Low-energy electron diffraction (LEED) is a widely used technique for the study of solid crystal surfaces and ultrathin films. It provides information about three properties of the crystal structure in the surface region, namely, the symmetry, the periodicity and the atomic arrangement. The first two are easily obtainable by mere inspection of the LEED pattern. Thus, even if only approximate normal-incidence conditions are established, the LEED pattern from, say, a face-centered-cubic {001} or {111} surface will clearly exhibit four-fold or three-fold symmetry, respectively, both in the positions and in the intensities of the diffracted beams. In addition, comparison of beam positions with the angles calculated from the grating formula will easily indicate whether the surface structure is bulk-like (so-called 1x1) or reconstructed (2x1 or c(2x2) or 3x3 or others).
So far, the use of LEED in surface science has been predominantly limited to gathering these two pieces of information: the symmetry and the periodicity of the surface structure. A third, equal if not more important, piece of information provided by a LEED pattern lies in the intensities of the diffracted beams. It is generally recognized that these intensities are related to, and therefore contain information about, the arrangement of atoms in the surface unit cell. Unfortunately, only a minority of workers has paid attention to these intensities and used them in order to determine the atomic arrangement. The reason is that quantitative LEED crystallography requires precise orientation of the sample, collection of large sets of intensity data and subsequent calculations of diffracted intensities by computer programs based on full-dynamical theories|all demanding and time-consuming tasks.
But there is another use of diffracted LEED intensities which is not time-c onsuming and can be very profitable, and which has not yet been universally recognized, namely, the use of intensities for surface characterization. Indeed, a surface structure is fully characterized, even if not yet known, by all three properties of its LEED pattern|symmetry, periodicity and intensity. For example, almost all {001} surfaces of the face- and body-centered-cubic metals have four-fold symmetry, and some of them have very similar periodicities (e.g., Ag and unreconstructed Au), but they produce very different LEED intensities.
For purposes of surface-structure analysis people collect curves of diffracted intensities versus incident electron energy [so-called I(V) curves or I(V) spectra] for a large number of beams, but for purposes of surface-structure characterization it is not necessary to collect large sets of intensity data or to carry out elaborate intensity calculations. In any LEED experiment, normal incidence can be established quite rapidly and with reasonable accuracy by visual means, and I(V) curves for one or two beams can be recorded in minutes with a spot-photometer or any of the video-LEED systems now available in many laboratories around the world. Such curves constitute a unique fingerprint of the surface under scrutiny. Even if spot- photometers or video-LEED systems are not available, mere visual recording of intensity maxima in a few diffracted beams and of the electron energies at which they occur is useful for characterization purposes.
Some examples of the importance of such characterization can be found in the literature. When growing epitaxial films of Fe on Cu{001} mere inspection of the LEED pattern does not allow one to decide whether the structure of the film is the stable bcc or the metastable fcc, because both structures produce four-fold-symmetric 1x1 LEED patterns. However, the corresponding I(V) curves are very different from one another and can therefore be used to make the distinction. In general, experimenters growing epitaxial films tend to consider the persistence of a 1x1 LEED pattern after deposition of one or two layer-equivalents as a proof of layer-by-layer growth. However, a 1x1 pattern may also be observed when the film growth occurs by islands and the LEED signal stems predominantly from bare areas of t e substrate. Comparison of the I(V) curves with those of the clean substrate makes it easy to distinguish between the two modes of growth. In other areas of surface science, e.g., the preparation of atomically clean surfaces of the rare-earth metals, it has been noticed that Fe impurities segregated onto the surface from the bulk during annealing treatments produce LEED patterns with the same symmetry and the same periodicity as the clean surface, but with completely different I(V) spectra. Thus, knowledge of the I(V) spectra of the clean surface is very useful.
One objection that has been made against routine use of I(V) spectra in order to characterize surface structures is that only a few people have the means and the expertise to calculate the I(V) spectra that may be expected in any particular study and that the experimental I(V) spectra needed in the study are unavailable or scattered throughout the literature and therefore difficult to find. The present collection of I(V) spectra is aimed at overcoming that objection, and is intended both as an encouragement for people to use LEED intensities for more complete characterization of surface structures and as a tool for doing so. This collection includes I(V) spectra from most clean metal surfaces studie by LEED to date, and the I(V) spectra from some clean and absorbate-covered semiconductor and alloy urfaces. The spectra are those of the low-index beams (10, 11, 20, etc...) at normal incidence and, whenver available, those of the specular beam (00) at specified values of the polar (theta) and the azimuthal (phi) agle of incidence. To avoid confusions about the indexing of LEED beams we have included a series of sketchs of schematic LEED patterns for all surfaces represented in the collection and have indicated the corresponding indices. The I(V) spectra are presented in plots of intensity (in arbitrary units) versus voltage (in Volts) on scales of either 20 of 40 Volt/inch, and are arranged in alphabetical order of the chemical symbol of the crystal involved. The collection is preceded by a table listing all surfaces included and the references to the sources of the data. The I(V) spectra that were not collected in the writers' laboratory were digitized from the journals' articles in which they were published. In some cases the indexing has been changed to be consistent with that of similar surfaces of other crystals represented in this collection.
We are aware of the fact that at the time of writing only a few workers in the areas of surface science and ultrathin film studies recognize the usefulness of and the need for LEED I(V) spectra. But we believe that, given the almost universal use of LEED in surface studies, such a recognition is unavoidable, particularly by those who study epitaxial growth of ultrathin films of stable and metastable phases and may not have the facilities or the inclination to do full-dynamical calculations of LEED intensities. The present collection of experimental I(V) curves will then be a useful reference for surface characterization in a variety of surface- sensitive experiments.
Please note, that this data can be sent to you in ASCII format. We plan to post it for direct access.
If you want to see more, then drop us a line.
10/28/98.
And finally, our disclaimer.