Hydrogenation structure insensitivity

Finke has reported remarkable catalytic lifetimes for the polyoxoanion- and tetrabutylammonium-stabi-lized transition metal nanoclusters [288-292]. For example in the catalytic hydrogenation of cyclohexene, a common test for structure insensitive reactions, the lr(0) nanocluster [296] showed up to 18,000 total turnovers with turnover frequencies of 3200 h [293]. As many as 190,000 turnovers were reported in the case of the Rh(0) analogue reported recently. Obviously, the polyoxoanion component prevents the precious metal nanoparticles from aggregating so that the active metals exhibit a high surface area [297]. 

This review covers the personal view of the authors deduced from the literature starting in the middle of the Nineties with special emphasis on the very last years former examples of structure-sensitive reactions up to this date comprise, for example, the Pd-catalyzed hydrogenation of butyne, butadiene, isoprene [11], aromatic nitro compounds [12], and of acetylene to ethylene [13], In contrast, benzene hydrogenation over Pt catalysts is considered to be structure insensitive [14] the same holds true for acetonitrile hydrogenation over Fe/MgO [15], CO hydrogenation over Pd [16], and benzene hydrogenation over Ni [17]. For earlier reviews on this field we refer to Coq [18], Che and Bennett [9], Bond [7], as well as Ponec and Bond [20]. 

Heterogeneously catalyzed hydrogenation of alkenes is generally considered to be a structure-insensitive reaction, as was deduced from numerous studies on more or less complex model catalyst systems [40-54]. However, the following sections will give examples of the opposite case. 

A study concerning the lowermost particle size necessary for admitting a reaction to occur is provided by Amiridis et al. [55] in their investigations of propene hydrogenation. For metal particle sizes greater than 2nm, hydrogenation of simple olefins is considered to be structure insensitive... 

The colloidal catalysts have been prepared in different particle sizes by the reduction of platinum tetrachloride with formic acid in the presence of different amounts of alkaloid. Optical yields of 75-80% ee were obtained in the hydrogenation of ethyl pyruvate with chirally modified Pt sols (Equation 3.7). The catalysts were demonstrated to be structure-insensitive since turnover frequencies (ca. 1 sec-1) and enantiomeric excess are independent of the particle size. 

Carbon monoxide oxidation is a relatively simple reaction, and generally its structurally insensitive nature makes it an ideal model of heterogeneous catalytic reactions. Each of the important mechanistic steps of this reaction, such as reactant adsorption and desorption, surface reaction, and desorption of products, has been studied extensively using modem surface-science techniques.17 The structure insensitivity of this reaction is illustrated in Figure 10.4. Here, carbon dioxide turnover frequencies over Rh(l 11) and Rh(100) surfaces are compared with supported Rh catalysts.3 As with CO hydrogenation on nickel, it is readily apparent that, not only does the choice of surface plane matters, but also the size of the active species.18-21 Studies of this system also indicated that, under the reaction conditions of Figure 10.4, the rhodium surface was covered with CO. This means that the reaction is limited by the desorption of carbon monoxide and the adsorption of oxygen. 

Catalysts prepared from iridium neutral binary carbonyl compounds and several supports have been studied extensively. Small Ir (x = 4, 6) clusters supported on several oxides and caged in zeolite, and their characterization by EXAFS, have been prepared [159, 179, 180, 194-196]. The nuclearity of the resulting metallic clusters has been related with their catalytic behavior in olefin hydrogenation reactions [197]. This reaction is structure insensitive, which means that the rate of the reac-hon does not depend on the size of the metallic particle. Usually, the metallic parhcles are larger than 1 nm and consequently they have bulk-like metallic behavior. However, if the size of the particles is small enough to lose their bulk-like metallic behavior, the rate of the catalytic reaction can depend on the size of the metal cluster frame used as catalyst. 

Since then, the group of structure-insensitive reactions has been very well documented by experimental data. This can be seen in several reviews (181, 223-225). It seems to be reliably established that reactions of simple molecules such as H2, CO, or S02 oxidations, HC/D2 exchange, and others, are mostly structure insensitive. Sometimes, the insensitivity is quite surprising, as with di-tm-butylacetylene hydrogenation (226). 

The results, presented in Fig. 10, are very similar to those already discussed for ethylene and ethane. Reaction (I), the hydrogenation of cyclopropane, has been shown earlier to be structure insensitive (103a, 103b). The activity pattern of this reaction is reminiscent of cyclohexane dehydrogenation (63). Initially, a small increase in activity is found, followed at 80% Cu by a rapid decline. These results show that reaction (II) is of the hydrogenolysis type and that reaction (I) is hydrogenation of an unsaturated bond. 

Benzene hydrogenation carried out at 140°C on a series of Pd/SiOz and Pd/Al203 catalysts is structure insensitive (165) [in agreement with others (196,197)], whereas very small Rh particles exhibited much lower activity than moderately and poorly dispersed Rh/Al203 catalysts (192). 

Xiao, F.-S., Weber, W. A., Alexeev, O., and Gates, B. C., Probing the limits of structure insensitivity Size-dependent catalytic activity of Al203-supported iridium clusters and particles for toluene hydrogenation. Stud Surf. Sci. Catal. 101,1135 (1996). 

The reaction is found to be zeroth order with respect to a-methylstyrene and approximately first order with respect to hydrogen in all solvents as shown in Table I. Reaction dependence on hydrogen in cyclohexane solvent is shown in Figure 2 and a typical Arrhenius plot is presented in Figure 3. Reaction rate is independent of Pd concentration (structure insensitive) in pure nonpolar solvents (cyclohexane, hexane (U.V.)) but becomes structure sensitive (i.e. dependent on Pd concentration) in solvents with impurities or which are more polar. The activation energy of 10.2 kcal/mol found in cyclohexane agreed well with the one determined by Germain et al. (6). 

Two structure insensitive reactions have been selected cyclohexene hydrogenation [6] oil surface Pt sites upon silica and but-l-ene isomerisation on acidic sites in bentonite. Both reactions were studied in differential reactors The former was investigated at 273-313K and lOlkPa Samples (5-lGmg) of catalyst were flushed with N2, pre-reduced in H at 423K for lh, flushed with N2 and then the reactant stream (lOlkPa total pressure cyclohexene N2 H2 = 1,7 89.5 10 1 200cm3.min 1 total flow rale) was introduced and analysed... 

The higher dehydrogenation activity keeps the 3-methylpiperidine concentration on the metal surface at lower levels thus suppressing the condensation reaction leading to the dimer. The increased (de)hydrogenation activity at higher dispersion can be explained in different ways. The rate of a structure insensitive reaction can be linearly correlated to the number of active sites thus to the dispersion. On the other hand metal particles with different shapes and dimensions could interact with molecules in the gas phase in a different way or could display different resistance against deposition of coke precursors. 

A little work on structure-insensitive reactions has been reported [18]. Both catalysts were very active for ethene hydrogenation, and rapid deactivation occurred even at 176 K. Ethyne and 1,3-butadiene react in a more controlled manner study of ethyne hydrogenation using both l4C-labeled ethyne and ethene showed that ethane formation took place directly from adsorbed ethyne, without the intervention of gas-phase ethene. 

Recently, FT synthesis reactions were shown to be independent of metal dispersion on Si02-supported catalysts with 6-22% cobalt dispersion (103). Turnover rates remained nearly constant (1.8-2.7 x 10 s ) over the entire dispersion range. Dispersion effects on reaction kinetics and product distributions were not reported. These tests were performed at very low reactant pressures (3 kPa CO, 9 kPa H2), conditions that prevent the formation of higher hydrocarbons and lead to methane with high selectivity and to CO hydrogenation turnover rates 10 times smaller than those obtained at normal FT synthesis conditions and reported here. These low reactant pressures also lead to kinetics that become positive order in CO pressure. Thus, the reported structure insensitivity (103) may agree only coincidentally with the similar conclusions that we reach here on the basis of our results for the synthesis of higher hydrocarbons on Co. 

Similar turnover rates recently reported on Co single crystals with different exposed crystal planes are consistent with the structure-insensitive nature of CO hydrogenation on Co (105, 106). Tlirnover rates on crystals with zig-zag grooved [Co (1120)] and close-packed [Co(OOOl)] surfaces differ by less than a factor of two, and CO hydrogenation activation energies are also similar on the two surfaces. The grooved surface favors chain growth... 

Acidic forms of zeolites are well suited as supports for metal functions which are employed for hydrogenation, since they can also withstand the presence of traces of sulfur compounds frequently found in feedstocks of petrochemical industry. It should be noted, however, that hydrogenation is a structure insensitive reaction so it will primarily depend upon the concentration of the accessible metal particles and the adsorption constant of the unsaturated hydrocarbon. This may offer an explanation as to why Pt catalysts, for example, are still active for hydrogenation, when their activity for dehydrocyclization or hydrogenolysis (i.e., for structure sensitive reactions) is completely lost (e.g., by poisoning). 

---