X-ray study of W(001) with and without hydrogen

We have carried out synchrotron x-ray diffraction studies of the clean W(001) surface reconstruction transition as a function of temperature, and of the room-temperature phases and phase transitions of W(001) with submonolayer coverages of hydrogen. For clean W(001) we find that below the transition temperature of ≅230 K, the ( x )R450 diffraction peaks have an intrinsic width, larger than the instrumental resolution, corresponding to finite-sized domains. Nevertheless, we are able to measure one and one-half decades of change in diffraction linewidth, and three decades in the amplitude. The detailed dependences of the surface peak widths are consistent with the predictions of the two-dimensional XY model with cubic anisotropy to leading order. We also find that the clean surface structure is commensurate at all temperatures measured, above and below the transition temperature. For hydrogen covered W(001) our data suggest a simple picture in which at very low coverages there is a trade-off with increasing coverage, analogous to that in a two-phase coexistence region, between hydrogen-poor and hydrogen-rich regions until a uniform but still disordered ( x )R450--H is attained. With further increase in coverage the surface layer exhibits a commensurate--incommensurate transition which, rather than being a solid--solid transition as previously assumed, is actually a lattice--gas melting transition into a domain-wall fluid phase. Such surface domain-wall fluid phases should occur commonly in chemisorbed systems