R. Baudoing

Theory of angular dependence in electron emission from surfaces: Application to Auger emission from several metal surfaces

Abstract An exact multiple scattering description of electron emission from inner sources is given using the same matrix formalism as applied in LEED theory, and some approximations are discussed. The model is derived for sources localized on atomic sites and represented in terms of spherical waves but it may be adapted for other types of inner sources. This model is used to study the anisotropy of the Auger emission from aluminum and some transition metals (Fe, Ni, Cu). The results obtained show that: (a) The intra-layer multiple scattering term dominates the calculated angular dependence. (b) The anisotropy due to multiple scattering is very important but too strong compared to experiment; this may be related to the over simplified model used for the source wave function, (c) The isotropic emitter does not account for experimental results (d) The intrinsic anisotropy of the emitter is of first importance in the observed angular dependence. Thus we conclude that not only a full multiple scattering description must be used, but also a realistic description of the Auger source is needed in particular to give account of the experimental behaviour of aluminum and of transition metals.

Angular resolved Auger emission from crystal surfaces: New experimental results and predictions from a simple model

Abstract In a first part, new angular resolved CVV Auger emission results (including line shape after background substraction and anisotropy measurements) are reported concerning the Al L 23 VV transition on clean Al(001) and Al(110), the Ni M 23 VV transition on clean Ni(001), and the Ni M 23 VV and S L 23 VV on the Ni(001)-(11,11)-S system. In a second part we discuss a semi-quantitative analysis of the Auger emission anisotropy with a model which includes at the same time the final state effects due to multiple elastic scattering of the ejected Auger electron on its escape path and a description of the Auger source in solids as realistic as possible. The band aspect of the valence states id described as s, p, d, … like contributions at the emitters lattice site, and we use atomic results for the matrix element angular momentum dependence. In a manner consistent with this model we have analyzed the available anisotropy measurements on Al(001), Al(110), Cu(001), Cu(110), Cu(111), Fe(001) and Ni(001)-(11,11)-S. The results are consistent for various faces of the same material and correspond, moreover, to the model predictions, indicating that the most important aspects of the Auger emission anisotropy are included. In a third part, on the Ni(001)-(11,11)-S system, we investigate the sensitivity of the emission anisotropy of adsorbate Auger transitions to structural parameters such as the adsorption site coordinance and the adsorbate-substrate distance. We find a large sensitivity in both cases, comparable to LEED. Moreover, satisfactory agreement is found only when the structural parameters are given the values known from LEED studies and when the angular momentum predicted by the above model is used for the ejected Auger electron.

Crystallography by LEED: IV. Application of the iso-intensity map method to the Ni(100)-(11, 11)-S system☆☆☆

Abstract In the present study, we illustrate in the case of the Ni(001) and Ni(001)-(1 1 , 11)-S surfaces, the application of a procedure of surface analysis by LEED using iso-intensity maps for the specular beam together with a very fast LEED program based on the RFS method. This procedure has several advantages: first, iso-intensity maps prevent one from leaving out important experimental information. Second, the fast LEED program allows one to study a large number of beams previously selected from the maps for their particular interest, thus avoiding finding a wrong result, as could hapen if one looks only at a few beams. It comes out from this study that the adsorption distance d = 1.3 A of the sulfur atoms is confirmed with the help of a cluster type superposition potential for the sulfur layer. The agreement between experiment and theory is generally excellent, except for a very strong experimental peak which is not well reproduced by calculation, whatever the changes in the physical or structural parameters. This has not been reported previously and is similar to what has been found by the same authors for the aluminium (001) surface.

Low-energy electron reflection and transmission on Ar films condensed on polycrystalline platinum

Abstract Low-energy electron reflection (LEER) and transmission (LEET) spectra have been recorded in the range 1–16 eV for argon films grown on polycrystalline platinum. The data are compared with low-energy electron diffraction (LEED) calculations of the elastic reflection and transmission of an electron beam incident on the (111) surface of a semi-infinite Ar solid azimuthally disordered. Structures in LEER spectra reflect essentially the band structure characteristics of the Ar(111) surface, whereas in transmission, similar structures are strongly modified by scattering from phonons, defects and imperfections of the film as evidenced by changes in LEET spectral line shapes as a function of incidence angle and film thickness. These findings are discussed in relation with the ability of the LEED formalism to describe elastic or quasi-elastic scattering from partially disordered films at very low energies.

A LEED data reduction method: Description and application to the relaxation of aluminum (001) and (110) surfaces

Abstract A LEED data reduction method is described and applied to Al (001) surface. The method consists of averaging rotation diagrams (azimuthal intensity profiles) obtained at constant electron energy between 500 and 1000 eV in the whole range of the colatitude angle of incidence. The method avoids certain difficulties, stemming e.g. from the energy dependence of the mean optical potential, that are present in other methods in which data obtained at different energies are averaged. The averaged data are analyzed using the two-beam Darwin theory. Values for the surface normal relaxation and mean optical potential are obtained from the study of the position, width and profiles of the averaged peaks. The (100) surface of aluminum is found not relaxed while the (110) surface is contracted. The values found for the mean optical potential are similar to those obtained from other sources. The precision of the averaging method for surface structure is not better than that of direct calculation. Experimental factors limiting the precision of the method (surface roughness, background interference due to Kikuchi lines) are specified.

Medium energy electron diffraction on aluminum

Abstract Measurements of the specular beam intensity in the medium range of energy (645 eV and 980 eV) are presented for the aluminum (001) and (110) surfaces. They have been performed with a LEED goniometer allowing to achieve high accuracy in the determination of the angular parameters of the diffraction experiment. The experimental results are displayed in the form of iso-intensity maps in (θ, φ) coordinates at a constant primary beam energy. θ denotes the colatitude angle of incidence and is varied in the range 30° to 80°. φ denotes the azimuth and is varied between 0° and 360°. Sections of the maps at constant angle of incidence are known as rotation diagrams. A model calculation ignoring intra-layer multiple scattering, but incorporating inter-layer multiple scattering processes shows the prominent role of the latter, in many cases, and the non-negligible contribution of the former in other cases. Sections of the maps at constant azimuth are called incidence-profiles. They are very sensitive to the azimuth and very hard to compute. Using the preceding model calculation, average incidence profiles have been computed, using as input parameters the refractive and absorptive potentials and the surface normal relaxation parameter obtained by a pseudo-kinematic analysis of experimental average incidence profiles as explained in another paper. The peak position and width have been compared to those obtained from the pseudo-kinematic calculation. From the result, it is shown that if averaging is performed according to a specific set of restrictive rules, the pseudokinematic analysis of the data has a semi-quantitative meaning.

Electronic surface resonances of Ni(001) and Ni(001)c(2 × 2)S observed by leed

The electronic surface resonance band structure E(k∥) (E = resonance energy, k∥ = reduced parallel momentum) is determined for Ni(001) and Ni(001) c(2 × 2)S surfaces from low-energy electron diffraction (LEED) intensity measurements. The LEED data consist of analog plots of 00 intensity versus accelerating potential up to 40 V for colatitude incidence angles ranging from 5° to 80° and for both the 01 and 11 azimuths. Features less than 1.5 eV wide are assigned to resonances. Sequences of resonances corresponding to the 11, 01 and 01 branches of the free electron band structure for Ni(001) and to the 10, 01, 10, 12 12, 12 12 and 1212 branches for Ni(001) c(2 × 2)S are found. For Ni(001) the observed surface resonance band structure is fitted accurately (±0.3 eV) by a two-dimensional (2-D) free electron formula with a resonance binding energy of 3.5 eV. For Ni(001) c(2 × 2)S most branches of the band structure conform approximately to the 2-D free electron description, but the apparent binding energies vary from one branch to another in the range 3–5 eV. Comparisons with related mehods and with other results obtained using Ni(001) substrate are given. Some improvements of the LEED method are suggested.

Angular resolved auger emission from clean and sulfur covered Ni(110) surface

The angular dependence of the nickel M 23 VV and of the sulfur L 23 VV Auger transitions are studied in detail, on clean and sulfur covered Ni(110) surfaces. New experimental data are presented for the anisotropy of both transitions as a function of polar and azimuthal angles of emission. Our model, which incorporates at the same time the multiple scattering effects in the final state wave function and the intrinsic anisotropy of the Auger emitter, is found to give a satisfactory account of the observed auger anisotropy. We find a large sensitivity to the position of the sulfur adsorbed atoms. The best agreement is obtained for the hollow site, slightly less than 0.9 A above the top nickel layer. This conclusion is consistent with previous LEED and MEIS studies, but does not agree with the long bridge site obtained from quantum chemistry calculations. Moreover the sulfur emitter on this particular Ni(110) face appears to have an intrinsic anisotropy.

Observation of LEED threshold effects on polycrystalline platinum

Abstract We report the observation of LEED threshold interference structure on polycrystalline samples of platinum. These structures, observed at low energy by high-resolution electron reflection measurements, are interpreted in terms of a recently developed theoretical framework. Our analysis shows how observation and detection of threshold effects is possible even on polycrystalline samples under certain conditions. These conditions are described with emphasis on the use of threshold effects for studying polycrystalline samples.

Theory of angular dependence in electron emission from surfaces: Application to secondary emission from Al(001) and Al(110)☆

The model of electron emission derived tor Auger emission in a previous paper, is extended to the study of angular resolved secondary emission and applied to the description of experimental results obtained on Al(001) and Al(110). We include exactly all the final state effects, and we give a simple description for the source wavefield which is taken as a set of plane waves created incoherently at a given depth and with any possible wavevector coupled to the direction of observation. Having no specific model for the cascade processes at the origin of the source wavefield, we have taken a uniform distribution of sources. The experimental data have been taken at a primary energy at 350 eV and at normal incidence, such that the anisotropy depends no longer on the incidence conditions. On one hand the polar angle plots are dominated by a cos2θ like dependence; on the other hand the modulations on polar or azimuthal plots are found less than 10%. Both effects are well accounted for by our model. As far as the correlation of experimental and calculated fine structures is concerned, a close examination of the individual contributions from deeper and deeper crystal slices suggests a non-uniform distribution of sources normal to the surface, the top and first slice having a small contribution.