A. López Agudo

Temperature-programmed reduction and zeta potential studies of the structure ofMo/O3Al2O3 andMo/O3SiO2 catalysts effect of the impregnation pH and molybdenum loading

Abstract The influence of the impregnation pH (at 2–11) and molybdenum content (6–18 wt.-% Mo03) on the surface structure ofMo/O3Al2O3 andMo/O3Si02 catalysts has been studied by temperature-programmed reduction, X-ray diffraction and zeta potential measurements. The results indicate that molybdena was relatively well dispersed on A1203 support, the dispersion being worse for the catalysts prepared in acidic medium than for those in basic medium; formation of a crystalline phase was observed only at a pH of 2 and for Mo03 loadings > 14 wt.-%. However, on SiO2 support only crystalline MoO3 species were found, irrespective of the preparation conditions. In this case the temperature-programmed reduction profiles suggest the presence of two Mo03 bulk phases which may probably differ in crystal size, and whose proportion depends on the pH of the impregnating solution.

Thiophene hydrodesulfurization on sulfided Ni, W and NiW/USY zeolite catalysts: effect of the preparation method

Abstract The effect of the preparation method and metal loading on the hydrodesulfurization ⊙HDS) of thiophene, and on the dispersion and location of Ni and/or W species in Ni, W and NiW sulfides supported on USY zeolite, was investigated by temperature-programmed reduction ⊙TPR), X-ray photoelectron spectroscopy, N 2 adsorption and acidity determination. The monometallic Ni catalyst prepared by ion exchange [Ni ⊙int)] exhibited higher initial HDS activity than those prepared by impregnation [Ni ⊙imp)] and by ion exchange and subsequent formation of small NiO clusters by NaOH treatment [Ni ⊙clus)]. For the binary NiW catalyst series with different metal contents, the catalysts prepared by impregnating W onto the Ni ⊙int) one showed activities comparable to those of the catalysts prepared by impregnating Ni onto the W/USY sample. These catalyst series showed much higher activities than those prepared by impregnating W onto the Ni ⊙clus). The characterization results revealed that catalysts displayed significant differences in dispersion, location and sulfidation of the Ni and W phases dependent on the preparation method. However, none of these factors alone accounted for the activity trend. In general, it was found that a combination of high acidity and metal dispersion led to catalysts with high initial HDS activity.

Effect of phosphorus concentration and method of preparation on the structure of the oxide form of phosphorus-nickel-tungsten/alumina hydrotreating catalysts

Two series of phosphorus containing NiW/Al2O3 catalysts were prepared by different preparation methods and varying phosphorus content from 0 to 7.6 wt.% P2O5. The influence of the phosphorus concentration and the preparation procedure on the structure and the dispersion of the compounds formed in the oxide form of the catalysts was studied. Diffuse reflectance spectroscopy (DRS), infrared spectroscopy (IR) and X-ray photoelectron spectroscopy (XPS) were used for the characterization of the catalysts. It was demonstrated that introduction of phosphorus in NiW/Al2O3 catalysts impedes the formation of NiAl2O4 and increases the amount of Ni2+(Oh) ions in the oxide form of the samples. This effect of phosphorus is better expressed when phosphorus is first introduced to alumina, followed by co-impregnation with nickel and tungsten. Surface AlPO4 is formed in the phosphorus containing samples irrespective of the preparation procedure. Phosphorus is distributed as a monolayer up to concentrations of 1.3 P at/nm2 in both series of catalysts. It was also found that the increased phosphorus content in the samples leads to an increased degree of polymerization of the tungsten species via WOW bonds. The presence of phosphorus changes the dispersion of the active components. Evaluation of the average particle size of tungsten species shows that it increases from 10Afor the phosphorus free sample up to 20–25Afor the samples with high phosphorus content. The relationship between the structure of the phosphorus promoted NiW/Al2O3 catalysts and their hydrodesulfurization activity is discussed.

Effect of support on the surface characteristics of supported molybdena catalysts

Abstract Various supported molybdena catalysts with submonolayer loading on TiO2, Al2O3, SiO2, SiO2 Al2O3, and carbon have been prepared by both impregnation and equilibrium adsorption procedures. The characterization of the oxide form of the catalysts by several techniques (X-ray diffraction, XPS, ESR, O2 chemisorption, and H2 reduction) revealed that the oxidic Mo phase was highly and homogeneously dispersed, essentially in monolayer form, on TiO2; well-dispersed and coordinatively heterogeneous on Al2O3; and poorly and heterogeneously spread on SiO2, SiO2 · Al2O3, and carbon, in which the presence of three-dimensional MoO3 crystallites was detected. Catalyst reducibility decreased in the order carbon > TiO2 ⪢ SiO2 > SiO2 · Al2O3 > Al2O3 for the series prepared by impregnation. For the equilibrium adsorption series, the ranking was very similar except that TiO2 and carbon changed places. Reducibility was governed by both the Mo dispersion and the molybdena-support interaction which was found to be weaker for TiO2 and carbon than for Al2O3. The conjunction of both high Mo dispersion and weak Mo-support interaction leads generally to high reducibility.

Preparation and characterization of equilibrium adsorption-prepared molybdena-alumina catalysts

Abstract The preparation of molybdena-alumina catalysts was studied using an equilibrium adsorption method at 50 °C in which the molybdenum anions from solution were adsorbed on the alumina surface at pH values between 5.70 and 7.70 near to its isoelectric point. In these experimental conditions a two-step adsorption isotherm was found, indicating that the loading of the carrier was very dependent on the solute concentration. By combining different characterization techniques, namely, temperature-programmed reduction and kinetics of reduction, reflectance spectroscopy, chemisorption of NO and O 2 probe-molecules, infrared spectroscopy of NO, and extraction by ammonia, it is shown that at low solute concentrations the Mo is highly dispersed, mostly as tetrahedrally coordinated (by oxygen ions) Mo species. However, poorer dispersion was found at intermediate and high solute concentrations. In this latter case, the combined use of chemisorption and extraction data indicated that on calcining at 500 °C the surface of the carrier becomes populated with both tetrahedrally (monolayer) and octahedrally (small clusters) coordinated Mo-species.

Determination of the active surface area of vanadia by electrophoretic migration and XPS measurements

Electrophoresis migration and X-ray photoelectron spectroscopy techniques were applied to measure the active surface area of vanadia in titania- and alumina-supported vanadium oxide catalysts. For V-oxide contents ranging between 0 and 2.3 × 1014 cm−2 for Al2O3 and 0 and 7.2 × 1014 cm−2 for TiO2, it was found that V-oxide has a very high degree of dispersion mainly as monolayer. The V 2pT 2p and V 2pAl 2p intensity ratios, taken from the XPS spectra, increase monotonically with the V content. However, for VTiO2 catalysts the straight line declines for V contents above 4.9 × 1014 cm−2, thus indicating the end of the first vanadium layer. For the VAl2O3 catalyst series both apparent surface coverage and V 2pAl 2p intensity ratio steadily increase with V content, indicating that in all these preparations, even with the highest V content, the V-layered structures do not fully cover the surface of the carrier.

Effect of fluorine on hydrogenation of cyclohexene on sulfided Ni (or Co)Mo/Al2O3 catalysts

The effect of fluorine incorporation on alumina support on the surface structure of unpromoted molybdenum, promoted cobalt—molybdenum and nickel—molybdenum catalysts, and their activity for hydrogenation of cyclohexene has been studied. The incorporation of 0.2 and 0.8 wt.-% fluorine on the alumina was carried out by impregnation with NH4F solutions. The catalysts in the oxidic state were characterized by X-ray diffraction and ammonia adsorption and in the sulfide state by X-ray photoelectron spectroscopy (XPS) and infra-red spectroscopy (IR) of adsorbed NO. The absence of significant changes in the binding energy values of Mo3d and Ni2p (or Co2p) levels in the XPS spectra of the fluorine-containing catalysts as compared to the fluorine-free counterpart does not support the existence of an electronic effect of fluorine. The quantitative XPS results showed, however, that fluorine clearly increases the dispersion of molybdenum and promoter, this being linearly correlated to surface fluorine content. The IR results of adsorbed NO also indicate that fluorine incorporation leads generally to minor sizes of MoS2 slabs, and more exposed promoter atoms, except for the cobalt in the CoMo/F(0.2)A catalyst. It is suggested that the increase in the dispersion of the supported active phase is a secondary effect of fluorine incorporation, which may result from the observed textural changes of the alumina and its small partial solubilization provoked by NH4F solution. It was found that the incorporation of fluorine enhances appreciably, moderately and considerably the hydrogenation activity of molybdenum, cobalt—molybdenum an nickel—molybdenum catalysts, respectively. Such increase in hydrogenation activity is not directly correlated to the exposed atoms probed by NO adsorption, and is only loosely related to molybdenum dispersion for the molybdenum and cobalt—molybdenum catalysts. The lack of similar reliable correlations for the nickel—molybdenum catalysts suggests that other structural parameters such as extent of reduction-sulfidation and certain configurations of molybdenum ions and sulfur vacancies may govern hydrogenation activity.

Physicochemical characterization of MoO3-NaY zeolite catalysts

Abstract A series of Mo-impregnated NaY zeolite catalysts, with a Mo loading in the range 1–15 wt%, MoO3, have been studied by X-ray diffraction, transmission electron microscopy, ir and diffuse reflectance spectroscopy, temperature-programmed reduction, specific surface area, water content, and extraction measurements, in order to characterize them structurally. The crystallinity, specific surface area, and water content of the Mo-NaY were found to decrease almost linearly with increasing Mo loading. These effects were accompanied by an increasing perturbation of the zeolite framework. The ir spectra and, particularly, the diffuse reflectance spectra suggest that Mo is predominantly present as a tetrahedrally coordinated MoO42− species, presumably within the zeolite cavities. The Mo extraction and temperature-programmed reduction results revealed the existence of a strong interaction between Mo species and zeolite, which apparently decreases with increasing Mo loading.

Migration of molybdenum into intracrystalline cavities in molybdate-impregnated NaY zeolite

Abstract Molybdenum-containing Y-type zeolites have been prepared by thermal decomposition of molybdate-impregnated samples according to a nonconventional procedure which uses a constantrate decomposition at very low water vapor pressure (10 N m−2). The combined results of X-ray diffraction, mid-infrared, surface area, and sorption (water and n-hexane) capacity measurements show that zeolite structure was largely retained when the impregnated material was activated at temperatures below 623 K. However, a progressive loss of crystallinity and a parallel decrease of the surface Mo Si ratio, as obtained from the X-ray photoelectron spectra, was observed at higher temperatures. Furthermore, on heating at higher temperatures (793 K), the electron spin resonance spectra indicated that a fraction of the Mo(V) ions are essentially isolated in the sodalite cages or hexagonal prisms of the zeolite. A mechanism which involves “vaporization” of MoO3 from the external surface of the zeolite by water vapor is proposed to describe the redistribution of molybdenum in the impregnated precursor.

Effect of fluoride on the structure and activity of NiW/Al2O3 catalysts for HDS of thiophene and HDN of pyridine

Abstract The effect of adding different amounts of fluoride to a γ-alumina used as support of NiW/Al 2 O 3 catalysts was investigated using X-ray diffraction, diffuse reflectance spectroscopy (DRS), infrared spectroscopy of adsorbed NO (IR-NO), X-ray photoelectron spectroscopy (XPS), surface acidity, specific surface area and catalytic activity measurements. The catalytic activity was evaluated by two model reactions, namely thiophene hydrodesulfurization (HDS) and pyridine hydrogenation (HDN), in the temperature range 573–623 K using a flow system operated at 2 MPa (HDS) or 3 MPa (HDN) total pressure. DRS results of the oxidic precursors suggested that the effect of fluoride incorporation to the alumina support is that it increases slightly the octahedrally coordinated fractions of Ni and W. The XPS results of the sulfided catalysts showed that the dispersion of the Ni and W phases decreased progressively with increasing fluoride content. This technique also showed that these phases were only partially sulfided under the present experimental sulfiding conditions, and that the extent of sulfidation of such phases apparently tends to decrease slightly with increasing fluoride content, in contrast with the results suggested by the IR-NO spectra. The sulfur content of sulfided samples, as measured by chemical analysis, revelated that the sulfidation degree increased only slightly with increasing fluoride content. The activity for thiophene HDS was almost invariant with fluoride addition, but the subsequent hydrogenation of the intermediate products (butenes) was found to decrease linearly with the fluoride content. Possible interpretations for these changes in activity and selectivity, which reflect changes in the relative distribution of the different active sites for the two reactions, are discussed in relation with the structural variations of the active phases described above. In contrast, the activity for pyridine HDN increased considerably with the incorporation of small amounts of fluoride; this increase is mainly attributed to the increase in surface acidity which was observed with fluoride addition.