Most materials interact with the world through their surface, which is why physicists and material scientists are so interested in the structural and electronic properties of surfaces. For many years, however, surface scientists have been unable to understand why the experimental and theoretical values for the lifetime of "hole" states on some surfaces can differ by factor of two or more. Now Richard Berndt at the University of Kiel in Germany, and co-workers in Aachen, also in Germany, San Sebastian in Spain and Bath in the UK, have shown that the anomaly is due to the presence of defects on the surface and the competing effects of two- and three-dimensional processes on the lifetime of the holes (J Kliewer et al. 2000 Science 288 1399).
Berndt and co-workers prepared gold, silver and copper surfaces with standard techniques and then used a scanning tunnelling microscope with a specially prepared tungsten tip to identify regions on the surface that were free from defects. The advantage of this technique over others, such as photoelectron spectroscopy, is that the same instrument that is used to check the surface can also measure the lifetime of the “holes”. (These holes are left behind on metal surfaces when electrons are excited to higher energies.) The lifetime of the holes can be determined by carefully measuring how the current through the tip changes as the voltage difference between the tip and the surface is varied.
However, the theory also had to be modified to resolve the discrepancy with experiment – which ranged from a factor of two in copper, four in gold and seven in silver. Berndt and co-workers extended previous treatments to calculate in detail how the hole lifetime is determined by electron-electron interactions and electron-phonon coupling. (Phonons are quantized vibrations of the metal lattice). They find that the lifetime is essentially determined by the two-dimensional electron gas near the surface, but that the interactions between these electrons are modified by the three-dimensional electron gas of the bulk crystal.
