Hyperthermal alkali-ion scattering from a metal surface: A theoretical study of the potential

The ${\mathrm{K}}^{+}$-W ion-atom scattering potential is calculated with the Hartree-Fock-Slater linear combination of atomic orbitals (HFS-LCAO) method. For hyperthermal (10–100 eV) ${\mathrm{K}}^{+}$ scattering from a W(110) surface, classical-trajectory calculations are performed, where the ${\mathrm{K}}^{+}$-W(110) ion-surface potential is represented by a sum of pairwise-calculated (HFS-LCAO) ${\mathrm{K}}^{+}$-W potentials. The results of these classical-trajectory calculations are compared with experiment and with the results of similar trajectory calculations using a sum of Ziegler-Biersack-Littmarck ‘‘universal’’ pair potentials. From these comparisons, it turns out that the HFS-LCAO pair potential is able to reproduce well on-top-site hyperthermal ${\mathrm{K}}^{+}$ scattering from a W(110) surface, contrary to the Ziegler-Biersack-Littmarck potential, which clearly does not work very well in this low-energy range. The inability of the HFS-LCAO pair potential to give a proper description of ${\mathrm{K}}^{+}$ scattering from the hollow site of the W(110) surface unit cell can be ascribed to the breakdown of a summation of pair potentials. This is clear from the difference between the sum of the calculated ${\mathrm{K}}^{+}$-W ion-atom potentials and a calculated ${\mathrm{K}}^{+}$-${\mathrm{W}}_5$ ion-cluster potential, the cluster representing the W(110) surface. The ion-cluster calculations indicate an extra repulsion of about 10% at the center of the W(110) surface unit cell. This extra hollow-site repulsion can be explained by analyzing the properties of the exchange (Pauli, Born) repulsions between the ${\mathrm{K}}^{+}$ ion and (i) one W atom and (ii) the W(110) surface (${\mathrm{W}}_5$ cluster) at the hollow site.