Metal Crystallite Size Research Articles

A microscopist's view of catalysis

Analytical electron microscopy, AEM, has provided valuable information on the structure, composition, location and size of metal clusters and/or crystallites supported on amorphous or crystalline metal oxide materials. Although this information is vital to the understanding of how a catalyst functions, the catalytic reaction is usually more complex. Many variables must be taken into account to evaluate properly a catalyst. Improper selection of process variables can make a good catalyst perform poorly. Consequently, focusing on specific metal site properties may not always be the appropriate direction to take when evaluating a supported metals catalyst problem.To help the microscopist appreciate the complexity of the problem, the following examples were chosen to illustrate this point. Reforming catalysis involves a complex set of reactions to convert naphtha feed to high octane gasoline. The catalyst is comprised of highly dispersed Pt on a high surface area alumina with Cl. The catalyst is dual functional requiring both metal site activity for dehydrogenation and acidity for isomerization and cyclization.

Microporous poly-( N,N-dimethyl-acrylamide)-(1-methacryloyl-ethylene-2-sulphonate)-( N,N'-methylene-bis-acrylamide) resins as hydrophilic supports for metal catalysts

The synthesis of poly-( N,N-dimethyl-acrylamide)-(1-methacryloyl-ethylene-2-sulphonate) (M4K) with 4 mol. % of N,N'-methylene-bis-acrylamide (MBAA) as crosslinker, and of poly-( N,N-dimethyl-acrylamide)-(1-methacryloyl-ethylene-2-sulphonic acid) (M4A) with 4 mol. % MBAA is described. Polymerization was performed under gamma rays irradiation of a water solution of monomers (M4K), or of the monomer mixture (M4A), at room temperature. High affinity to protonic solvents was observed for both materials. Palladium catalysts were prepared from M4K through ion-exchange with [Pd(NH 3) 4] 2+ and from M4A through ion-exchange with palladium acetate; the subsequent Pd(II) reduction with sodium borohydride in ethanol yielded the final catalysts with a uniform distribution of metal throughout the resin particles. The average size of metal crystallites obtained by X-ray diffraction was about 4 nm and did not show any dependence on metal content in the polymer in the range from 0.5 to 8.8 w w % of metal. The rate of hydrogenation of p-nitrotoluene and cyclohexene in catalytic tests with 1 and 0.5 w w % Pd catalysts was proportional to the metal content in catalysts, which confirmed a very good accessibility of these supports. Catalysts with low metal content also exhibit a significant deactivation in the course of the hydrogenation of p-nitrotoluene to p-toluidine in methanol under ambient conditions.

Chemisorption and catalytic parameters of alumina-supported copper oxide catalysts

Abstract The textural properties of CuO Al 2 O 3 catalysts were determined from the adsorption of nitrogen at 77 K. The surface area decreases and the mean pore radius increases with the increase of copper loading and with the rise of calcination temperature. CuAl2O4 was detected when the sample containing 20 wt% Cu was calcined at 1173 K. The data obtained from the chemisorption of hydrogen on reduced catalysts allowed the determination of the metal surface area, the crystallite size and the extent of metal dispersion. Hydrogenation of cyclopentadiene to cyclopentane, on the reduced catalysts depends on the efficiency of the dispersion of the metal at the surface of the support and on the metal surface area. This reaction depends also, but to a lesser extent, on the metal loading and crystallite size. The catalytic hydrogenation of cyclopentadiene is a structural insensitive reaction independent of the total surface area, as determined from the physical adsorption of nitrogen at 77 K on the catalyst.

A new product distribution formulation for fischer-Tropsch synthesis. Effect of metal crystallite size distribution

A new product distribution formulation is presented for Fischer-Tropsch synthesis (FTS). This formulation, which contains the effect of the metal crystallite size on the product selectivity, can explain all possible Fischer-Tropsch product distributions. The Anderson-Shulz-Flory distribution formulation can be regarded as its approximation in special cases.

Kinetics of heterogeneous catalytic hydroformylation of propylene on rhodium-cobalt catalysts. Effect of metal dispersity

Selectivity of heterogeneous catalytic hydroformylation of propylene on rhodium-cobalt catalysts increases with decreasing size of metal crystallites. The data are interpreted in terms of a reaction mechanism.

ChemInform Abstract: Structure‐Sensitivity of the Deactivation of Pd/SiO2 for Methylcyclopropane Hydrogenolysis by CO.

Silica-supported palladium catalysts ranging from 5 to 80% metal exposed have been characterized as to their reactivity for methylcyclopropane hydrogenolysis as a function of poisoning by CO. Both the level of surface coverage by CO and the metal crystallite size are seen to be important in the net deactivation behaviour of the catalyst. In addition, pretreatment conditions are important in determining patterns both of reaction and of deactivation structure sensitivity; these effects appear to be related to differences in surface morphology induced by both the atmosphere and the temperature of the final treatment before reaction.In general, CO poisoning in this system is structure sensitive: as CO coverage increases a monotonic decrease in activity occurs for large Pd particles, while catalysts with percentage metal exposed >40% exhibit a near-step drop in activity between 20 and 30% CO coverage. A maximum in turnover frequency is observed near 65% exposed metal for coverage levels >30%.

Evidence for the bimodal distribution of Pt metal crystallites on a zeolite support using XRD and HREM

Zeolites are of interest as supports for reforming catalysts because of their high selectivity and because they provide a means of achieving and maintaining high dispersion of noble metals. This translates into high activity and also the maintenance of activity for an extended period of time. One of the goals of catalyst characterization in general is to measure those properties assumed to be responsible for performance in a given reaction. In this study, we have applied powder x-ray diffraction (XRD) and high-resolution electron microscopy (HREM) techniques to determine the size, shape, distribution, and location of Pt metal crystallites on a synthetic zeolite support in an attempt to explain CO chemisorption results.

Structure-sensitivity of the deactivation of Pd/SiO2 for methylcyclopropane hydrogenolysis by CO

Silica-supported palladium catalysts ranging from 5 to 80% metal exposed have been characterized as to their reactivity for methylcyclopropane hydrogenolysis as a function of poisoning by CO. Both the level of surface coverage by CO and the metal crystallite size are seen to be important in the net deactivation behaviour of the catalyst. In addition, pretreatment conditions are important in determining patterns both of reaction and of deactivation structure sensitivity; these effects appear to be related to differences in surface morphology induced by both the atmosphere and the temperature of the final treatment before reaction.In general, CO poisoning in this system is structure sensitive: as CO coverage increases a monotonic decrease in activity occurs for large Pd particles, while catalysts with percentage metal exposed >40% exhibit a near-step drop in activity between 20 and 30% CO coverage. A maximum in turnover frequency is observed near 65% exposed metal for coverage levels >30%.

Calorimetric study of the coadsorption of hydrogen and carbon monoxide over ruthenium graphitized carbon black catalysts

In order to explain the different catalytic behaviour in the carbon monoxide hydrogenation of carbon-supported ruthenium particles with different metal crystallite sizes, a calorimetric study of carbon monoxide and hydrogen adsorption and coadsorption has been carried out. Preadsorbed hydrogen does not inhibit carbon monoxide adsorption while preadsorbed carbon monoxide largely prevents subsequent hydrogen adsorption. It was found that on the smaller ruthenium crystallites pre-covered with carbon monoxide molecules the hydrogen interaction is energetically stronger. This observation is used to postulate the presence of active centres with different adsorption properties, and consequently with different catalytic properties, depending on the metal crystallite size.

Support and crystallite size effects in ethylene oxidation catalysis

The influence of the support and metal crystallite size of silver catalysts under ethylene oxidation conditions is investigated. It is shown that the specific activity of silver can vary by as much as a factor of 50 depending on the carrier employed. Among the supports investigated, only α-Al 2O 3, ZrO 2, SiC, and TiO 2 of the rutile structure, were found to offer non-zero apparent selectivities towards epoxide formation. The influence of metal crystallite size on the kinetic parameters was also found to depend on the carrier employed. Thus, the specific activity of Ag/α-Al 2O 3 was found to exhibit a maximum at an average crystallite size of approximately 400A˚while that of Ag/SiO 2 increases with crystallite size and appears to level off at an average crystallite size of approximately 500A˚.

Selective leaching of NiAl3 and Ni2Al3 intermetallics to form Raney nickels

Annealed single-phase NiAl3 and Ni2Al3 materials were leached with 20wt% aqueous NaOH solution to remove the aluminium. At temperatures of 274 to 323 K, NiAl3 leached according to linear kinetics, yielding porous nickel which was friable and disintegrated. At these temperatures Ni2Al3 was unreactive, but at 343 to 380 K it leached according to parabolic kinetics, producing a strong, tightly adherent rim of residual material. The Ni2Al3 reaction proceeded in two steps, firstly to produce a two-phase mixture of Ni2Al3 plus nickel, and secondly to produce nickel alone. In both stages the detailed microstructure of the prior alloy was preserved, implying that the mechanism is selective dissolution. The surface adsorption properties of the nickel residues were obscured by reprecipitated aluminA. However, metal crystallite size measurements showed that a large nickel surface area was potentially available.

Diffusion Model of Solid-State Catalytic Hydrogenation of Organic Compounds. The Influence of the Size of Metal Crystallites

Abstract The kinetics of the reaction was experimentally studied in the platinum black/thymine system. It was shown that the data obtained and the kinetic data of reference2 could be interpreted by the diffusion model and not by the available models of catalyst poisoning. The diffusion model was developed for reactions on dispersed metals. We calculated the dependence of both the reaction zone specific volume and the maximum compound/catalyst ratio ensuring complete conversion of the initial reagent on the radius of the catalyst metal crystallites for a series of values of the process parameter.

Studies of Cu−Ni supported catalysts for enantioselective hydrogenation

The effect of support content and composition of supported metal phase on the crystallite size, phase composition and catalytic properties of supported Ni and Cu−Ni catalysts has been investigated. It has been established that with increasing metal content, crystallite sizes and enantioselectivity increase. Copper added to Ni catalysts decreases metal phase dispersity and enantioselectivity.

Influence of preparative procedure on the activity and selectivity of FeZSM-5 catalysts in syngas conversion

The ZSM-5 supported iron catalysts have been investigated with respect to physical and chemical properties and to activity and product selectivity for conversion of synthesis gas to hydrocarbon products. The catalyst preparations consisted of (1) direct decomposition of (C5H5Fe(CO)2)2 on ZSM-5, (2) impregnation of ZSM-5 with an aqueous solution of iron nitrate, and (3) physical admixture of precipitated iron oxide with ZSM-5. Some H2 and CO chemisorption studies and magnetization measurements were conducted to measure the average metal crystallite size. At the metal loading of about 8 wt%, the catalysts prepared by the first method produced higher metal dispersion compared to the other two methods of preparation, although the amount of aromatic compounds in liquid hydrocarbon products was very low. Infrared studies of chemisorbed pyridine on these samples indicate that ion exchange of Fen+ for acidic protons occurs in aqueous solution impregnated catalysts, while no ion exchange is found in organometallic impregnated or physically admixed catalysts.

Zeolite-encapsulated platinum catalysts: Preparation, characterization by transmission electron microscopy, and their shape selective behavior toward various nitrogen base poisons during the catalytic oxidation of aqueous formaldehyde

Procedures are described for the encapsulation of Pt metal crystallites by the hydrogen forms of zeolite Y, mordenite, and ZSM-5. The Pt metal crystallite size and distribution in these well-encapsulated Pt/zeolite samples have been established through the use of transmission electron microscopy. Particle size ranges for the Pt crystallites are as follows: Pt/HY, 10–20 Å; Pt/H-MOR, 6–15 Å; and Pt/H-ZSM-5, 5–20 Å. Supported Pt metal crystallites effectively catalyze the oxidation of aqueous formaldehyde to carbon dioxide and water in an autoclave reactor at 95 °C and 207 kPa O 2. The rate of this oxidation is extraordinarily sensitive to the presence of nitrogen bases that effectively poison the Pt sites. Nitrogen bases investigated in this study include glycine, ethylenedi-aminetetraacetic acid, pyridine, quinoline, and 4-methylquinoline. Reactor evaluation of well-encapsulated Pt/zeolite catalysts for formaldehyde oxidation in the presence of small amounts of these nitrogen bases demonstrates clearly that the Pt crystallites reside in regions of the zeolite structure accessible only through the crystalline zeolite pore structure. Nitrogen bases too large to penetrate the zeolite pores are found to have little or no effect on the formaldehyde oxidation rate. In the case of Pt/ZSM-5, poisoning by quinoline occurs rapidly but 4-methylquinoline shows little effect on the rate of formaldehyde oxidation even in high concentrations and for prolonged periods of exposure to the catalyst. Complete details will be given for these selective poisoning experiments with the encapsulated Pt/zeolite catalysts.

Hydrogenolysis of ethane over Ni/Al2O3 catalysts of high nickel dispersity

The effect of nickel dispersity on the rate of ethane hydrogenolysis and the number of B5 centers has been studied. A maximum in the rate of ethane hydrogenolysis is observed in the range of nickel dispersity of 2–3.5 nm. There is a minimum of the apparent activation energy of ethane hydrogenolysis in this range of metal crystallite size.

Influence of the crystallite size of nickel on the course of the hydrogenolysis of propane and n-butane over Ni/Al 2O 3 catalysts

The effect of the degree of nickel dispersion (crystallite size) on the reaction rate, selectivity and activation energy of the hydrogenolysis of propane and n-butane was studied. Maximum rates of propane and n-butane hydrogenolysis were observed in the range of nickel dispersion 3–4 nm and 2–3.5 nm, respectively. Minimum apparent activation energies of propane and n-butane hydrogenolysis were observed in the range of Ni crystallite size 2–3.5 nm. Maximum selectivity of the reaction towards methane was found in the same range of metal crystallite size.

Structural sensitivity of CO adsorption and [formula omitted] coadsorption on [formula omitted] catalysts

The surface structure of a series of Ni SiO 2 catalysts was probed by CO adsorption and H 2 CO coadsorption using quantitative chemisorption measurements and IR spectroscopy. Preparation method, pretreatment conditions, and metal crystallite size were found to affect the nature of the metal surface for crystallites in the range 2.5–9.0 nm. CO was found to adsorb in linear, bridge, and multiple-CO forms on these supported nickel catalysts. Multiple-CO adsorbs on “defect sites” of low coordination, which increase in number as the heterogeneity of the nickel surface increases. On large crystallites exhibiting a large fraction of bridge-CO adsorption, coadsorption of hydrogen aided in probing the types of surface planes present on the crystallites. A high frequency peak in the bridge-CO region was assigned to twofold bridge-CO on Ni(111) while a low-frequency peak was assigned to two- or fourfold bridge-CO on Ni(100) planes. The coadsorption of H 2 also resulted in an increase in the intensity of the peak for CO adsorption on defect sites, possibly due to increased dipole-dipole interactions between linear and subcarbonyl CO species. CO and CO H 2 adsorption were found to be sensitive probes for both surface smoothness in terms of the presence of defect sites and surface structure in terms of the types of planes exposed.

Intrapellet metal crystallite size distribution in impregnated catalysts: Nickel on alumina.

Intrapellet metal crystallite size distribution in impregnated catalysts: Nickel on alumina.

Preparation and characterization of well-defined Ni/SiO 2 catalysts

Silica-supported nickel catalysts were prepared by three different methods of preparation including impregnation and precipitation. The nature of the precursor compounds, as well as the activation of these catalysts were studied by electron microscopy, X-ray diffraction, thermal analysis, and infrared spectroscopy. From this battery of techniques it was found that the method of preparation affects the properties of the catalysts in four ways: 1) the type of precursor metal compound deposited on the support; 2) the decomposition/reduction pathway that the metal precursor takes to the metallic state; 3) the degree of precursor-support interaction resulting from preparation and pretreatment; and 4) characteristics of the reduced catalyst including metal crystallite size and size distribution.