Tim Reid

Carbon nanotubes: sorted by DNA.

and George Hull, then at Bell Laboratories, measured the thermal conductivity of various isotopic compositions of bulk germanium and found them to be significantly different3. More recently, the end of the Cold War resulted in a number of laboratories in the former Soviet Union making a variety of stable isotopes — including isotopes of germanium, silicon, gallium and many other semiconductors — available for peaceful research purposes. These isotopes have been used to explore various aspects of state-of-the-art silicon CMOS device fabrication (for example, 30Si has been used to probe the time dependence of chemical reactions and impurity diffusion at interfaces). In parallel, much effort has been made to identify the effect of isotopic composition and randomness on a variety of electrical, magnetic, thermal and optical properties of semiconductors4,5. However, despite their scientific importance, isotope effects in silicon and germanium have had very little impact on the development of new semiconductor products. Isotope effects in diamond, on the other hand, may be different. Diamond is well known for being ‘number one’ for properties such as hardness, thermal conductivity, bandwidth for optical transmission and chemical stability. Even though it is considered to be an insulator, recent advances in doping and lithography technology have made diamond an attractive material for the fabrication of high-power semiconductor devices. Thanks to its wide bandgap of ~5.5 eV (compared with 1.1 eV for silicon and 1.43 eV for gallium arsenide), its various optical emissions are observable at room temperature, and a number of groups around the world are exploring the possible use of diamond-based devices for applications in quantum information. Isotope engineering of bulk diamond was independently pioneered by Hisao Kanda of the National Institute for Research in Inorganic Materials and colleagues in Japan6 and Tom Anthony of General Electric and co-workers in the US7. However, atomic layer-by-layer stacking of 12C and 13C isotopes has been extremely challenging because the microwave plasma-assisted chemical vapour deposition process that is needed to obtain high-quality diamond films involves simultaneous etching and growth processes at the substrate surface. Watanabe and co-workers successfully optimized the competition between etching and growth to perform atomic-layer-level control of the growth of 12C and 13C diamond films for the first time. Watanabe and co-workers confirmed that electrons and holes were preferentially confined in the 12C layers of the 12C/13C structures at temperatures of 77 K by detecting electroluminescence. The charge carriers were confined in the 12C layers because the bandgap of 12C is 17 meV smaller than that of 13C. This is the first example of isotope engineering of a semiconductor’s electronic and optical properties that might have a major impact on real-world applications, such as solidstate lasers composed exclusively of diamond that operate at extreme ultraviolet wavelengths (Fig. 1). An important challenge for the future is to work with the 17 meV difference between the bandgaps of 12C and 13C: although this is large for an isotope effect, it is still too small for room-temperature operations. However, by combining isotope engineering in diamond with another wide-bandgap semiconductor, such as silicon carbide, it may be possible to develop electronic and optical devices that can work at room temperature. ❐