Applied Catalysis A, General (v.196, #1)

Dehydrogenation of ethane with carbon dioxide over supported chromium oxide catalysts by Shaobin Wang; K Murata; T Hayakawa; S Hamakawa; K Suzuki (1-8).
The oxidative dehydrogenation of ethane into ethylene by carbon dioxide over an unsupported Cr2O3 and several supported Cr2O3 catalysts on metal oxides such as Al2O3, SiO2, TiO2, and ZrO2 was investigated and the effect of support on the catalytic activity was studied. The unsupported Cr2O3 shows medium catalytic activity in this reaction; the support will exert a quite different effect on catalytic behavior. The catalytic activity varies with the nature of supports. Cr2O3/SiO2 catalysts exhibit an excellent performance in this reaction. Cr2O3 loading also affects the catalytic activity; 8 wt.% Cr2O3/SiO2 catalysts can produce 55.5% ethylene yield at 61% ethane conversion at 650°C. Characterization indicates that the distribution of chromium oxide on supports and surface chromium species structure are influenced by the nature of supports. The acidity/basicity and redox property of catalysts determines the catalytic activity in the dehydrogenation of ethane by carbon dioxide.
Keywords: Ethane dehydrogenation; Ethylene; CO2; Cr2O3; Support;

Comparative study of vanadium aluminophosphate molecular sieves VAPO-5, -11, -17 and -31 by M. Hassan Zahedi-Niaki; S.M. Javaid Zaidi; S. Kaliaguine (9-24).
Vanadium aluminophosphate molecular sieves VAPO-5, -11, -17 and -31 were synthesized hydrothermally and fully characterized using XRD, A.A, SEM, N2 and n-butane adsorption, diffuse reflectance UV-Vis, XPS, 27 Al , 31 P , 51 V NMR and ESR techniques.XRD data of VAPOs were consistent with their corresponding AlPOs structural patterns and indicated that all VAPOs were very well crystallized. The 51 V NMR spectra of calcined air-exposed (hydrated) VAPOs samples revealed that vanadium cations are present as V5+ both in distorted octahedral and tetrahedral coordination and quite probably in a framework position. DRS UV-Visible spectroscopy shows that in low vanadium content calcined samples, V5+ ions are mostly in tetrahedral coordination without any presence of polymeric extraframework oxy-vanadium phase. ESR and XPS spectroscopies indicated the presence of residual V4+ in calcined samples.VAPOs were found to be active in vapor phase oxidation of toluene with molecular oxygen. VAPO-11 and -31 showed high oxidation catalytic activity and selectivity for benzaldehyde. The order of activity were VAPO-11 > VAPO-31 > VAPO-5 ≥ VAPO-17. VAPO-11 better selectivity is related to its unique structural properties compared to those of the other VAPOs synthesized in this study.
Keywords: Vanadium aluminophosphate molecular sieves; VAPO-5; VAPO-11; VAPO-17; VAPO-31; Framework vanadium; ESR; XPS; Catalytic partial oxidation; Toluene;

Different compositions of calcium–copper Ca10.5−x Cu x (PO4)7 (0 ≤  x  ≤ 1) and calcium–sodium–copper phosphates Ca10−x/2Na x Cu0.5(PO4)7 (0 ≤  x  ≤ 1) belonging to the whitlockite-type structure were synthesised and characterised. They crystallise in the rhombohedral space group R3c (Z  = 6) and are isostructural with the tricalcium phosphate β-Ca3(PO4)2. The catalytic behaviour of these series of phosphates was studied in butan-2-ol conversion. Pure tricalcium phosphate β-Ca3(PO4)2 was found to be inactive while the samples containing Cu2+ ions exhibited mainly a dehydrogenation activity producing methyl ethyl ketone. An increase in Cu2+ concentration in the catalysts improved the dehydrogenation activity. Structural characteristics of these Cu2+ ions were investigated by spectroscopic techniques and correlated to the catalytic behaviour. UV–VIS and EPR showed that the Cu2+ ions are distributed in the phosphate lattice between two different positions: the Ca(4) and the Ca(5) sites. Progressive substitution of Cu2+ located in the Ca(4) sites by sodium resulted in a decrease in the activity. The Ca9.5NaCu0.5(PO4)7 that contains cupric ions only in the Ca(5) sites is inactive. 31P MAS NMR investigations of Ca10.5(PO4)7 and Ca10Na(PO4)7 showed that the occupancy level of the Ca(4) sites does not modify notably the symmetry of the (PO4)3− groups. UV–VIS and XPS analysis performed on the samples after a catalytic test showed that the dehydrogenation reaction lowers the amount of Cu2+ ions in the phosphate by reducing them to Cu+. It was proven that the active sites are the Cu2+ ions hosted by the Ca(4) sites.
Keywords: Calcium–copper phosphates; Calcium–sodium–copper phosphates; Butan-2-ol dehydrogenation; 31 P NMR; UV–VIS; EPR characterisations;

Oxidation of ethane into acetaldehyde and acrolein over silica containing cesium and a very small amount of additives by Zhen Zhao; Yusuke Yamada; Atsushi Ueda; Hiroaki Sakurai; Tetsuhiko Kobayashi (37-42).
Many samples containing different elements M in the system of M/SiO2 (M : Si = 1 : 1000, molar ratio) or Cs/M/SiO2 (Cs : M : Si = 10 : 1 : 1000, molar ratio) were prepared and screened for the oxidation of ethane by use of oxygen as oxidant. It has been found that the elements M (M = V, Bi, In, Ga, P, Zr, Zn, La) can give good aldehyde yields in the Cs/M/SiO2 system. The promoting effects of different alkali metals (Li, Na, K, Rb, Cs) on the catalytic performance of V/SiO2 (V : Si = 1 : 1000) in ethane oxidation were investigated. Among them, cesium gave the best promoting effect on V/SiO2 for aldehyde formation. The presence of alkali metals increases the basicity and neutralizes the acid site of catalyst; thus it enhances the selectivity to acetaldehyde and controls the formation of formaldehyde. Increase in basicity promotes the cross-aldol condensation of acetaldehyde and formaldehyde to give acrolein. The pathway for acrolein formation is mainly through the cross-aldol condensation of acetaldehyde and formaldehyde over Cs/V/SiO2 catalysts.
Keywords: Ethane oxidation; Acetaldehyde; Acrolein; Silica; Alkali metal; A small amount of additives; Pathway for acrolein formation;

Oxidative dehydrogenation of ethylbenzene on activated carbon catalysts by M.F.R. Pereira; J.J.M. Orfão; J.L. Figueiredo (43-54).
A detailed kinetic study of the oxidative dehydrogenation of ethylbenzene in the presence of an activated carbon catalyst is presented. The partial pressures of ethylbenzene and oxygen, as well as temperature, were varied one at a time, covering a wide range of experimental conditions. The experimental results were found to be well described by a kinetic model in which the main reaction occurs by a redox mechanism involving quinone/hydroquinone groups on the surface of the carbon catalyst. The activation energies determined for the main reaction and for the formation of carbon oxides are comparable to those determined on coked oxide catalysts.
Keywords: Oxidative dehydrogenation; Activated carbon; Ethylbenzene;

The evolution of H2O during temperature programmed pyrolysis (TPP) of pyridine and pyrrole deposited NiMo and CoMo catalysts and γ-Al2O3 was investigated between room temperature and 1000°C. γ-Al2O3 influenced TPP of pyrrole but had little effect on that of pyridine. The oxidic and sulfided NiMo and CoMo catalysts influenced the TPP of both pyridine and pyrrole. This was supported by increased yields of H2O for the deposited catalysts compared to the undeposited catalysts. The mechanism of H2O formation involving one OH group, SH groups and ring hydrogens is proposed. The difference between the mechanisms of the TPP of pyrrole and pyridine, including a potential role of Brønsted acidic sites during the TPP of the latter, is discussed.
Keywords: Hydroprocessing catalysts; Nitrogen-rings; Pyrolysis; Regeneration;

Maximizing H2 production by combined partial oxidation of CH4 and water gas shift reaction by P.S. Maiya; T.J. Anderson; R.L. Mieville; J.T. Dusek; J.J. Picciolo; U. Balachandran (65-72).
A dense ceramic membrane reactor has been constructed to exclusively transport oxygen for the partial oxidation of methane (CH4) to syngas (a mixture of CO and H2) at temperatures of 850–900°C. H2 production is enhanced in a second catalytic reactor through the water gas shift reaction, in which the CO reacts with steam that is injected into the reactor at a controlled rate to produce CO2 and H2. Experiments and thermodynamic calculations were used to establish the optimal lower temperature (≈400°C) and lower steam-to-CO ratio (≈2) to achieve thermodynamic efficiency while maximizing H2 production. No unusual synergisms were observed by the combination of the two processes and the experimental results are in good agreement with thermodynamic predictions.
Keywords: Ceramic membrane reactor; Oxygen transport; Syngas; H2 production; Water gas shift reaction;

Stability of supported transition metal catalysts in the hydrogenation of nitriles by Yinyan Huang; Valeria Adeeva; Wolfgang M.H Sachtler (73-85).
Gas phase hydrogenation of acetonitrile and butyronitrile over supported transition metal catalysts was carried out in a gas phase microflow reactor at 1 and 10 bar; butyronitrile was also hydrogenated in a stirred autoclave at an initial pressure of 24 bar. Catalyst stability is higher in the liquid phase than in the gas phase operation, suggesting desorption of site blocking molecules into the liquid. Agglomeration of metal particles has been detected by TEM and EDX, with Pt/NaY it starts at 110°C, with Ru/NaY it is obvious at 160°C. Metal agglomeration and formation of an overlayer contribute to catalyst deactivation. Ru/zeolite catalysts are more active than Ru/Al2O3, both in the gas and the liquid phase. No significant effect of support acidity on the selectivity of the reaction has been detected; the results give no indication of carbide or nitrile formation as major causes of catalyst deactivation.
Keywords: Acetonitrile hydrogenation; Butyronitrile hydrogenation; Transition metals; Metal agglomeration; Overlayer formation;

A series of Bi x Gd1−x VO4 mixed oxide has been prepared by coprecipitation and characterised by XPS, IR spectroscopy and by the reaction of methanol oxidation used as a catalytic surface probe. Within the tetragonal phase domain (0 ≤  x  ≤ 0.64), zircon-type and scheelite-type structures are observed for x less than 0.3 or higher than 0.3, respectively. The zircon-type phase shows higher selectivities in methyl formate and carbon dioxide than the scheelite-type phase which mainly forms formaldehyde. The selectivities in methyl formate and carbon dioxide increase with the bismuth content in the zircon-type phase domain. Whatever these structures, the gadolinium substitution by the larger size bismuth cation induces an effect similar to that of the external pressure, leading to a shortening of the Bi–O and Gd–O distances which tends to increase the electronic density on these cations. However, a Bi → VO4 charge transfer which occurs in the Bi–O–V group partly compensates the preceding effect.
Keywords: Bismuth–gadolinium–vanadium mixed oxide; XPS; Electronic effects; Methanol oxidation;

Liquid-phase hydrogenation of citral over an immobile silica fibre catalyst by Tapio Salmi; Päivi Mäki-Arvela; Esa Toukoniitty; Ahmad Kalantar Neyestanaki; Lasse-Pekka Tiainen; Lars-Eric Lindfors; Rainer Sjöholm; Ensio Laine (93-102).
A new knitted silica fibre catalyst was introduced into liquid-phase hydrogenation of organic compounds. The fibre catalyst combines the advantages of catalyst slurries and large catalyst particles: diffusional limitations are suppressed and the catalyst is immobile. Nickel-impregnated silica fibres were used in the hydrogenation of citral to citronellal and citronellol in a tubular glass reactor where the catalyst was packed in two layers. The liquid phase was recirculated while the gas phase was flowing concurrently through the reactor to the vent. The highest selectivities to the desired product, citronellol, exceeded 90%. The catalyst performance was competitive with a conventional crushed nickel catalyst.
Keywords: Citral hydrogenation; Liquid phase; Silica fibre catalyst; Nickel;

The sulfidation state in a series of nickel- and cobalt-promoted Mo/Al2O3 catalysts was investigated using 35 S radioisotope pulse tracer methods (RPTM). Several pulses of 35 S -labeled H2S ([ 35 S ]H2S) were introduced into catalysts in a nitrogen stream until the radioactivity in the recovered pulse approached the radioactivity of the introduced pulse. From the amount of [ 35 S ]H2S introduced, the amount of sulfur accumulated on the catalyst was estimated. The amounts of sulfur accumulated on the catalysts increased with increasing temperature for all catalysts and the cobalt-promoted catalysts showed features similar to those of the nickel-promoted catalysts. It was found that the sulfided states of the catalysts close to the stoichiometric states, where Ni, Co and Mo are present as NiS, Co9S8 and MoS2, was attained at 400°C for all catalysts. Moreover, it was suggested that only molybdenum was sulfided below 300°C; the sulfidation of nickel oxide and cobalt oxide in the Ni– or the Co–Mo/Al2O3 catalyst is more difficult at lower temperatures and the sulfidation of nickel and cobalt species occurs only at higher temperatures (above 300°C).
Keywords: Sulfidation; Nickel- and cobalt-promoted Mo/Al2O3 catalysts; 35S-radioisotope pulse tracer method;

Metal–support interaction in Co/SiO2 and Co/TiO2 by Roberto Riva; Hans Miessner; Roberto Vitali; Gastone Del Piero (111-123).
Cobalt supported on silica and titania catalysts were investigated by XPS, TPR, TPD, XRD and TEM in order to elucidate the Co/support interactions and their effect on the dispersion and reducibility of cobalt. The reducibility of cobalt was studied both with TPR experiments, in which the temperature is raised at a steady rate, and with XPS after in-situ reduction treatments at constant temperature. Silica-supported catalysts were prepared with various Co loadings (2–27.5 wt.%) and they exhibited significant structural and morphological differences. A significant Co/titania interaction was found, while no conclusive proof of any interaction was found for Co/silica. The degree of interaction between cobalt and the support affected not only the response of cobalt to reduction, but also its dispersion. In fact, cobalt spreads on titania during reduction and tends to sinter on silica.
Keywords: Cobalt; Silica; Titania; Support; XPS;

Polymerization of acetone and acetylene, the decomposition products of methylbutynol (MBOH), on the surface of Y/MgO catalyst was inferred using high-temperature transmission FTIR spectroscopy. It was found that MBOH decomposes into acetone and acetylene revealing the basic surface character of Y/MgO catalyst. However, modification of MgO with Y3+ cations decreased the catalyst activity towards MBOH decomposition relative to that observed for pure MgO. The consecutive polymerization of the decomposition products is responsible for this low activity, since polymerized acetone is strongly adsorbed on the catalyst surface, and hence, blocks the surface active sites.
Keywords: Surface reactions; Acetone; Acetylene; Methylbutynol; Y/MgO catalyst;

Skeletal isomerization of n-butenes to isobutene over acid-treated natural clinoptilolite zeolites by Hyun Chul Lee; Hee Chul Woo; Ryong Ryoo; Kyung Hee Lee; Jae Sung Lee (135-142).
The proton form of the natural clinoptilolite zeolite (HNZ) was modified by treatment with various acids. Effects of modification on catalytic performance were investigated for the skeletal isomerization of n-butenes to isobutene. Among HNZs treated with various acids, only those modified with boric acid (B-HNZ) showed an improved catalytic performance of skeletal isomerization in terms of activity and stability. Furthermore, B-HNZ exhibited higher isobutene selectivity compared to unmodified HNZ at the same conversion of n-butenes. The performance was comparable to that of a widely known ferrierite catalyst at the same reaction conditions. The main effect of boric acid treatment on HNZ was the selective increase in the number of acid sites of moderate strength required for efficient catalysts in skeletal isomerization without collapse of its framework. These acid sites of moderate strength were considered to be selective and stable reaction sites for skeletal isomerization of n-butenes. Coke deposition poisoned non-selective strong acid sites and imposed a spatial restriction on zeolite channels, leading to suppression of undesired dimerization reactions. Factors affecting selectivity to isobutene are discussed.
Keywords: Clinoptilolite; Skeletal isomerization; Acidity; Selectivity; Coke deposition;

Deactivation kinetics of Mo-supported Raney Ni catalyst in the hydrogenation of xylose to xylitol by Jyri-Pekka Mikkola; Hanna Vainio; Tapio Salmi; Rainer Sjöholm; Tapio Ollonqvist; Juhani Väyrynen (143-155).
The activity of a Mo-promoted Raney nickel catalyst was studied in the hydrogenation of xylose to xylitol. Kinetic measurements carried out in a laboratory scale pressurised slurry reactor (40–70 bar H2 and 80–130°C) with recycled catalysts revealed that the catalyst deactivates during the use, but an asymptotic activity level is finally attained. Water and water–ethanol mixtures were used as solvents. The formation kinetics of the main product, xylitol as well as the by-products, xylulose, d-arabinitol, furfural and xylonic acid were registered quantitatively in the experiments.Catalyst characterisation studies carried out with nitrogen adsorption, XRD, ESCA-XPS and gravimetric reduction with hydrogen suggested that the main reasons for the deactivation is the decay of accessible active sites through collapse of the pore structure and leaching of the promoter metal, Mo and alumina. Also, accumulation of organic species in the pores may slightly contribute to the deactivation process. Catalyst deactivation was more rapid in aqueous milieu than in water–ethanol solutions. The deactivation rate was retarded, if the catalyst was treated with ethanol at elevated hydrogen pressure and temperature between the hydrogenation experiments.A rate model based on plausible surface reaction mechanisms was proposed for the generation of the main and by-products. The rate equations were based on a semi-competitive adsorption model for hydrogen and organic species. The catalyst deactivation kinetics was described with a reversible semi-empirical model, which lumped the physical and chemical reasons for deactivation to a simple two-parameter system. The deactivation model was combined with the rate equations and the model of the slurry reactor. The kinetic and deactivation parameters were determined with a sequential technique, by using non-linear regression analysis. The model was able to reproduce the hydrogenation behaviour of Raney nickel very well: it predicted the product distribution and the catalyst deactivation within a wide range of process parameters.
Keywords: Xylose hydrogenation; Raney nickel; Kinetic modelling; Catalyst deactivation;