Supported metals, small particles specific catalysts

To maximize the rate of a reaction, one needs the maximum exposure of metal- or catalytically active atoms to the reactants. Hence there is a great desire to stabilize small particles on catalyst supports. In the next two subsections on transition metals we will provide a detailed description of changes in the chemical reactivity of transition metals when the particle size decreases. This provides a short background to aid in understanding the effects of particle size on catalysis. In the next subsection we discuss cluster size dependence effects and in the subsections that follow we will summarize the specific effort on supported Au clusters. 

The saturation coverage during chemisorption on a clean transition-metal surface is controlled by the fonnation of a chemical bond at a specific site [5] and not necessarily by the area of the molecule. In addition, in this case, the heat of chemisorption of the first monolayer is substantially higher than for the second and subsequent layers where adsorption is via weaker van der Waals interactions. Chemisorption is often usefLil for measuring the area of a specific component of a multi-component surface, for example, the area of small metal particles adsorbed onto a high-surface-area support [6], but not for measuring the total area of the sample. Surface areas measured using this method are specific to the molecule that chemisorbs on the surface. Carbon monoxide titration is therefore often used to define the number of sites available on a supported metal catalyst. In order to measure the total surface area, adsorbates must be selected that interact relatively weakly with the substrate so that the area occupied by each adsorbent is dominated by intennolecular interactions and the area occupied by each molecule is approximately defined by van der Waals radii. This... 

Specifically, catalysts are typically in the form of a ceramic support carrying small amounts of metals such as chromium, nickel, or platinum. Alumina and silica are commonly used in the construction of the ceramic support. Die catalysts lose their activity progressively via various deactivation mechanisms (Pavel and Elvin, 1994). Tliermal regeneration is often employed for regaining catalytic activity, if applicable, but some of the particles break during this process. Once the catalyst particles become too small to be useful, they constitute a waste disposal problem, since catalysts may contain heavy metals that are considered hazardous, or other harmful components. 

A similar explanation may well be valid for a study by Yates and Sinfelt of the specific activity of rhodium catalysts supported on silica for the hydrogenolysis of ethane to methane 49). As in the example just discussed, there appears to be a sharp contrast in order of magnitude between the specific activity of catalysts with particle size below 40 A (the sensitive range of Poltorak) and above that size (130-2500 A) where bulk behavior is expected. In this case, I speculate that, with very small particles, hydrocarbon surface residues which appear to play an important role in the hydrogenolysis of ethane may well perturb the metallic character of the small rhodium particles, just like adsorbed oxygen in the case of Poltorak and co-workers. 

In the original work on catalytic ammonia synthesis, Haber [41] had used an osmium catalyst, but this metal was much too expensive to be the basis of the large-scale industrial plants. In the long search for alternatives to the Mittasch catalyst, alkali-promoted ruthenium was found to exhibit specific activity, which is even superior to the iron catalyst [42] and which was subsequently developed to an industrial catalyst [43]. The Mittasch catalyst is cheap and the alumina promoter provides a high specific surface area. This situation is different with Ru catalysts that are prepared as small particles on a suitable support. Figure 1.1 was a t)q)ical electron microscopic picture from such a catalyst particle on MgAl204 (spinel) support [44]. 

Surface Area, Porosity, and Permeability. Some very interesting and important phenomena involve small particles and their surfaces. For example, SO2 produced from mining and smelting operations that extract metals such as Cu and Ph from heavy metal sulfide ores can be oxidized to SO3 in the atmosphere, thus contrihutingto acid rain problems. The reaction rate depends not only on the concentration of the SO2 bnt also on the siuTace area of any catalyst available, such as airborne dnst particles. The efficiency of a catalyst depends on its specific surface area, defined as the ratio of siuTace area to mass (17). The specific snrface area depends on both the size and shape, and is distinctively high for colloidalsized species. This is important in the catalytic processes nsed in many indnstries for which the rates of reactions occurring at the catalyst siuTace depend not only on the concentrations of the feed stream reactants bnt also on the sinface area of catalyst available. Since practical catalysts freqnently are snpported catalysts, some of the sinface area is more important than the rest. Since the supporting phase is usnally porous the size and shapes of the pores may influence the reaction rates as well. The final rate expressions for a catalytic process may contain all of these factors sinface area, porosity, and permeability. 

Specific activities of Pd, Pt and Rh catalysts in propene oxidation are reported in Table 1.7. Contrary to what was observed in alkane oxidation, propene oxidation is not very sensitive to the nature of the metal. Quite similar TOP were measured over unsupported metals, while Pt and Pd seemed to be shghtly more active than Rh when supported on alumina. Propene oxidation is not very sensitive to metal particle size. However, intrinsic activity would be rather higher on small particles. As TOP are higher or much higher on unsupported metals, it seems that alumina could play a negative role in propene oxidation. The intermediary formation of partially oxidized compounds (acrolein, alcohols,...) is not excluded. Alumina might store and stabilize these intermediates, slowing down the total oxidation. 

Specific featiu>es of palladlvun catalysts vdiich pass alkaline treatment (procedure employed often in the catalyst preparation [10]) may be derived from the data in Table 1. Due to adsorption on carbon support, the hydroxy-species are stabilized against agglomeration (which occurs in solutions) and, so, still small particles of metallic palladium can be produced. TEH examination showed that Pd crystallites in catalysts 1 and 2 are indistinguishable in size (1-2 nm 3-8 nm as predominant size in sample 3 -1). Moreover, fraction of the palladlvun exposed proved to be substantially higher in catalysts 2 than in catalysts 1. And... 

Nitrogen adsorption experiments showed a typical t)q5e I isotherm for activated carbon catalysts. For iron impregnated catalysts the specific surface area decreased fix>m 1088 m /g (0.5 wt% Fe ) to 1020 m /g (5.0 wt% Fe). No agglomerization of metal tin or tin oxide was observed from the SEM image of 5Fe-0.5Sn/AC catalyst (Fig. 1). In Fig. 2 iron oxides on the catalyst surface can be seen from the X-Ray diffractions. The peaks of tin or tin oxide cannot be investigated because the quantity of loaded tin is very small and the dispersion of tin particle is high on the support surface. 

---