Cedric J. Powell

Attenuation Lengths of Low-Energy Electrons in Solids

Abstract A compilation is presented of measured attenuation lengths of low-energy electrons in solids in the energy range (40 to 2000 eV) normally employed in X-ray photoelectron and Auger-electron spectroscopy. The techniques used to obtain electron attenuation lengths are summarized, and it is pointed out that the accuracy of measurement needs both to be defined adequately and to be improved for more meaningful intercomparisons of data and theory. An approximate expression is derived to predict attenuation lengths using either dielectric data (derived from optical or electron-energy-loss data) or average excitation energies estimated from electron binding energies for given materials at electron energies greater than about 500 eV. Good agreement is found between the predictions of this formula and some measured attenuation lengths (e.g. for Al, C, Mo, W) but further work is required to validate the formula and to extend it to lower electron energies.

Evaluation of Calculated and Measured Electron Inelastic Mean Free Paths Near Solid Surfaces

An analysis is given of the consistency of calculated and measured electron inelastic mean free paths (IMFPs) near solid surfaces for electron energies between 50 and 104 eV, the energy range of relevance for surface analysis by Auger electron spectroscopy and x-ray photoelectron spectroscopy. This evaluation is based on IMFPs calculated from experimental optical data and on IMFPs measured by elastic-peak electron spectroscopy (EPES). We describe the methods used for the calculations and measurements, and we identify the various sources of uncertainty. Most of our evaluation is based on IMFPs for seven elemental solids (Al, Si, Ni, Cu, Ge, Ag, and Au) for which there were at least two sources of IMFP calculations and at least two sources of IMFP measurements for each solid. Our comparison of the calculated IMFPs showed a high degree of consistency for Al, Ni, Cu, Ag, and Au. The comparison of measured IMFPs showed greater scatter than for the calculated IMFPs, but reasonable consistency was found for the ...

elsepa—Dirac partial-wave calculation of elastic scattering of electrons and positrons by atoms, positive ions and molecules ☆

The fortran 77 code system elsepa for the calculation of elastic scattering of electrons and positrons by atoms, positive ions and molecules is presented. These codes perform relativistic (Dirac) partial-wave calculations for scattering by a local central interaction potential V(r)V(r). For atoms and ions, the static-field approximation is adopted, with the potential set equal to the electrostatic interaction energy between the projectile and the target, plus an approximate local exchange interaction when the projectile is an electron. For projectiles with kinetic energies up to 10 keV, the potential may optionally include a semiempirical correlation–polarization potential to describe the effect of the target charge polarizability. Also, for projectiles with energies less than 1 MeV, an imaginary absorptive potential can be introduced to account for the depletion of the projectile wave function caused by open inelastic channels. Molecular cross sections are calculated by means of a single-scattering independent-atom approximation in which the electron density of a bound atom is approximated by that of the free neutral atom. Elastic scattering by individual atoms in solids is described by means of a muffin-tin model potential. Partial-wave calculations are feasible on modest personal computers for energies up to about 5 MeV. The elsepa code also implements approximate factorization methods that allow the fast calculation of elastic cross sections for much higher energies. The interaction model adopted in the calculations is defined by the user by combining the different options offered by the code. The nuclear charge distribution can be selected among four analytical models (point nucleus, uniformly charged sphere, Fermis distribution and Helms uniform–uniform distribution). The atomic electron density is handled in numerical form. The distribution package includes data files with electronic densities of neutral atoms of the elements hydrogen to lawrencium (Z=1Z=1–103) obtained from multiconfiguration Dirac–Fock self-consistent calculations. For comparison purposes, three simple analytical approximations to the electron density of neutral atoms (corresponding to the Thomas–Fermi, the Thomas–Fermi–Dirac and the Dirac–Hartree–Fock–Slater models) are also included. For calculations of elastic scattering by ions, the electron density should be provided by the user. The exchange potential for electron scattering can be selected among three different analytical approximations (Thomas–Fermi, Furness–McCarthy, Riley–Truhlar). The offered options for the correlation–polarization potential are based on the empirical Buckingham potential. The imaginary absorption potential is calculated from the local-density approximation proposed by Salvat [Phys. Rev. A 68 (2003) 012708]. Program summary Title of program:ELSEPA Catalogue identifier: ADUS Program summary URL:http://cpc.cs.qub.ac.uk/cpc/summaries/ADUS Program obtainable from: CPC Program Library, Queens University of Belfast, N. Ireland License provisions: none Computer for which the program is designed and others in which it is operable: Any computer with a FORTRAN 77 compiler Operating systems under which the program has been tested: Windows XP, Windows 2000, Debian GNU/Linux 3.0r0 (sarge) Compilers: 1Compaq Visual Fortran v6.5 (Windows); GNU FORTRAN, g77 (Windows and Linux) Programming language used: FORTRAN 77 No. of bits in a word: 32 Memory required to execute with typical data: 0.6 Mb No. of lines in distributed program, including test data, etc.:135 489 No. of bytes in distributed program, including test data, etc.: 1 280 006 Distribution format: tar.gz Keywords: Dirac partial-wave analysis, electron elastic scattering, positron elastic scattering, differential cross sections, momentum transfer cross sections, transport cross sections, scattering amplitudes, spin polarization, scattering by complex potentials, high-energy atomic screening functions Nature of the physical problem: The code calculates differential cross sections, total cross sections and transport cross sections for single elastic scattering of electrons and positrons by neutral atoms, positive ions and randomly oriented molecules. For projectiles with kinetic energies less than about 5 MeV, the programs can also compute scattering amplitudes and spin polarization functions. Method of solution: The effective interaction between the projectile and a target atom is represented by a local central potential that can optionally include an imaginary (absorptive) part to account approximately for the coupling with inelastic channels. For projectiles with kinetic energy less that about 5 MeV, the code performs a conventional relativistic Dirac partial-wave analysis. For higher kinetic energies, where the convergence of the partial-wave series is too slow, approximate factorization methods are used. Restrictions on the complexity of the program: The calculations are based on the static-field approximation. The optional correlation–polarization and inelastic absorption corrections are obtained from approximate, semiempirical models. Calculations for molecules are based on a single-scattering independent-atom approximation. To ensure accuracy of the results for scattering by ions, the electron density of the ion must be supplied by the user. Typical running time: on a 2.8 GHz Pentium 4, 1 the calculation of elastic scattering by atoms and ions takes between a few seconds and about two minutes, depending on the atomic number of the target, the adopted potential model and the kinetic energy of the projectile. Unusual features of the program: The program calculates elastic cross sections for electrons and positrons with kinetic energies in a wide range, from a few tens of eV up to about 1 GeV. Calculations can be performed for neutral atoms of all elements, from hydrogen to lawrencium (Z=1Z=1–103), ions and simple molecules.

Oxygen as a surfactant in the growth of giant magnetoresistance spin valves

We have found a novel method for increasing the giant magnetoresistance (GMR) of Co/Cu spin valves with the use of oxygen. Surprisingly, spin valves with the largest GMR are not produced in the best vacuum. Introducing 5×10−9 Torr (7×10−7 Pa) into our ultrahigh vacuum deposition chamber during spin-valve growth increases the GMR, decreases the ferromagnetic coupling between magnetic layers, and decreases the sheet resistance of the spin valves. It appears that the oxygen may act as a surfactant during film growth to suppress defects and to create a surface which scatters electrons more specularly. Using this technique, bottom spin valves and symmetric spin valves with GMR values of 19.0% and 24.8%, respectively, have been produced. These are the largest values ever reported for such structures.

Comparison of electron elastic-scattering cross sections calculated from two commonly used atomic potentials

We have analyzed differential cross sections (DCSs) for the elastic scattering of electrons by neutral atoms that have been derived from two commonly used atomic potentials: the Thomas–Fermi–Dirac (TFD) potential and the Dirac–Hartree–Fock (DHF) potential. DCSs from the latter potential are believed to be more accurate. We compared DCSs for six atoms (H, Al, Ni, Ag, Au, and Cm) at four energies (100, 500, 1000, and 10 000 eV) from two databases issued by the National Institute of Standards and Technology in which DCSs had been obtained from the TFD and DHF potentials. While the DCSs from the two potentials had similar shapes and magnitudes, there can be pronounced deviations (up to 70%) for small scattering angles for Al, Ag, Au, and Cm. In addition, there were differences of up to 400% at scattering angles for which there were deep minima in the DCSs; at other angles, the differences were typically less than 20%. The DCS differences decreased with increasing electron energy. DCSs calculated from the two ...

Quantitative surface analysis by X-ray photoelectron spectroscopy

Abstract Measurements have been made of the relative intensities of the principal features in X-ray photoelectron spectra of indium, lead, and aluminum oxide and compared with those expected from a simple model for the photoemission process. Systematic effects in the determination of line intensities are discussed and a suitable procedure for determining intensities is described. The satisfactory agreement between computed and measured intensities confirms the validity and utility of the photoemission model and associated data, and indicates that quantitative analyses of homogeneous single-phase surfaces can be obtained by X-ray photoelectron spectroscopy.

Elemental binding energies for X-ray photoelectron spectroscopy

Abstract A report is presented here of an analysis of core-electron binding-energy (BE) data for solid elements from four sources. For each set of data, a single instrument had been used to measure BEs for many elements. In order to compare BEs from these four sources, it was necessary to recalibrate the BE scales of three instruments; this recalibration was performed using reference data from the UK National Physical Laboratory and a draft calibration standard. Mean BEs were then computed for those elements and core levels for which there were two or more measurements. Finally, individual differences from the mean values were calculated; this analysis disclosed several discrepancies in data due either to differences in specimens or to possible mistakes. Plots of differences in BE values from mean values versus BE indicated that the BE scales of the four instruments were close to linear. The mean BE values for each element appear to be a useful reference set for the determination of chemical shifts in X-ray photoelectron spectroscopy.

Magnetoresistance values exceeding 21% in symmetric spin valves

We report values of the giant magnetoresistance (GMR) effect exceeding 21% in symmetric spin valves, the highest values ever reported for such structures. The key elements in this achievement are the use of a Co/Cu/Co/Cu/Co multilayer in which the center Co layer is substantially thicker than the outer Co layers and the use of the antiferromagnetic insulator NiO at the top and bottom to pin the adjacent Co layers magnetically. The relative Co layer thicknesses suggest that some specular scattering of conduction electrons may occur at the metal/insulator interfaces and may enhance the GMR.

Proposed formula for electron inelastic mean free paths based on calculations for 31 materials

A new general formula is proposed for determining electron inelastic mean free paths (IMFPs) for 200–2000 eV electrons in solids. The new formula is based on separate IMFP calculations for 27 elements and 4 compounds using an algorithm due to Penn. This formula is believed useful for determining the IMFP dependence on electron energy for a given material and the material-dependence for a given energy. The new formula should also be a reasonable guide to electron attenuation lengths which have been difficult to determine with the needed accuracy.

The quest for universal curves to describe the surface sensitivity of electron spectroscopies

Abstract An overview is given of data for electron attenuation lengths (ALs) and inelastic mean free paths (IMFPs) in the energy range of interest for electron spectroscopy of surfaces with emphasis on the search for so-called universal curves. Reliable knowledge of ALs and IMFPs is important for determining the surface sensitivity of electron spectroscopies and for quantitative surface analysis by AES and XPS. Suitable formulae for ALs and IMFPs are needed as user guides but existing formulae are of limited accuracy due to problems in AL measurements and approximations in IMFP calculations. AL measurements and the various AL and IMFP formulae are evaluated. A new formula due to Tanuma et al. shows promise of being a useful guide to the energy- and material-dependences of IMFPs and may be a reasonable though less accurate guide for ALs.