Surface segregation chemisorption effect

In this section, we extend the above formalism to that for an alloy surface within the CPA, which serves as the model for the pre-chemisorption substrate. The model discussed here is based on that of Ueba and Ichimura (1979a,b) and Parent et al (1980). For a comprehensive introduction to alloy surfaces see Turek et al (1996). A feature of surface-alloy models, which is different from bulk ones, is that the CP in layers near the surface is different from that in the bulk, due to the surface perturbation. Moreover, the alloy concentration in the surface layers may be quite different from that in the bulk, a feature known as surface segregation. (See Ducastelle et al 1990 and Modrak 1995 for recent reviews.) We assume that both of these surface effects are confined to the first surface layer only. 

However, surface segregation (cs / Cb) has a radical effect on AE, as can clearly be seen in Fig. 6.3(b). Cu/Ni alloys are known (Kelley and Ponec 1981, Ouannasser et al 1997) to have an enriched Cu concentration in the surface layer for all bulk concentrations. As a result, the alloy shows a more Cu-like behaviour than it would if it were non-segregated. In particular, AE has a value significantly closer to that for pure Cu than in the case where cs = Cb, and this occurs at all bulk concentrations c. The smallest change in AE occurs in Cu-rich alloys, which is understandable, because these alloys have mostly Cu in the surface layer anyway, so the effect of surface segregation is relatively small. Thus, surface segregation has a lesser effect in these alloys than in Ni-rich ones, which have mostly Ni in the bulk, but may have a Cu majority in the surface layer. Clearly, then, the concentration cs of the surface layer is the primary parameter in determining the chemisorption properties of the DBA. 

At the higher metal level (2.0-4.5% Ni with up to 2% Sb) used to study artificially contaminated materials, XRD results have shown the formation of Ni-Sb alloys (NiSb x<0.08) whereas XPS data have indicated that a non-reducible antimony oxide, a well dispersed reducible Sb phase together with reducible Sb (that form an alloy with reducible Ni), were present. Selective chemisorption data for unsupported Ni-powders showed that one surface structure can effectively passivate 2-3 Ni atoms with respect to H2 chemisorption. XPS examination confirmed that Sb segregates at the surface of Ni particles where it can drastically affect the electron properties of neighboring Ni atoms thus reducing their activity. 

There is evidence of a promoting action of chromium on nickel catalysts for the reaction of hydrogenation of valeronitrile in our conditions. Introduction of chromium increased the initial specific activity and the selectivity. The promoting effect of chromium on activity could be correlated to the increase of the metallic surface. Another explanation could be that the Cr+ segregated at the surface of the catalyst may play the role of a Lewis acid center and may be responsible for a better chemisorption of valeronitrile on the catalysts, through nitrogen lone pair electrons or the n orbital of the CN bond. However, further examination of the results obtained (see Fig. 3)... 

Experimental results bear out these theoretical predictions. Work with Ti02 deposited on platinum foil (14, 15) indicates that under UHV conditions in the absence of oxygen little, if any segregation of titanium occurs, In agreement with the theoretical predictions (11). In contrast, an AES study (16) of stoichiometric PtjTi showed that the surface composition is close to the bulk composition, with consequent effects on H2 and CO chemisorption. 

In a previous paper, it was shown that there is a rough correlation between the catalytic activity in methanol synthesis and the whole chemisorption activity towards CO (35). Therefore, the low catalytic activity of copper-rich catalysts may be attributed to the segregation of much of the copper as well crystallized metallic particles (36). The further decrease in activity for Cat D and, especially. Cat E is in agreement with the hypothesis of the formation of alloys at the surface of the copper particles (26,27), taking into account the poisoning effect of small amount of cobalt (37). 

---