Attenuation of Low-Energy Electrons by Solids: Results from X-Ray Photoelectron Spectroscopy

Surface and bulk components in angle-resolved core-level photoemission spectroscopy of graphite

“…Due to the nonmonotonic variation of the bulk signal values for the IMFP of the C 1s photoelectrons vary between ϳ13 and 26 Å, with the mean value of 16Ϯ 5 Å in good agreement with the measurements of IMFP in carbon by Steinhardt et al 11 and Zemek et al 12 In their analysis of a smaller ARXPS data set Speranza and Minati 4 drew somewhat different conclusions about the nature of the surface and bulk components, finding a surface core-level shift which varied with emission angle, and that the asymmetry parameter of the surface component was much larger than that reported for the bulk ͑0.18 vs ϳ0.12͒. However, Speranza and Minati 4 found a common natural linewidth of 95 meV, in close agreement with the value reported here for the bulk component, for both surface and bulk components, which is much narrower than the typical natural linewidth reported in ͑surface sensitive͒ SXPS studies.…”

“…Such artificially constrained fits also lead to a variable surface core-level shift and surface-to-bulk intensity ratios which imply an electron inelastic mean-free path more than three times that measured for carbon materials. 11,12 Balasubramanian et al 1 explained the lack of a bulk C 1s component in some SXPS studies of graphite in terms of the final state of the bulk photoelectron. However, this hypothesis does not explain the observations of Lizzit et al 2 of a low-energy C 1s component only for HOPG and not for other graphite types nor the angular dependence of the low binding-energy component.…”

“…By careful analysis of the data we show that the C 1s line of HOPG consists of both surface and bulk components: the line shape of the surface component is found to be consistent with that observed in SXPS studies, a constant ͑to within experimental error͒ surface-bulk energy splitting is observed, and the variation of the relative intensity of surface and bulk components is consistent with measurements of the inelastic mean-free path ͑IMFP͒ of electrons of the appropriate kinetic energy in graphite. 11,12 Some of the origins of the spread in results so far reported in the literature are discussed. Measurements were performed using a Scienta ESCA300 x-ray photoelectron spectrometer at the National Centre for Electron Spectroscopy and Surface Analysis ͑NCESS͒, Daresbury, U.K. Spectra were obtained with monochromated Al K ␣ radiation ͑1486.6 eV͒ at a resolution of 294Ϯ 5 meV, as determined from the Gaussian broadening of the Fermi edge of a silver calibration sample measured at room temperature.…”

Correction for the effects of adventitious carbon overlayers in quantitative XPS analysis

“…Using higher values for both simply increases the computed overlayer thickness (and vice versa) without a †ecting the computed substrate composition. Initially, a value of 15 at 1000 eV Ó was selected for the carbonaceous overlayer, in line with the lowest reported values for graphite or evaporated carbon, which range from 15 to 31 To obtain Ó15 Ó.3 the excellent corrections shown in Figs 1 and 2, the substrate attenuation lengths were then varied to obtain the most consistent behaviour. Higher values for j result in increased correction, so that the horizontal line is curved upwards as the layer thickness increases, while lower values result in progressive undercorrection and a corresponding downwards curve.…”

Characterization of graphene films grown on CuNi foil substrates

“…Measurements of the attenuation length λ of electrons in thin carbon films by Steinhardt, Hudis, and Perlman using XPS resulted in a value of 19 Å at a kinetic energy of 1169 eV [31]. Therefore, the value of λ measured by our group is somewhat lower than both the theoretical value of λ i and the previously measured experimental value of λ.…”