence band maximum and absorption revealing the conduction band minimum. The techniques also reflect the density of states around the bandgap—a sensitive fingerprint of atomic bonding configurations.

Valence and conduction spectra of nanodiamond clusters compared to those of bulk diamond and graphite, as obtained using x-ray emission and absorption techniques, respectively. The energy scale of the absorption spectra was calibrated to the π resonance of highly oriented pyrolytic graphite, set to 285.38 eV.

The sample nanodiamond sizes (4 ± 1 nm) and crystallinity were verified by electron diffraction and high-resolution transmission electron microscopy. The samples were also heated and cooled several times to remove impurities. The carbon K-edge emission and absorption spectra showed the same basic features as bulk diamond and graphite. In particular, the researchers observed no blueshift in the position of the nanodiamond valence and conduction band edges in comparison to those of bulk diamond. Furthermore, the nanodiamond absorption spectra showed a pre-edge peak at 286.7 eV not observed in the bulk. Comparison to the density of unoccupied states derived for bucky diamond suggests that the feature is the signature of the pentagonal and hexagonal bonding configurations found on buckyball-like surfaces.

Pre-edge peaks in nanodiamond absorption spectra are compatible with the calculated signatures (dashed line) of a mixture of pentagons and hexagons found on buckyball-like surfaces.

Interest in and funding for nanoscience has increased dramatically in recent years, reflecting the field's enormous potential for theoretical breakthroughs as well as practical applications. Studies such as this one are a necessary first step toward understanding the molecular building blocks that will be utilized in the nanotechnologies of the future.

Research conducted by J.-Y. Raty (Lawrence Livermore National Laboratory and University of Liege, Belgium) and Giulia Galli, C. Bostedt, T.W. van Buuren, and L.J. Terminello (Lawrence Livermore National Laboratory).

Research funding: U.S. Department of Energy, Lawrence Livermore National Laboratory, and Fonds National de la Recherche Scientifique. Operation of the ALS is supported by U.S. Department of Energy, Office of Basic Energy Sciences.

Publication about this research: J.-Y. Raty, G. Galli, C. Bostedt, T.W. van Buuren, and L.J. Terminello, "Quantum Confinement and Fullerenelike Surface Reconstructions in Nanodiamonds," Phys. Rev. Lett. 90, 037401 (2003).