Catalyst ensemble effect

The high dispersity inside the nano-honeycomb matrix and the high surface area of the nanopartides leads to very good electrocatalytic activity. The electrocatalytic activities of nanosized platinum particles for methanol, formic add and formaldehyde electrooxidation have been recently reported [215]. The sensitivity of the catalyst particles has been interpreted in terms of a catalyst ensemble effect but the detailed microscopic behaviour is incomplete. Martin and co-workers [216] have demonstrated the incorporation of catalytic metal nanopartides such as Pt, Ru and Pt/Ru into carbon nanotubes and further used them in the electrocatalysis of oxygen reduction, methanol electrooxidation and gas phase catalysis of hydrocarbons. A related work on the incorporation of platinum nanopartides in carbon nanotubes has recently been reported to show promising electrocatalytic activity for oxygen reduction [217]. 

We have studied the steady-state kinetics and selectivity of this reaction on clean, well-characterized sinxle-crystal surfaces of silver by usinx a special apparatus which allows rapid ( 20 s) transfer between a hixh-pressure catalytic microreactor and an ultra-hixh vacuum surface analysis (AES, XPS, LEED, TDS) chamber. The results of some of our recent studies of this reaction will be reviewed. These sinxle-crystal studies have provided considerable new insixht into the reaction pathway throuxh molecularly adsorbed O2 and C2H4, the structural sensitivity of real silver catalysts, and the role of chlorine adatoms in pro-motinx catalyst selectivity via an ensemble effect. 

In order to verify the presence of bimetallic particles having mixed metal surface sites (i.e., true bimetallic clusters), the methanation reaction was used as a surface probe. Because Ru is an excellent methanation catalyst in comparison to Pt, Ir or Rh, the incorporation of mixed metal surface sites into the structure of a supported Ru catalyst should have the effect of drastically reducing the methanation activity. This observation has been attributed to an ensemble effect and has been previously reported for a series of silica-supported Pt-Ru bimetallic clusters ( ). 

An apparent particle size effect for the hydrodechlorination of 2-chlorophenol and 2,4-dichlorophenol was observed by Keane et al. [147], Investigating silica supported Ni catalysts (derived from either nickel nitrate or nickel ethane-diamine) with particles in the size range between 1.4 and 16.8 nm, enhanced rates for both reactions were observed with increased size over the full range (Figure 13). As electronic factors can be ruled out in this dimension, the observed behavior is traced back to some sort of ensemble effect, known from CFC transformations over Pd/Al203... 

PtMo alloys are not as effective as PtRu for methanol, or ethanol, oxidation. As shown in Figure 29, the d band vacancy per Pt atom for the PtMo/C catalyst continues to increase until 0.6 V vs RHE, in contrast to the behavior of PtRu/C. ° The authors attribute this difference to the lack of removal of the Cl fragments from the particle surface by the oxy-hydroxides of Mo. However, the difference in the electrocatalytic activity of PtRu and PtMo catalysts may be attributed to ensemble effects as well as electronic effects. The former are not probed in the white line analysis presented by Mukerjee and co-workers. In the case of methanol oxidation, en-... 

A brief overview of bimetallic catalysts is presented. Electronic vs. ensemble effects are discussed, and literature is reviewed on single crystal bimetallics, and supported bimetallic clusters. Bimetallic cluster compounds are considered as models. Structural considerations, effects of potential poisons, particles from bimetallic cluster compounds, and catalytic activity/selectivity studies are briefly reviewed and discussed. 

All of the bimetallic SMAD catalysts have shown unusual catalytic properties. Several laboratories are now reporting similar findings.(22,23) Both geometric/ensemble effects and electronic effects have been considered in trying to explain these unusual properties. 

The composition of an alloy surface is often very different than the alloy s bulk composition due to segregation effects. The overall activity of a catalyst is determined by the distribution of active sites. This distribution may be very heterogeneous both in terms of the local environments that define each site and their chemical reactivities. The reactivity of any specific active site can be affected by contributions from strain, ligand and ensemble effects. Computational methods are well suited to exploring these effects because one can simulate model systems where only one effect dominates as well as model systems where multiple effects are important. 

One function in some heterogeneous catalysts is to bind and dissociate simple diatomic molecules such as Nj and CO. The strength of the binding interaction with the metal surface provides the thermodynamic drive for these cleavage reactions. Reactions such as CO and N2 dissociation appear to require more than one surface metal atom, and this phenomenon is referred to as a metal ensemble effect (22). Thus, the conversion of CO + H2 to... 

As described in Chapter 3, the reasons given to explain this dilution effect of copper on nickel were that the hydrogenolysis reaction required a group or ensemble of nickel atoms on the catalyst surface and the presence of copper prevented the formation of the appropriately sized ensembles and the reaction was inhibited. On the other hand, the reactions involving C-H bond breaking take place on single atom sites so the surface dilution by copper has no effect until the surface is almost completely covered by the copper. The rate increase observed in the deuterium exchange on cyclopentane (Fig. 12.7b), however, is not easily rationalized by this surface ensemble effect alone. 

In conclusion by using rhenium as adsorbent instead of platinum, it is possible to achieve the ensemble control by sulfur passivation, at sulfur levels comparable to those applied for nickel and much lower than that which would have been required on a non-alloyed Pt-catalyst. A similar ensemble effect is achieved by alloying alone on Pt-Sn catalysts ... 

Abstract Transition metal carbides and phosphides have shown tremendons potential as highly active catalysts. At a microscopic level, it is not well understood how these new catalysts work. Their high activity is usually attributed to ligand or/and ensemble effects. Here, we review recent studies that examine the chemical activity of metal carbides and phosphides as a function of size, from clusters to extended surfaces, and metal/carbon or metal/phosphorous ratio. These studies reveal that the C and P sites in these compounds cannot be considered as simple spectators. They moderate the reactivity of the metal centers and provide bonding sites for adsorbates. 

The carbides of the early transition metals exhibit chemical and catalytic properties that in many aspects are very similar to those of expensive noble metals [1], Typically, early transition metals are very reactive elements that bond adsorbates too strongly to be useful as catalysts. These systems are not stable under a reactive chemical environment and exhibit a tendency to form compounds (oxides, nitrides, sulfides, carbides, phosphides). The inclusion of C into the lattice of an early transition metal produces a substantial gain in stability [2]. Furthermore, in a metal carbide, the carbon atoms moderate the chemical reactivity through ensemble and ligand effects [1-3]. On one hand, the presence of the carbon atoms usually limits the number of metal atoms that can be exposed in a surface of a metal carbide (ensemble effect). On the other hand, the formation of metal-carbon bonds modifies the electronic properties of the metal (decrease in its density of states near the Fermi level metal—>carbon charge transfer) [1-3], making it less chemically active... 

From not only the scientific but the technological point of view, bimetallic nanoparticles composed of two different metal elements are of greater interest and importance than monometallic nanoparticles [7,8]. Scientists have especially focused on bimetallic nanoparticles as catalysts because of their novel catalytic behaviors affected by the second metal element added. This effect of the second metal element can often be explained in terms of an ensemble and/or a ligand effect in catalyses. Such effects appear in bimetallic catalysts composed of both zerovalent metal atoms and another metal ions [9,10]. In this case, however, metal ions do not construct nanoparticles but are located close to them to exhibit an ensemble effect. This chapter covers the bimetallic nanoparticles composed of only zerovalent metals in homogeneous systems the supported or heterogeneous systems of metal nanoparticles are not covered. 

The rate of reaction expressed as molecules reacted (or formed) per unit time per catalytic site (or per exposed atom of active metal for metal catalysts) is called the turnover frequency. For supported metal catalysts the calculation requires knowledge of the dispersion, i.e., the fraction of the active metal available for adsorption of reactants. Boudart coined the term demanding (structure-sensitive) for catalyzed reactions for which the turnover frequency varies with the dispersion. Related to this is the ensemble effect, where the active site requires a specific multiatom grouping.f ... 

One of the questions which remains to be answered is whether the conclusion from the experiments with labeled carbon that CO dissociation is a fast step (cf. Section V,C) complies with the IR observation that CO js is abundantly present at the catalyst surface (cf. Section IV,B). A possible solution is to assume that CO storage and CO dissociation take place on different sites (48, 59). When FT catalysts such as Ni or Ru are alloyed with an inert metal such as copper, the activity decreases drastically. The alloy studies of Bond and Turnham (73) and Araki and Ponec (48) consistently indicate that the decrease in activity originates from a decrease in the preexponential rather than from an increase in the activation energy in the Arrhenius equation (48, 57, 73, 74). This is indicative of an ensemble effect... 

---