Oxidative dehydrogenation of propane over (Mo)-Sm-V-O catalytic system. Role of the different phases

B. P. Barbero¹, R. Prada Silvy² and L. E. Cadús¹

¹ Instituto de Investigaciones en Tecnología Química (INTEQUI), UNSL – CONICET, Casilla de correo 290, 5700, San Luis, Argentina.
E-mail: bbarbero; lcadus@unsl.edu.ar
² Université Catholique de Louvain, Croix du Sud 2, 1348 Louvain-la-Neuve, Belgium.

Abstract — The (Mo)-Sm-V-O catalytic system has been exhaustively studied in the propane oxidative dehydrogenation reaction. In order to obtain different surface arrangements, simple oxides (V₂O₅, Sm₂O₃ and MoO₃), SmV mixed oxides with different Sm/V molar ratio and SmVO₄ impregnated with vanadium, samarium or molybdenum were prepared. The function that the possible arrangements play has been identified. A slight samarium excess favors total combustion causing a strong drop of selectivity, therefore, it is necessary to avoid it. On the other hand, vanadium excess constituting surface vanadium oxide species (VO_x) notably increases the catalytic activity while a higher vanadium amount leads to crystalline V₂O₅ formation and the catalyst behavior tends to that of bulk V₂O₅. Molybdenum at low concentrations constitutes surface molybdenum oxide species (MoO_x) which showed to be highly selective in propane ODH. High contents of molybdenum favor the formation of crystalline MoO₃, thus, causing an important catalyst deactivation. Finally, a comparison with other known efficient vanadium based catalysts is made and hence, the potentiality of (Mo)-Sm-V-O catalysts is shown.

Keywords — Oxidative Dehydrogenation. Propane. Vanadium. Samarium. Molybdenum.

I. INTRODUCTION

One of the major interests to the refining and petrochemical industry is the upgrading of cheaper feedstock. In this sense, the oxidative dehydrogenation (ODH) reaction of light alkanes so as to obtain the corresponding olefin is an attractive alternative (Kung, 1994; Cavani and Trifiro, 1995; 1997; Albonetti et al., 1996; Baerns and Buyevskaya, 1998). However, the practical application of this type of process still remains an unresolved scientific and technological challenge in spite of the broad research effort devoted to this goal. The main aim is to obtain olefin selectivity and productivity to make the process economically feasible. Thus, the catalyst should be optimized in order to minimize the unselected oxidation pathways which lead to low yields.

Among the large number of catalytic systems essayed in the ODH of light alkanes, vanadium based catalysts are the most efficient whether the vanadium oxide is deposited on an oxide support or combined with other oxides (Mamedov and Cortés Corberán, 1995; Bañares, 1999). The physico-chemical properties of such catalysts, and hence, its catalytic performance, are highly dependent on dispersion, repartition, coordination, and oxidation state of V atoms (Wachs and Weckhuysen, 1997; Kung and Kung, 1997; Blasco and López Nieto, 1997; Grzybowska-Swierkosz, 1997; Khodakov et al., 1998; Burrows et al., 1999; Solsona et al., 2001a, b; Argyle et al., 2002; Keränen et al., 2002).

In previous papers (Barbero and Cadús, 2002a, b, 2003a, b), the (Mo)-Sm-V-O solid system has been exhaustively characterized by studying the possible physico-chemical arrangements. In this work, the performance of (Mo)-Sm-V-O catalytic system in ODH of propane is compiled and the role of the different phases is deduced. A comparison with other known efficient vanadium based catalysts is also made and hence, the potentiality of (Mo)-Sm-V-O catalysts is shown.

II. EXPERIMENTAL SECTION

A. Catalysts preparation

Simple oxides. V₂O₅ was prepared by calcination in air of NH₄VO₃ (Mallinckrodt) at 450^oC for 4 h. Sm₂O₃ and MoO₃ from commercial origin were used.

Mixed oxides. Samples with atomic ratios Sm/V = 0.8, 1 and 1.2 were prepared by the citrate method (Courty, 1973). Sm(NO₃)₃.6H₂O (Fluka), NH₄VO₃ (Merck) and citric acid (Aldrich) were used as reagents. A citric acid solution with a 10% excess over the number of ionic equivalents of cations was prepared. Transparent solutions of samarium nitrate and ammonium metavanadate were prepared and added to the citric acid solution in such concentration that the Sm/V atomic ratio was equal to that in the resulting catalysts. The resulting solution was concentrated in a Rotavapor at 70^oC under reduce pressure until the solution became viscous. Then, it was dried to obtain amorphous organic precursor. The solid obtained was decomposed in air at 400^oC for 4 h and finally calcined at 600^oC for 3 h. These catalysts were denoted as SmVx, where x = 0.8, 1 or 1.2 indicates the Sm/V atomic ratio.

Impregnated catalysts. SmVO_4, obtained by the citrate method as described above, was impregnated with:

1. An aqueous solution of Sm(NO₃)₃.6H₂O by the dry impregnation method. The resulting solid was dried in vacuum at 70^oC for 2.5 h and then, calcined at 400^oC for 3 h and at 600^oC for 3 h.
2. An aqueous solution of NH₄VO₃ (Merck) by incipient wetness impregnation in multiple steps. After each impregnation step, samples were dried at 100^oC and then, dried solids were calcined at 600^oC for 12 h.
3. An aqueous solution of (NH₄)₆Mo₇O₂₄.4H₂O by dry impregnation. Impregnated samples were dried in vacuum at 50^oC for 1 h and then, calcined in air at 450^oC for 3 h and at 550^oC for 3 h.

The resulting catalysts were denoted as xM/SmVO₄, being x = 3 or 12 mol% and M = Sm, V or Mo.

B. Catalytic activity

The catalysts were tested in a fixed – bed, U-type quartz microreactor operated at atmospheric pressure. In each test, 500 mg of sample were used. The temperature was varied between 350 and 550^oC. The feed was a mixture of 4 vol.% propane and 4 vol.% oxygen balanced with helium. The total flow rate was 50 ml min^-1 measured at room temperature. The reaction products were analyzed by a Delsi Di 200 on-line gas chromatograph equipped with a thermal conductivity detector. Helium was the carrier gas. Two columns were used: a Hayesep Q column to separate hydrocarbons and CO₂, and a Carbosieve SII column to separate oxygen, carbon monoxide and methane. The homogeneous contribution was tested in the empty reactor. These runs showed no activity below 550^oC. Results were similar with and without the use of quartz particles. The measurements was made after at least 30 min time-on-stream, which has been verified it is sufficient time to achieve the steady state. The conversion of propane, X (%), is defined as the percentage of propane feed which has reacted. The selectivities, S (%), are defined as the percentage of propane that reacts to a specific product.

III. RESULTS

The catalytic behavior of all the catalysts in the propane oxidative dehydrogenation was studied. The results are shown in Table 1. Propylene, CO₂ and CO were the main products. At high reaction temperatures, cracking products (C₂H₄ and CH₄) were also detected. No other oxygenated products were found.

Simple oxides. V₂O₅ was an active catalyst in the propane oxidative dehydrogenation but its selectivity to propylene rapidly decreases with the increase of propane conversion (S_C3H6 decreases from 81% to 14% as X_C3H8 increases from 1.5% to 35%). Thus, the yield to propylene was lower than 5% although the reaction temperature was above 500^oC. Sm₂O₃ was very active but its selectivity to propylene was practically negligible, being CO₂ the main product obtained. The selectivity to cracking products was also high. MoO₃ was not very active under the conditions used in this study. The conversion was only 0.5% at 550^oC.

Mixed oxides. SmVx catalysts showed a great activity in propane ODH. Propane conversion was similar on all the catalysts (x = 0.8, 1 and 1.2) in the range of tested reaction temperatures. However, selectivity depended strongly on the Sm/V atomic ratio. SmV0.8 presented high selectivity to propylene at low conversion levels but selectivity decreased quickly as conversion increased. This behavior was similar to that observed on pure V₂O₅. SmV1 showed relatively good selectivity at low conversion (S_C3H6 = 54% at X_C3H8 = 1.9%) and it decreased slightly as conversion increased. The selectivity to propylene on SmV1.2 drastically dropped being CO₂ the major reaction product. At reaction temperatures higher than 450^oC, the selectivity to cracking products was also important (>1.5%).

Comparing the yield to C₃H₆ as a function of the Sm/V atomic ratio at different reaction temperatures, it has been observed that the curve presents a maximum at a ratio Sm/V = 1/1.

Impregnated catalysts. SmVO₄ used as support of the impregnated catalysts resulted slightly different to SmV1. At low reaction temperatures, both propane conversion and selectivity to propylene on SmVO₄ were lower than on SmV1. The results on impregnated catalysts indicate that the SmVO₄ catalytic behavior is significantly modified by the addition of Sm, V or Mo.

The 3Sm/SmVO₄ catalyst was less active that SmVO₄ and the selectivity to C₃H₆ dropped drastically. High levels of CO were obtained, however, the main product of reaction was CO₂. Considering the variation of S_C3H6 as a function of X_C3H8, the behavior was similar to that of SmV1.2 catalyst.

The addition of 3% vanadium over SmVO₄ increased notably the activity at low reaction temperatures (7% conversion at 350^oC) and the level of selectivity to propylene was relatively good (57%). At low conversion levels, catalysts impregnated with vanadium exhibited better selectivity and propylene yield than SmVO₄. Selectivity decreased as conversion increased. This decrease was more marked when vanadium loading increased becoming similar to the behavior of pure V₂O₅. Below certain reaction temperature, the selectivity to propylene improved significantly due to vanadium impregnation.

SmVO₄ with 3% molybdenum was almost as active as pure SmVO₄ but it showed a much higher selectivity to propylene (S_C3H6 = 84% at 1.5% conversion and 350^oC reaction temperature). Selectivity to propylene decreased with the increase of conversion and this decrease was more rapid as molybdenum loading increased. Propane conversion also decreased with the increase of molybdenum loading. The highest yield to propylene was obtained with 3Mo/SmVO₄.

IV. DISCUSSION

The nature of (Mo)-Sm-V-O catalytic system has been thoroughly studied in previous works. The results indicate that this system presents promissory alternatives for the propane oxidative dehydrogenation.

Table 1. Results of catalytic test.

Figure 1. Selectivity to propylene as a function of the propane conversion on V₂O₅ and SmVO₄.

Among the simple oxides studied in this work (V₂O₅, Sm₂O₃ and MoO₃), only V₂O₅ showed a relevant activity but the yield to propylene was lower than 5%. The formulation of a more complex catalytic system, as it is the case of SmVx mixed oxides, could yield a best performance. In fact, yield to propylene on SmVO₄ pure phase resulted more than the double that on V₂O₅.

The active centers on V₂O₅ as well as on SmVO₄ are constituted by vanadium ions. However, its catalytic behavior in propane ODH is noticeably different (Fig. 1). This indicates that the type of coordination that the vanadium presents in each oxide determines its performance as catalyst. This observation is coherent with the different V-O bonds existing. The vanadium ions present octahedral coordination in V₂O₅ (Haber et al., 1997) and there are terminal oxygen atoms (V=O) while in SmVO₄, vanadium ions are in tetrahedral coordination and there are only bridging oxygen atoms (V-O-V).

In order to explain the function each oxygen atom type (or the coordination type of the vanadium ions) plays in the oxidative dehydrogenation, it is convenient to consider the scheme of parallel and consecutive reactions proposed by several research teams (Kung, 1994; Blasco and López Nieto, 1997) and that can be represented as follows:

where k₁, k₂ and k₃ are the kinetic constants. In agreement with this scheme, the selectivity at low conversion levels will be related to the k₁/k₂ ratio, while at high conversion levels the selectivity will be associated to the k₁/(k₂+k₃) ratio.

Blasco and López Nieto (1997) have obtained theoretical curves of selectivity to propylene as a function of the conversion of propane varying the k₂/k₁ and k₃/k₁ ratios. Comparing these curves with the results obtained experimentally on our catalysts, it can be observed that the behavior of SmVO₄ corresponds to a k₂/k₁ ratio greater than 1, i.e. there is an important contribution of the reaction of direct combustion of propane. On the other hand, the k₃/k₁ ratio would be very small, indicating that propylene consecutive combustion is almost negligible. The curve of selectivity to propylene as a function of propane conversion obtained on V₂O₅ (Fig. 1) is comparable to that theoretically obtained with k₂/k₁ ratio ca. 0 and k₃/k₁ ratio approx. 10. This indicates that propane direct combustion is negligible but propylene consecutive combustion is relevant.

Consequently, either terminal oxygen atoms or octahedrally coordinated vanadium ions would be responsible for propylene combustion, while tetrahedrally coordinated vanadium ions would be more selective.

From these observations, with regard to a catalyst design it can be suggested that the challenge is to find a host matrix in which the cation constituting the active site is in a desirable coordination. Then, the way to increase the amount of these sites would be found. Likewise, samarium showed to be an appropriate element, since V₂O₅ (octahedrally coordinated vanadium ions) easily reacts with samarium to constitute SmVO₄ (tetrahedrally coordinated vanadium ions) (Barbero and Cadús, 2002a). Other advantageous feature of SmVO₄ is its high thermal stability with respect to V₂O₅.

With the aim of optimizing the oxide structure holding the catalytically active vanadium iones, the effect of Sm-V-O system non-stoichiometry was studied. It has been found that when the composition of SmVx mixed oxides is different from the stoichiometry corresponding to SmVO₄, the catalytic behavior varies notably. The selectivity to propylene on the sample with samarium excess (SmV1.2) drops drastically. It has been reported (Zhang et al., 1999) that an excess of Y₂O₃ in nYVO catalysts prepared by the citrate method is responsible for total oxidation. Unlike what happens with MgO in MgVO catalysts, Y₂O₃ does not promote the propylene desorption. This would be due to the strong basic sites existing on Y₂O₃ which are very active in the total oxidation (Zhaorigetu et al., 1996). Thus it could be suggested that free samarium present in SmVO₄ favors the direct combustion of propane causing the decrease of the selectivity to propylene. To confirm this, a probable arrangement including SmVO₄ and Sm₂O₃ was synthesized artificially by impregnating SmVO₄ with samarium (3Sm/SmVO₄). As a matter of fact, a similar behavior to SmV1.2 was obtained on this catalyst.

Another fact to highlight is that the selectivity to cracking products on SmV1.2 increases significantly. This would be due to the presence of samarium excess since pure Sm₂O₃ also showed a high selectivity to cracking products.

On the other hand, the sample with vanadium excess (SmV0.8) exhibits a great increase of selectivity to propylene at low conversion levels (around 20% at 3% conversion) with respect to SmV1. However, the selectivity decreases with increasing conversion similarly to that on pure V₂O₅. The preparation method of SmV0.8 catalysts does not assure about how vanadium excess is distributed. Vanadium could have remained dispersed in a matrix of SmVO₄ or it could have segregated to the surface during the preparation, resulting in a SmVO₄ recovered by vanadium. In an attemp to study in detail the behavior of this last arrangement, catalysts were prepared by impregnating SmVO₄ with vanadium.

SmVO₄ used as support of impregnated catalysts was prepared by the citrate method in the same way as SmVx catalysts. Although this synthesis method confers high purity and good reproducibility, this SmVO₄ resulted slightly different to SmV1. By an enlarged physico-chemical characterization that has been reported (Barbero and Cadús, 2003a), it was found that SmVO₄ contains a slight excess of samarium on the surface and this explains the differences on the propylene selectivity at initial levels of conversion (X_C3H8~0%) between the SmVO₄ used as support of impregnated catalysts (S_C3H6~36%) and SmV1 (S_C3H6~ 55%).

The impregnation of SmVO₄ with vanadium increases notably the selectivity to propylene at low conversion levels. This indicates that propane direct combustion would be controlled. Probably, the impregnated vanadium blocks the active sites responsible for direct combustion. An important increase of conversion on 3V/SmVO₄ at low reaction temperatures was also observed. The differences found in the catalytic behavior of 3V/SmVO₄ and 12V/SmVO₄ make us think that different vanadium species exist on the surface as a function of vanadium loading. The characterization of these catalysts has been made by several techniques and reported (Barbero and Cadús, 2003a). Briefly, it has been determined that the vanadium surface coverage in 3V/SmVO₄ catalyst is lower than the theoretical monolayer (approx. 50 %) and that vanadium is found as surface vanadium oxide species (VO_x) characterized by a reduction signal at about 500^oC in the temperature programmed reduction profile and the appearance of bands at 980-1010, 680-720 and 505-540 cm^-1 in Raman spectrum. In the case of 12V/SmVO₄ catalyst, the surface coverage is higher than two theoretical monolayers. The formation of V₂O₅ crystals was observed by XRD and confirmed by the results of temperature programmed reduction studies, the appearance of a band at 1020 cm^-1 in the FT-IR spectra and the characteristic bands of bulk V₂O₅ in Raman spectra. Consequently, the increase of the activity at low reaction temperature could be attributed to surface VO_x species while V₂O₅ crystals on the surface of 12V/SmVO₄ catalyst would be the reason why the catalytic behavior of the latter is similar to that of pure V₂O₅.

Taking into account that the behavior of SmV0.8 catalyst is also similar to that of bulk V₂O₅, it can be suggested that the vanadium excess has segregated during the preparation forming V₂O₅ crystals. This would indicate that SmVO₄ cannot include a higher amount of vanadium ions than that corresponding to stoichiometry. Consequently, one of the optimal structures of Sm-V-O system in which vanadium ions would exist in tetrahedral coordination, would be pure SmVO₄. Then, the addition of one or more cations could improve SmVO₄ performance. As it has been discussed above, surface vanadium species increase SmVO₄ catalytic activity. The addition of another appropriately chosen element which helps to design a surface architecture similar to that observed with vanadium could be of interest.

Molybdenum has been reported as a highly selective element in dehydrogenation reactions forming Mg-Mo-O (Yoon et al., 1996; Abello et al., 1996; Cadús et al., 1996, 1997) and Mo-Mn-O (Cadús and Ferretti, 2000, 2002) systems. The promotive effect of molybdenum on V-Mg-O (Oganowski et al., 1996; Dejoz et al., 1999) has been studied as well. Moreover, SmVO₄ was impregnated with different molybdenum loadings. As with vanadium, the object here was to discriminate different species that may be formed on the surface and to determine the role that these species play in propane ODH.

The results of catalytic activity show that the impregnation of SmVO₄ with 3% Mo does not modify significantly propane conversion but brings about a marked increase of yield to propylene (12.3% at 500^oC). As in the case of impregnated catalysts with vanadium, it has been calculated that the surface coverage is lower that 50% of the theoretical monolayer. From Raman spectroscopy results reported in elsewhere (Barbero and Cadús, 2003b), the existence of surface molybdenum oxide species (MoO_x) was detected. These species would be very selective to propylene and, consequently, they are responsible for a higher yield. With higher molybdenum loadings (12% Mo), a remarkable decrease of conversion as well as of yield and selectivity to propylene is observed. The surface coverage of this catalyst was about two theoretical monolayers. Both Raman spectroscopy and TPR studies show that molybdenum is present on 12Mo/SmVO₄ catalyst as surface polymolybdate species and MoO₃ crystals. The formation of MoO₃ crystals on the surface would explain of the catalytic behavior since pure MoO₃ is practically inactive in propane oxidative dehydrogenation under the experimental conditions used in this study. Propane conversion was only 0.5% at 550^oC.

It has been reported that propane direct combustion step brings about CO₂ mainly, while propylene consecutive combustion produces both CO and CO₂. Taking this into account and also the scheme of parallel and consecutive reactions presented above, a way to

compare the performance of several catalysts could be based on the values of CO/CO₂ ratio. Thus, it can be observed that the impregnation of SmVO₄ with vanadium causes an increase of the average CO/CO₂ ratio from 0.6 to the range of 1.2-1.7. This ratio grows with the increase of vanadium loading, approaching the average value of 2.0 obtained with the V₂O₅. In the case of catalysts impregnated with molybdenum, the average CO/CO₂ ratio is also higher than that obtained on SmVO₄ but lower than that resulting on catalysts with vanadium (0.7-1.0 range). Therefore, it can be inferred that vanadium impregnation favors the consecutive combustion of propylene and simultaneously controls the propane combustion. The same conclusion was reached comparing the shape of curves of selectivity to propylene as a function of the conversion of propane with the theoretical curves reported by Blasco and López Nieto (1997). Molybdenum produces a similar effect, i.e. it decreases the rate of propane combustion causing an increase of selectivity at low conversion levels and favors the consecutive combustion of propylene but to a smaller extent than vanadium. Thus, it can be explained why it is possible to obtain the highest yield (13%) on the catalyst with low molybdenum loading.

Finally, the high yield showed by (Mo)-Sm-V-O catalysts led us to compare their performance with the other vanadium based catalysts. Although a direct comparison of propylene yields is difficult since the results reported in literature have been obtained under very different conditions (reaction temperature, C₃H₈:O₂ ratio, W/F, etc.), it can give an idea about the potentiality of our system. Among vanadium based catalysts, V-Mg-O catalysts are known for exhibiting high efficiency in ODH of propane (Chaar et al., 1988; Gao et al., 1994, Corma et al., 1992, 1993). Good results have been also obtained with supported vanadium oxide catalysts (Corma et al., 1992; Eon et al., 1993, 1994; Concepción et al., 1995; Solsona et al., 2001a) and different rare earth vanadate (Au et al., 1996; Fang et al., 1999). Table 2 shows the higher yields presented by several research teams. It is observed that propylene yield levels obtained with our (Mo)-Sm-V-O catalysts are among the best ones reported in literature. In consequence, we think that this catalytic system has a great potentiality for being employed in propane ODH. Further studies tending to optimize both the catalyst and the operative conditions could lead to enhance propylene yield and productivity.

Table 2. Results of catalytic activity reported in the literature.

V. CONCLUSIONS

The (Mo)-Sm-V-O catalytic system has been evaluated in propane ODH and the function that each of the possible arrangements plays has been identified. Briefly, vanadium ions into SmVO₄ structure are catalytically more convenient than in V₂O₅ when these oxides are used as catalysts of propane oxidative dehydrogenation. A slight samarium excess favors the total combustion causing a strong drop of selectivity, therefore it is necessary to avoid it. On the other hand, vanadium excess constituting surface vanadium species (VO_x) increases notably the catalytic activity while at higher vanadium loadings, crystalline V₂O₅ is formed and catalyst behavior tends to that of bulk V₂O₅. Molybdenum is an appropriate element to improve SmVO₄ performance when it is at low concentrations constituting surface molybdenum species (MoO_x) which showed to be highly selective in propane ODH. High contents of molybdenum favor the crystalline MoO₃ formation, which causes an important catalyst deactivation.

The ways to attain a higher increase of propylene yield in order to make this reaction type competitive, would be:

• To optimize the catalyst design including elements in a host matrix able to stabilize the cations constituting the catalytic sites in a suitable coordination.
• To optimize the reactor design and the selection of the operative conditions.

ACKNOWLEDGEMENTS
Financial support from CONICET and from Universi-dad Nacional de San Luis are gratefully acknowledged.

REFERENCES
1. Abello, M.C., M.F. Gómez and L.E. Cadús, "Oxidative dehydrogenation of propane over molybdenum supported on MgO - γ Al₂O₃", Ind. Eng. Chem. Res., 35, 2137-2143 (1996).         [ Links ]
2. Albonetti, S., F. Cavani and F. Trifiro, "Key aspects of catalyst design for the selective oxidation of paraffins", Catal. Rev. Sci. Eng., 38 (4), 413-424 (1996).         [ Links ]
3. Argyle, M.D., K. Chen, A.T. Bell and E. Iglesia, "Effect of catalyst structure on oxidative dehydrogenation of ethane and propane on alumina-supported vanadia", J. Catal., 208, 139-149 (2002).         [ Links ]
4. Au, C.T., W.D. Zhang and H.L. Wan, "Preparation and characterization of rare earth orthovanadates for propane oxidative dehydrogenation", Catal. Lett., 37, 241-246 (1996).         [ Links ]
5. Baerns, M. and O. Buyevskaya, "Simple chemical processes based on low molecular-mass alkanes as chemical feedstocks", Catal. Today, 45, 13-22 (1998).         [ Links ]
6. Bañares, M.A., "Supported metal oxide and other catalysts for ethane conversion: a review", Catal. Today, 51, 319-348 (1999).         [ Links ]
7. Barbero, B.P. and L.E. Cadús, "Evaluation and characterization of Sm-V-O catalytic system for propane oxydehydrogenation: Sm₂O₃-impregnated V₂O₅ catalysts", Appl. Catal. A, 234, 245-258 (2002a).         [ Links ]
8. Barbero, B.P. and L.E. Cadús, "V₂O₅-SmVO₄ mechanical mixture: oxidative dehydrogenation of propane." Appl. Catal. A, 237, 263-273 (2002b).         [ Links ]
9. Barbero, B.P. and L.E. Cadús, "Vanadium species: Sm-V-O catalytic system for oxidative dehydrogenation of propane", Appl. Catal. A, 244, 235-249 (2003a).         [ Links ]
10. Barbero, B.P. and L.E. Cadús, "Molybdenum role: Mo-Sm-V-O catalytic system for propane oxidative dehydrogenation", Appl. Catal. A, 252, 133-147 (2003b).         [ Links ]
11. Blasco, T. and J.M. López Nieto, "Oxidative dehydrogenation of short chain alkanes on supported vanadium oxide catalysts", Appl. Catal. A, 157, 117-142 (1997).         [ Links ]
12. Burrows, A., C.J. Kiely, J. Perregaard, P.E. Hojlund-Nielsen, G. Vorbeck, J.J. Calvino and C. López-Cartes, "Structural characterisation of a VMgO catalyst used in the oxidative dehydrogenation of propane", Catal. Lett., 57, 121-128 (1999).         [ Links ]
13. Cadús, L.E. and O. Ferretti, "Highly effective molybdena-manganese catalyst for propane oxidative dehydrogenation", Catal. Lett., 69, 199-202 (2000).         [ Links ]
14. Cadús, L.E. and O. Ferretti, "Characterization of Mo-Mn-O catalyst for propane oxidative dehydrogenation", Appl. Catal. A, 233, 239-253 (2002).         [ Links ]
15. Cadús, L.E., M.C. Abello, M.F. Gómez and J.B. Rivarola, "Oxidative dehydrogenation of propane over Mg-Mo-O catalysts", Ind. Eng. Chem. Res., 35, 14-18 (1996).         [ Links ]
16. Cadús, L.E., M.F. Gómez and M.C. Abello, "Synergy effects in the oxidative dehydrogenation of propane over MgMoO₄ – MoO₃ catalysts", Catal. Lett., 43, 229-233 (1997).         [ Links ]
17. Cavani, F. and F. Trifiro, "Some aspects that affect the selective oxidation of paraffins", Catal. Today, 36, 431-439 (1997).         [ Links ]
18. Cavani, F. and F. Trifiro, "The oxidative dehydrogenation of ethane and propane as an alternative way for the production of light olefins", Catal. Today, 24, 307-313 (1995).         [ Links ]
19. Chaar, M.A., D. Patel and H.H. Kung, "Selective oxidative dehydrogenation of propane over V-Mg-O catalysts", J. Catal, 109, 463-467 (1988).         [ Links ]
20. Concepción, P., J.M. López Nieto and J. Pérez-Pariente, "Oxidative dehydrogenation of propane on VAPO-5, V₂O₅/ALPO₄-5 and V₂O₅/MgO catalysts. Nature of selective sites", J. Mol. Catal. A: Chem., 99, 173-182 (1995).         [ Links ]
21. Corma, A., J.M. Lopez-Nieto, N. Paredes, M. Perez, Y. Shen, H. Cao and S.L. Suib, "Oxidative dehydrogenation of propane over supported vanadium oxide catalysts", Stud. Surf. Sci. Catal., 72, 213-220 (1992).         [ Links ]
22. Corma, A., J.M. López Nieto and N. Paredes, "Influence of the preparation methods of V-Mg-O catalysts on their catalytic properties for the oxidative dehydrogenation of propane", J. Catal., 144, 425-438 (1993).         [ Links ]
23. Courty, Ph., H. Ajot, Ch. Marcilly and B. Delmon, "Oxydes mixtes ou en solution solide sous forme très divisée obtenus par décomposition thermique de précurseurs amorphes", Powder Technol., 7, 21-38 (1973).         [ Links ]
24. Dejoz, A., J.M. López Nieto, F. Márquez and M.I. Vázquez, "The role of molybdenum in Mo-doped V-Mg-O catalysts during the oxidative dehydrogenation of n-butane", Appl. Catal. A, 180, 83-94 (1999).         [ Links ]
25. Eon, J.G., P.G. Pries de Oliveira, F. Lefebvre and J.C. Volta, "A comparison between gamma-alumina and aluminium niobate supported vanadium oxides in propane oxidative dehydrogenation", Procceding of the II World Congress and IV European Workshop Meeting, New Developments in Selective Oxidation., B.1 (1993).         [ Links ]
26. Eon, J.G., R. Olier and J.C. Volta, "Oxidative dehydrogenation of propane on γ -Al₂O₃ supported vanadium oxides", J. Catal., 145, 318-326 (1994).         [ Links ]
27. Fang, Z.M., Q. Hong, Z.H. Zhou, S.J. Dai, W. Zheng and H.L. Wan, "Oxidative dehydrogenation of propane over a series of low-temperature rare earth orthovanadate catalysts prepared by the nitrate method", Catal. Lett., 61, 39-44 (1999).         [ Links ]
28. Gao, X., P. Ruiz, Q. Xin, X. Guo and B. Delmon, "Effect of coexistence of magnesium vanadate phases in the selective oxidation of propane to propene", J. Catal., 148, 56-67 (1994).         [ Links ]
29. Grzybowska-Swierkosz, B., "Active centres on vanadia-based catalysts for selective oxidation of hydrocarbons", Appl. Catal. A, 157, 409 (1997).         [ Links ]
30. Haber, J., M. Witko and R. Tokarz, "Vanadium pentoxido I. Structures and properties", Appl. Catal. A, 157, 3-22 (1997).         [ Links ]
31. Keränen, J., A. Auroux, S. Ek and L. Niinistö, "Preparation, characterization and activity testing of vanadia catalysts deposited onto silica and alumina supports by atomic layer deposition", Appl. Catal. A, 228, 213-225 (2002).         [ Links ]
32. Khodakov, A., J. Yang, S. Su, E. Iglesia and A.T. Bell, "Structure and properties of vanadium oxide – zirconia catalysts for propane oxidative dehydrogenation", J. Catal., 177, 343-351 (1998).         [ Links ]
33. Kung, H.H., "Oxidative dehydrogenation of light (C₂ to C₄) alkanes", Adv. Catal., 40, 1-38 (1994).         [ Links ]
34. Kung, H.H. and M.C. Kung, "Oxidative dehydrogenation of alkanes over vanadium-magnesium-oxides", Appl. Catal. A, 157, 105-116 (1997).         [ Links ]
35. Mamedov, E.A. and V. Cortés Corberán, "Oxidative dehydrogenation of lower alkanes on vanadium oxide-based catalysts. The present state of the art and outlooks", Appl. Catal. A, 127, 1-40 (1995) and references therein.         [ Links ]
36. Oganowski, W., J. Hanuza, H. Drulis, W. Mista and L. Macalik, "Promotional effect of molybdenum, chromium and cobalt on a V-Mg-O catalyst in oxidative dehydrogenation of ethylbenzene to styrene", Appl. Catal. A, 136, 143-159 (1996).         [ Links ]
37. Solsona, B., T. Blasco, J.M. López Nieto, M.L. Peña, F. Rey and A. Vidal-Moya, "Vanadium oxide supported on mesoporous MCM-41 as selective catalysts in the oxidative dehydrogenation of alkanes", J. Catal., 203, 443-452 (2001a).         [ Links ]
38. Solsona, B., A. Dejoz, M.I. Vazquez, F. Márquez and J.M. López Nieto, "SiO₂-supported vanadium magnesium mixed oxides as selective catalysts for the oxydehydrogenation of short chain alkanes", Appl. Catal. A, 208, 99-110 (2001b).         [ Links ]
39. Wachs, I.E. and B.M. Weckhuysen, "Structure and reactivity of surface vanadium oxide species on oxide supports", Appl. Catal. A, 157, 67-90 (1997).         [ Links ]
40. Yoon, Y.S., W. Ueda and Y. Moro-oka, "Selective conversion of propane to propene by the catalytic oxidative dehydrogenation over cobalt and magnesium molybdates", Topics in Catal., 3, 265-275 (1996).         [ Links ]
41. Zhang, W., C. Au and H. Wan, "The active phase and support effect on V₂O₅/Y₂O₃ catalysts for propane oxidative dehydrogenation", Appl. Catal. A, 181, 63-69 (1999).         [ Links ]
42. Zhaorigetu, B., R. Kieffer and J.P. Hindermann, "Oxidative dehydrogenation of propane on rare earth vanadates. Influence of the presence of CO₂ in the feed", Stud. Surf. Sci. Catal., 101, 1049-1052 (1996).        [ Links ]