Electron Energy Loss Spectroscopy Study of Hydrogen Chemisorption on Metal Surfaces: Surface Sites and Scattering Selection Rules.

The object of this thesis is to study the vibrational modes of hydrogen on single crystal metal surfaces with electron energy loss spectroscopy (EELS) to make accurate assignments for the binding sites. Specular and off-specular collection are used to distinguish between dipole and non -dipole scattering mechanisms. Modes with perpendicular vibrational components usually have dipole scattering contributions although for hydrogen the intensity is generally weak and sometimes zero. In order to make the analysis procedure more straightforward, we show the ability to use a generalized selection rule for parallel modes with the in-plane scattering geometry of the spectrometer. Thus surfaces of C(,2v) symmetry have been treated. The rule, which is developed theoretically, states that, due to parallel momentum conservation, a parallel mode which is polarized perpendicular to the scattering plane is not excited. By using two azimuthal orientations of the crystal, the polarization of a hydrogen vibration is assigned on W(110) unambiguously for the first time. Due to the particular occupied surface site of H/Ni(110), however, the rule could not be applied to this system. A consistent trend on the strength of the dipole cross section for perpendicular modes of H with respect to the surface packing density is found both in our studies and those of other groups. This is explained with a qualitative discussion of the surface charge profile and the H position with respect to the positive charge background edge in the jellium model. In particular, for close-packed surfaces where the H is inside the positive charge background region, the dynamic dipole moment is effectively screened and the observed dipole scattering intensity is zero. On the other hand, in the case where the H is outside the positive background, the dipole scattering intensity is relatively strong. The model appears to break down for loosely packed surfaces where crystalline effects make it difficult to easily designate the position of the positive background edge. The design of the Penn EELS spectrometer is discussed with details of the electron optics calculations included. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of school.) UMI.