Nickel-silica catalysts, structure

II. Texture and Structure of Nickel-Silica Catalysts 1. Introduction... 

Other approaches to the problem of the structure of nickel-on-silica catalysts revealed, however, that the above suggestions are only a first step to the truth. 

Various metal and metal oxide nanoparticles have been prepared on polymer (sacrificial) templates, with the polymers subsequently removed. Synthesis of nanoparticles inside mesoporus materials such as MCM-41 is an illustrative template synthesis route. In this method, ions adsorbed into the pores can subsequently be oxidized or reduced to nanoparticulate materials (oxides or metals). Such composite materials are particularly attractive as supported catalysts. A classical example of the technique is deposition of 10 nm particles of NiO inside the pore structure of MCM-41 by impregnating the mesoporus material with an aqueous solution of nickel citrate followed by calicination of the composite at 450°C in air [68]. Successful synthesis of nanosized perovskites (ABO3) and spinels (AB2O4), such as LaMnOs and CuMn204, of high surface area have been demonstrated using a porous silica template [69]. 

Nickel, on the other hand, on alumina and on silica supports was found to have only five nearby sulfurs (square pyramidal) with Ni-Mo coordination numbers from 1 to 1.5. Ni-Mo-S supported on carbon was observed to have Ni-S coordination numbers of 6 in a trigonal-prismatic configuration. In addition, Ni (at low Ni concentrations) was found to have one nearby Ni, which could indicate that, in some catalysts, Ni is present as pairs on the MoS2 surface. The overall structure of the Ni-Mo-S was believed to be similar to that of millerite (i.e., Ni is located in the center of the MoS edge in a square-pyramidal configuration, with one sulfur extending perpendicular to the surface) (62-64). 

A Ni/Si02 catalyst prepared by incipient wetness was shown to consist, initially, of the silica support filled with the nickel salt. On reduction the nickel particles coalesced to form large crystallites because of their weak interaction with the support. Coprecipitated catalysts, however, were composed of layered silicate structures that were more difficult to reduce to metallic nickel. When reduction did take place, though, the reduced crystallites were fixed within the confines of the support so small metal particles with a narrow size range were produced. ... 

Studies of catalysts deactivation by coke are abundant in the literature most of them are usually conducted at high temperatures (around 500°C) using metal catalysts supported on oxides with low surface area such as silica, aluminas or silica-alumina [2 and references therein]. The deactivation by coke of zeolite catalysts has also been studied and such studies have mostly been done for high temperature reactions such as the conversion of n-hexane or the isomerization of xylenes [2,4]. However, low temperature coke formation (20-25°C) combining the effect of high acidity and size specificity for a high coking component such as nickel, has not yet been considered from the point of view of the presence of compounded effects of crystalline structure and location of metal particles. 

Since catalytic oxidation is a surface reaction, an inexpensive support material is normally coated with the noble metal. The support material can be made of ceramic, such as alumina, silica-alumina, or of a metal, such as nickel-chromium. The support material is arranged in a matrix shape to provide high surface area, low pressure drop, uniform flow of the waste gas through the catalyst bed, and a structurally stable surface. Structures that provide these characteristics are pellets, honeycomb matrices, or mesh matrices. 

Cusumano et al. (128) studied the reaction over Pt on alumina and on silica supports and concluded that the TOF was about the same for both catalysts, which did show quite different atomic rates AR. The later work of Sinfelt et al. (269) on reactions over copper-nickel alloys led also to the suggestion that cyclohexane dehydrogenation over Ni does not require a large ensemble of surface atoms and thus may be structure insensitive on a geometric basis. For ethane hydrogenolysis studied on the same CuNi alloys, it was found that the activity decreased much more rapidly than did the fraction of Ni atoms on the surface of the alloys. This implies that ethane hydrogenolysis requires an ensemble of surface atoms and should show antipathetic structure sensitivity. This reaction will be discussed in connection with Fig. 15 (below). 

The selectivity of the metal sulfate catalyst is influenced by many factors besides its acidic property, such as geometric structure involving a pore structure, arrangement of basic sites, polarity of the surface, etc. For example, the relative values of the first-order rate constants (per imit acidity at pK — 3) of the depolymerization catalyzed by nickel sulfate, cupric sulfate, and silica-alumina were found to be 1100 300 1. The difference may be attributed to the differences in acid-base bi-functional catalysis of these catalysts. This view may be said to have originated in 1948 when Turkevich and Smith (45) showed that the isomerization of 1-butene to 2-butene is catalyzed by metal sulfates, sulfuric acid, phosphoric acid, etc., but little by acetic acid, hydrogen chloride, etc. The high catalytic activity of the catalysts of the former group is considered as due to acid-base bifunctional catalysis as illustrated by Fig. 14. Independently, Horiuti (45a) advanced the same idea... 

The nickel dispersion of the catalyst on alumina support was less than that on silica support. This may be due to the strong interaction between nickel and alumina and undeveloped support pore structure than that of silica support. However, high catalytic activity and resistance to carbon deposition were obtained on the nickel catalyst supported on alumina. This indicated that metal dispersion was not the decisive factor that influenced the catalyst performance. Actually, the catalytic performance of the catalysts were integrative effect of nickel loading, metal dispersion, support, promoter, preparation and activation. 

A mixture of Ni°/NiO, produced by thermal decomposition of nickel acetate, dispersed on either silica or cordierite supports, was found to be catalytically active for the decomposition of methane without the need for any pre-treatment. Other authors used Ni catalysts supported on zirconia to produce H2 and a high yield of multiwalled carbon nanotubes. Raman spectroscopy suggested that carbon nanotubes formed at temperatures higher that 973 K had more graphite-like structure than those obtained at lower temperatures. They also reported that feed gas containing methane and hydrogen caused slow deactivation of the catalyst, and carbon yield increased with increasing Hg partial pressure in the feed gas. For a commercial Ni catalyst (65% wt Ni supported on a mixture of silica and alumina) it was found that catalyst deactivation depends on the... 

---