Loading, surface oxide-support interaction

The Pti samples (182) were prepared as colloids, protected by a PVP polymer film. Layer statistics according to the NMR layer model (Eqs. 28-30) for samples with x = 0,0.2, and 0.8 are shown in Fig. 63. The metal/ polymer films were loaded into glass tubes and closed with simple stoppers. The NMR spectrum and spin lattice relaxation times of the pure platinum polymer-protected particles are practically the same as those in clean-surface oxide-supported catalysts of similar dispersion. This comparison implies that the interaction of the polymer with the surface platinums is weak and/or restricted to a small number of sites. The spectrum predicted by using the layer distribution from Fig. 63 and the Gaussians from Fig. 48 show s qualitative agreement w ith the observed spectrum for x = 0 (Fig. 64a). 

The metals supported on a 2-D surface are investigated by periodic models, with the aim of characterising the properties of the supported metal particles as function of their cluster size from isolated supported metal atoms to regular overlayers. We will discuss two different noble metals Pd and Pt. The properties of these metals are similar, but when supported on an oxide surface they show in many cases different behaviour. One important distinction between them concerns the shape of the supported clusters, which are known to depend on the strength of the metal-support interaction, and therefore on parameters such as the choice of the substrate, working temperature and loading [11]. Pt is reported to form 2-dimensional particles followed by transformation into 3-dimensional particles [12], while Pd tends to form 3-dimensional but flat (raft-like) particles when supported on a Z1O2 support [13]. 

The nature of the supported metal oxide species depends upon a number of factors the preparation method (wet chemical synthesis plus calcination), chemical interactions between the support and surface layers, and surface density (surface oxide weight loading and specific surface area of the support oxide) [5]. Figure 11.1 schematically demonstrates the various dehydrated surface structures commonly observed for a mono-oxo metal oxide isolated, oligomeric, polymeric, and crystalline species. Several reviews comprehensively catalog the expected surface structures for transition metal mono- and polyoxoanions in four-, five-, and sixfold coordination under nonreaction conditions [3, 23-25],... 

The analysis of surface coverage places primary importance on the concept of support hydroxyl titration. As has been verified repeatedly for supported metal oxide systems immediately after deposition, precursors of surface oxides (typically salts) interact weakly with the oxide support through hydrogen bonding and electrostatic forces [8, 11, 47] (except in the case of grafting). However, postdeposition thermal treatment leads to formation of a thermally stable, two-dimensional overlayer of surface metal oxide distinctly different in structure than the precursor molecule (under dehydrated conditions) [48]. Numerous infrared studies have shown a loss of support hydroxyl density with an increase in surface oxide loading [8, 27, 28,... 

Therefore, the loaded metals virtually interacted with an oxidized surface rather than the native carbide surface. Schweitzer et al. developed a synthesis method that could allow a direct contact of Pt with the genuine Mo C surface [65]. In the WGS reaction, the resulting Pt/MOjC catalysts exhibited a higher activity than the most active oxide-supported Pt catalysts and a commercial Cu-Zn-Al catalyst. Moreover, the experimental rates were more consistent with those predicted by the perimeter model than by the particle surface model, suggesting that the rate-determining step for WGS on Pt/Mo C catalysts occurred at the perimeters of the Pt particles. 

---