Selective oxidation sites densities

Having identified the existence of separate oxidation sites, it is desirable to determine the densities of sites which contribute to the different selectivities among different catalysts. Since the products of the thermal desorption experiments described above are directly correlated with the active sites, it is possible to measure the number of active sites by measuring the quantities of desorbed products, provided that each and every active site produces only one molecule of reaction product. To achieve this condition, it must be established that the adsorbate is fully equilibrated with the surface, that there is no multiple reaction per site during equilibration, that there is no readsorption and reaction of desorbed products, and that all reaction products are quantitatively determined. 

We also examine the structure sensitivity of CO hydrogeneration on Co and Ru crystallites supported on various metal oxide supports at conditions that favor high selectivity (>80%) to C5+ products. We describe procedures for the synthesis of catalytic materials with high active site densities and controlled intrapellet distributions of sites. Finally, we review the extensive literature that has previously described metal dispersion, support, and transport effects on FT synthesis rate and selectivity. 

The benefits of nonuniform activity distributions (site density) or diffusive properties (porosity, tortuosity) within pellets on the rate of catalytic reactions were first suggested theoretically by Kasaoka and Sakata (Ml). This proposal followed the pioneering experimental work of Maatman and Prater (142). Models of nonuniform catalyst pellets were later extended to more general pellet geometries and activity profiles (143), and applied to specific catalytic reactions, such as SO2 and naphthalene oxidation (144-146). Previous experimental and theoretical studies were recently discussed in an excellent review by Lee and Aris (147). Proposed applications in Fischer-Tropsch synthesis catalysis have also been recently reported (50-55,148), but the general concepts have been widely discussed and broadly applied in automotive exhaust and selective hydrogenation catalysis (142,147,149). 

Reasonable estimates of surface site density are important for accurate modeling of solute adsorption (Davis and Kent, 1990). The most common methods for estimating surface site density include measurement of adsorption maxima of protons, cations, or anions at pH values where adsorption of the specific solute is greatest. Most measurements of surface site density for Fe, Al, and Mn oxides range from 1 to 7 pmoles/m (Davis and Kent, 1990). Dzombak and Morel (1990) reviewed the adsorption literature for ferrihydrite and selected a surface site density of 3.84 j,moles/m in constructing a coherent thermodynamic database for adsorption of cations and anions by ferrihydrite. 

It can be seen that the activities of the amorphous alloys are lower than those of the polycrystalline catalysts. Formation of the corresponding diol was not observed on the amorphous catalysts, while the crystalline catalysts either produced the diol selectively, or a mixture of the diol and the hydroxy ketone was formed. The fundamental reason for the lower activity and higher selectivity of the amorphous alloys is their rather small surface area. Of the amorphous alloys studied, Ni-B and Ni-P alloy powders prepared by chemical reduction exhibited higher activities than those of Ni-P alloys prepared by electrolytic reduction or rapid quenching. This difference in activity can be attributed to an oxide layer covering the surface of m-P foils [Ij. It is necessary to point out, however, that the comparison of activities is based on unit catalyst weight. Obviously, this comparison does not take into account the real surface area of the nickel samples, nor active site densities. 

Acetic acid is known to undergo a vapor-phase ketonization reaction with formation of acetone on Brpnsted acids in general, and on proton-zeolites in particular. On large-pore zeolites in their proton form, the ketonization reaction is followed by acid-catalysed self-condensation amounting to mesitylene, mesityl oxide and phorone as main products [1], the chemistry being essentially identical to that in mineral acids. In H-pentasil zeolites with suitable acid site density, phorone isomerises to isophorone, which is cracked to yield 2,4-xylenol [1]. With propionic acid a similar chemistry occurs, but the formation of phenolics is severely suppressed by transition-state shape-selectivity effects... 

Theory has been used predominantly to probe the nature of the sites on vanadium clusters and model vanadium oxide surfaces. Cluster and p>eriodic DFT calculations [68,69] have been carried out in order to imderstand the electronic and structural properties of the exposed (100) surface of (VO)2P207. Both cluster and slab calculations reveal that surface vanadium sites can act as both local acid and base sites, thus enhancing the selective activation of n-butane as well as the adsorption of 1-butene. Vanadium accepts electron density from methylene carbon atoms and, thus aids in the subsequent activation of other C-H bonds. Calculations reveal that that the terminal P=0 bonds lie close to the Fermi level and thus present the most nucleophihc oxygen species present at the surface for both the stoichiometric as well as phosphate-terminated surfaces. These sites may be involved in the nucleophilic activation of subsequent CCH bonds necessary in the selective oxidative conversion of butane into maleic anhydride. Full relaxation of the surface, however, tends to lead to a contraction of the terminal P=0 bonds and a lengthening of the P V bonds. This pushes the P V states, initially centered on the oxygen atoms, higher in energy and thus increases their tendency to be involved in nucleophilic attack . 

Oxides commonly studied as catalytic materials belong to the structural classes of corundum, rocksalt, wurtzite, spinel, perovskite, rutile, and layer structure. These structures are commonly reported for oxides prepared by normal methods under mild conditions [1,5]. Many transition metal ions possess multiple stable oxidation states. The easy oxidation and reduction (redox property), and the existence of cations of different oxidation states in the intermediate oxides have been thought to be important factors for these oxides to possess desirable properties in selective oxidation and related reactions. In general terms, metal oxides are made up of metallic cations and oxygen anions. The ionicity of the lattice, which is often less than that predicted by formal oxidation states, results in the presence of charged adsorbate species and the common heterolytic dissociative adsorption of molecules (i.e., a molecule AB is adsorbed as A+ and B ). Surface exposed cations and anions form acidic and basic sites as well as acid-base pair sites [1]. The fact that the cations often have a number of commonly obtainable oxidation states has resulted in the ability of the oxides to undergo oxidation and reduction, and the possibility of the presence of rather high densities of cationic and anionic vacancies. Some of these aspects are discussed in this chapter. In particular, the participation of redox sites in oxidation and ammoxidation reactions and the role of redox sites in various oxides that are currently pursued in the literature are presented with relevant references. 

The following sections describe the development of a chemisorption method that applies the dissociative chemisorption of methanol to quantitatively determine the density of active surface sites of metal oxide catalysts for methanol selective oxidation. 

The most recently developed anode for the cathodic protection of steel in concrete is mixed metal oxide coated titanium mesh The anode mesh is made from commercially pure titanium sheet approximately 0-5-2mm thick depending upon the manufacturer, expanded to provide a diamond shaped mesh in the range of 35 x 75 to 100 x 200 mm. The mesh size selected is dictated by the required cathode current density and the mesh manufacturer. The anode mesh is supplied in strips which may be joined on site using spot welded connections to a titanium strip or niobium crimps, whilst electrical connections to the d.c. power source are made at selected locations in a suitably encapsulated or crimped connection. The mesh is then fitted to the concrete using non-metallic fixings. 

Ab initio methods allow the nature of active sites to be elucidated and the influence of supports or solvents on the catalytic kinetics to be predicted. Neurock and coworkers have successfully coupled theory with atomic-scale simulations and have tracked the molecular transformations that occur over different surfaces to assess their catalytic activity and selectivity [95-98]. Relevant examples are the Pt-catalyzed NO decomposition and methanol oxidation. In case of NO decomposition, density functional theory calculations and kinetic Monte Carlo simulations substantially helped to optimize the composition of the nanocatalyst by alloying Pt with Au and creating a specific structure of the PtgAu7 particles. In catalytic methanol decomposition the elementary pathways were identified... 

---