Abstract

Using MD simulations, we have calculated the surface phonon spectral density functions for the (100), (110), and (111) surfaces of Ni and Cu using Finnis-Sinclair (FS) potentials. These simulated phonon spectral densities are compared to the experimental inelastic helium atom scattering and HREELS data which are available for the three basal faces of Ni and Cu. We find that the overall shape of the calculated surface and second layer phonon spectral densities qualitatively reproduce those obtained from force constant fits, i.e. lattice dynamical modelling, of the experimental phonon dispersion data. Good agreement is also found between the calculated and experimental geometric separations between the surface and second layer for a given interface. However, on all surfaces the phonon frequencies calculated with Finnis-Sinclair potentials are lower than the experimentally measured values. The best agreement between our calculated results and the experimentally measured phonon frequencies was for the (100) and (110) surfaces, while the poorest agreement was on the (111) surfaces. From this we conclude that Finnis-Sinclair model potentials derived from bulk properties systematically underestimate the many body binding potential at the surface. This underestimation of the many body binding term is also manifested in the magnitude of the calculated surface stress. Our results indicate that the Finnis-Sinclair model potentials are quite adequate for a good qualitative and semi-quantitative description of the bonding changes at the surfaces of Ni and Cu.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call