Abstract
Low-energy-electron-diffraction (LEED) intensity profiles are calculated for the (100) and (111) faces of copper and compared with experimental measurements. Calculations using a self-consistent electron-ion-core potential due to Snow and Waber are compared with calculations using a potential constructed from a simple overlap of atomic charge densities. The differences are found to be inconsequential as far as analyzing LEED spectra. Five partial-wave components are used to describe the vibronically renormalized electron-ion-core elastic-scattering vertex. The value of the inner potential V0 is determined by comparing the position of the Fermi level in Snow and Waber's band calculation with work-function measurements. Except for one set of data on the (111) face, this value of V0 gives a good placement of peak positions for electron energies ≲ 240 eV. It is found that the fine structure in the calculated profiles for the (111) face is considerably more sensitive to the value of the mean-free-path parameter used to describe the imaginary part of the one-electron self-energy than that in the calculated profiles for the (100) face. Analysis of the data indicates that the upper-layer spacing is the same as the bulk value (to within ∼ 5%) for both faces.