No abstract

No abstract

No abstract

No abstract

No abstract

No abstract

No abstract

The importance of recognizing a 1 x 1 structure preceding a superstructure in a chemisorption experiment is demonstrated by the system N-Ti(OOOl). Low-energy-electrondiffraction analysis of the 1 x 1 structure reveals that the N atoms are located in the octahedral holes 1.22 ±0.05 A below the first layer of Ti atoms. Interatomic distances are such as to justify the view of the surface structure as a single double-layer of TiN.The present Letter reports the first successful structure analysis by low-energy electron diffraction (LEED) of an underlayer of chemisorbed atoms on a clean metal surface. All structures of chemisorbed species on metal surfaces investigated so far are of the overlayer type, i.e., with the adatoms located above the outermost layer of the substrate. 1 The present note discusses the case of a lx 1 room-temperature structure of nitrogen on Ti(0001), in which the atomic arrangement reflects very closely the structure of the bulk compound TiN, the N atoms being located underneath the top layer of Ti atoms in what are usually called octahedral interstitial positions.The reaction of a clean Ti(0001) surface with nitrogen gas at room temperature goes through the following stages as observed by LEED and Auger electron spectroscopy (AES). With increasing concentration of nitrogen on the surface, the LEED pattern of the clean surface maintains its geometry but exhibits changes in the intensities of the diffracted beams. We interpret this observation as indicative of the formation of a Ti(0001)-lx 1-N structure. Completion of the lx 1 structure, at 4-5 L (1 L= 1 langmuir= 10" 6 Torr sec), is associated with minima in the rates of changes of LEED spectra on adsorption 2 ; with an inflection point in the rate of the plasmon-decay Auger line at 12 eV 3 ; and with a maximum in intensity of the interfacial Auger peak at 23 eV. 4 After completion of the lx 1 structure, the nitrogen concentration on the surface as revealed by AES keeps increasing until, around exposures of 8-9 L, a VlTxV~3-30 o structure is observed which is completed at approximately 15 L. The nitrogen AES signal remains constant after about 9 L. 4 It is clear, in the light of these observations, that the /3x/3-30° structure cannot be a low-coverage structure of the type reported, e.g., by Forstmann, Berndt, and Biittner 5 -we have carried out intensity calculations that indeed confirm this fact.In all experiments carried out so far, both the lx 1 and the /3x V3-30 0 phases appeared to be reproducibly well crystallized. The background in the LEED patterns remained at least as low for the lx 1 phase as it was for the clean substrate surface, then increased slightly in the early stages of formation of the /3XV1T-30 0 phase but finally dropped again to a very low level after completion of the latter structure. Reproducibility of peak shapes, peak positions, and peak intensities in LEED spectra of both the lx 1 and V~3" x/3-30° phases was excellent, not only with a given Ti sample in a first run but also with a different Ti sample in a second run...