Effective work functions for ionic and electronic emissions from mono- and polycrystalline surfaces

The effective work functions (phi(+), phi(e) and phi(-)) for positive-ionic, electronic, and negative-ionic emissions from mono- and polycrystalline surfaces are surveyed comprehensively and also investigated critically for the main purposes of (1) evaluating the most probable values of phi(+), phi(e) and phi(-) for a variety of surface species, (2) explicating both thermionic contrasts (Delta phi* equivalent to phi(+) - phi(e) and Delta phi** equivalent to phi(-) - phi(e)) and their dependence on experimental conditions, and (3) demonstrating the necessity of employing phi(+) (not phi(e)) for quantitative analysis of those data on positive ion emission from polycrystalline surfaces. Careful examination of both theoretical results and experimental data on the work functions yield several conclusions. By both theory and experiment, clean monocrystalline surfaces are verified to have Delta phi* = 0.0 eV within an error of +/- 0.05 eV. Next, as the density of local surface irregularities increases, the homogeneity in the work function over the whole surface area decreases and, hence, Delta phi* increases. Also, the most probable values of phi(+) and phi(e) are recommended for many mono- and polycrystalline surfaces, mostly (similar to 70%) with a standard deviation of +/- 0.02-0.08 eV. Compared with the probable or typical values of phi(e) accepted in influential handbooks, the most probable values of phi(e) recommended here are typically (similar to 70%) equal to each other within a narrow gap of less than similar to 0.1 eV, but some (similar to 20%) are different by similar to 0.2 eV or more (up to similar to 1 eV). Furthermore, polycrystalline surfaces of Nb, Mo, Ta, W, Re, Ir, Pt, etc. hold Delta phi* approximate to 0.3-0.8 eV since each surface has a mean value that is different between phi(+) and phi(e). Also, at the degree of monocrystallization (delta(m)) below similar to 50%, the theoretical value of Delta phi* depends little on delta(m) and agrees well with experimental data on each polycrystalline surface. As delta(m) increases beyond similar to 80%, Delta phi* decreases rapidly to 0, showing again a good agreement between theory and experiment. In particular, those surfaces of delta(m) > 97% generally have Delta phi* approximate to 0 within the uncertainty of about +/- 0.05 eV, which is apparently equivalent to the usually called monocrystalline surfaces (delta(m) = 100%). Additionally, even when both phi(+) and phi(e) are changed by up to similar to 1 eV by gas adsorption, Delta phi* itself remains little changed and, thus, the so-called work function (phi) recommended with polycrystalline surfaces in handbooks should not be cited as phi(+) since phi usually coincides with phi(e) except where otherwise stated. In the case of polycrystalline surfaces, phi(+) instead of phi(e) should always be adopted to analyze accurately data on any positive ion emission, irrespective of its process or mechanism. Also, those metals covered with a two-dimensional graphitic film usually have phi(+) approximate to phi(e) approximate to 4.5 eV, which corresponds to monocrystal graphite. Finally, for any species of mono- and polycrystalline surfaces, both theory and experiment verify phi(-) = phi(e) and hence, Delta phi** = 0. The featres of dissociative self-surface ionization of heated ionic crystals are outlined together with typical data on phi(+), phi(-) and phi(e), which originate from the thermionic properties of the crystal itself. A brief description is given to typical methods and techniques to prepare clean and/or monocrystalline surfaces, to determine local work functions of real monocrystalline surfaces, and also to form graphitic carbon films on various surfaces. In 12 tables and 29 figures based on 1350 references published to date (mainly similar to 1970-2006), we show data on each work function of mono- and polycrystalline surfaces and their temperature coefficient, as well as their dependence upon experimental conditions. Also, we illustrate a comparison of each work function between theory and experiment and the most probable values of phi(+) and which are generally citable as reliable references. A comparison between the most probable values (phi(e)) recommended here and the probable or typical ones (phi) accepted elsewhere are shown, along with working conditions for keeping phi(+) as high as possible for promoting positive ionization efficiency. Also, we present relationships between phi(+) and ionic desorption energies, and typical data on negative ion emission due to thermal stimulation. Thus, we provide an extensive and up-to-date database of the effective work functions of both mono- and polycrystalline surfaces, and also summarize their peculiarities governing the emissions of positive and negative ions and electrons. (C) 2007 Elsevier Ltd. All rights reserved.