The adsorption of Na on Al(l 11) at room temperature has been studied by surface extended x-rayabsorption fine-structure (SEXAFS) experiments as well as by parameter-free calculations. For coverages of BNU^O. 16-0.33, the SEXAFS analysis shows that Na atoms occupy an unusual sixfoldcoordinated substitutional site. The Na-Al bond length is determined as 3.31 A, consistent with metallic bonding. Ab initio density-functional-theory calculations for several adsorbate geometries show that the substitutional site has the lowest total energy. PACS numbers: 78.70.Dm Studies of alkali-metal (AM) adsorption have played an important role in the development of theories of chemisorption. The model for AM adsorption first proposed by Gurney [1], in which the nature of the adsorbate bonding is determined by the energy position and width of a resonance derived from the alkali valence s level, has strongly influenced subsequent theoretical and phenomenological descriptions of adsorption. Much of the interest has centered on the large change in work function, and phenomena related hereto, and on the characteristic dependence of the work function on coverage, found for many metals and semiconductors. Recently, more detailed insight into the nature of the adsorption bond has been obtained from experimental studies [2] using photoemission, inverse photoemission, electron energy loss, and Penning spectroscopies. These studies appear to confirm the main features of the Gurney model in which the resonance derived from the valence s level of the adsorbed atom is mostly unoccupied at low coverage, but moves down in energy through the Fermi level with increasing coverage due to adsorbate-adsorbate interactions. The interpretation that the bonding is largely ionic at low coverages [3] is, however, the subject of considerable current controversy: An alternative description in terms of the dominance of adatom polarization has been proposed [4,5].With a few exceptions noted below, studies of the adsorbate geometry have not played a central role in discussions of the nature of the adsorption mechanism and very few quantitative determinations of the adatom geometry have in fact been carried out. The recent emphasis of structural studies has rather been, on the one hand, the characterization of the variety of adsorbate phases [6] as a function of temperature and coverage and, on the other hand, studies of the reconstruction [7,8] of fee (110) and bec (100) metal surfaces induced by small concentrations of adsorbed AM's. Thus, theoretical models of AM adsorption have almost by default assumed that the adsorption site is one of high symmetry on an unperturbed substrate. In accord with this assumption, self-consistent calculations [5,9,10] for Na adsorption on Al(lll) arrived at the conclusion that Na occupies the threefoldcoordinated site. LEED structure determinations [6,11,12] for Ni(100)-c(2x2)-Na, Al(100)-c(2x2)-Na, and Rh(100)-c(4x2)-Cs indicate that for these systems the AM atoms occupy the fourfold sites of highest symmetry. The Cu(ll l)...

No abstract

Simple, rapid and nondestructive measurements of magnetic properties (magnetic susceptibility, saturation isothermal remanent magnetization and coercivity of isothermal remanence) coupled with more time consuming and sophisticated analyses such as thermomagnetic and Mbssbauer effect studies, provide a basis for identifying forms of iron oxide present in the soil.A t two of the sites studied, Caldy Hill, Merseyside, and Llyn Bychan. N. Wales, the secondary ferrimagnetic oxide formed in surface soil as a result of recent forest fires is shown to be non-stoichiometric magnetite approximating to the formula Fe2.904. No evidence for the presence of maghemite was found in any of the soil samples from these sites or in so% from the AMWY region of S.E. France. IntroductionNUMEROUS studies record the formation of secondary ferrimagnetic iron oxides in soils developed on sedimentary rocks. In a recent review article, Mullins (1977) summarises evidence for the origin, nature and composition of the oxides so formed and concludes that several possible mechanisms are involved, including burning, dehydration of lepidocrocite, and the alternate wetting and drying normally involved in pedogenesis. He cites several authors who have used a variety of techniques to identify the ferrimagnetic minerals in soils but recommends caution in evaluating the conclusions reached (p. 234). Although Neumeister andPeschel (1968), andVadyunina andKovtun (1974) claim to have identified magnetite, the majority of authors, whether on the basis of experimental methods designed to simulate the pedogenic processes inferred (e.g. Mullins, 1971, Taylor andSchwertmann, 1974), X-Ray diffraction studies or, more recently, Mbssbauer effect

No abstract

No abstract