J. Chil. Chem. Soc., 48, N 2 (2003)

Ru/SiO₂ AND CuRu/SiO₂ PREPARED BY SOL-GEL:
EFFECT OF pH AND WATER AMOUNT.

Fresia Orellana, Jorge Plaza De Los Reyes y Silvio Urizar

Departamento de Química, Facultad de Ciencias, Universidad del Bío-Bío, Casilla 5-C, Concepción
(forellan@ubiobio.cl)
(Received: October 7,2002 ­ Accepted: January 25, 2003)

ABSTRACT

Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ catalysts were prepared by cogelation. The effect of pH and the amount of water on the physical-chemical properties and the catalytic properties of them were analyzed. The acid medium increased the formation of microporosity and the formation of small size of metallic ruthenium particle. The variation of the water quantity used in the gelation not modify considerably the texture neither the dispersion of the noble metal. The reduction level of all prepared catalyst was similar, showing displacement in some of reduction peaks. The toluene hydrogenation indicated that the conversion and the quantity of hydrogenate product varied with the pH and the quantity of water used in the gelation. The addition of copper to Ru/SiO₂ catalysts diminished the reactive conversion and their hydrogenant activity.

Key Words: Sol-gel, tetraethoxysilane, hydrodechlorynation, 1,2-dichloroethane, selectivity

INTRODUCTION

Ruthenium based catalysts have become an important subject matter in various research works (1). They are mainly used in hydrogenation and hydrogenolysis reactions because of their high selectivity (2). Unfortunately, these catalysts get easily deactivated due to carbonaceous material built (3) and/or the loss of the noble metal when being subjected to high temperatures (4). Because of what has been previously mentioned, achieving metallic ruthenium stabilization to be used as catalysts active component has become an interesting goal. Some authors (5,6) report positive autcomes, when noble metal catalysts are added an inactive metal, such as Cu, Ag or Au (5, 6). The second metal changes the noble metal catalytic behavior, which has been explained regarding changes in metals electronic features, or changes in active sites geometry (7).

The use of the so-called sol-gel technique, aimed at generating metallic catalysts, has made the metal direct incorporation to the support network (8). As a result, its holding down highly assists such metal's immovability. The advantage of this procedure consists on the catalyst texture control, thus making it possible to obtain small and homogenous metal particles (9). This technique takes place in two stages: first, the hydrolysis of the alcoxyde, and once this has been completed, a condensation takes place. Variation in the amount of water added for gelation will affect porosity development, while variation in pH alters hydrolysis rate. The catalyst texture also depends on the temperature used to set up the gel and to remove solvents that have been used in the homogenization of both the support and metal precursors.

Because of what has been previously stated, in this study the effect of copper addition to Ru/SiO₂ catalysts has been studied. Both Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ catalysts were obtained through sol-gel and by means of cogelation, respectively. The latter is used with the purpose of achieving a close mixture of both metallic components.

Copper has been chosen as bimetallic component due to its inactive behavior in the case of hydrogenation reactions (10), especially under the conditions that have been used in this study. It was evaluated the effect of pH in the preparation solution, as well as of the amount of water used for gelation on surface features, on texture characteristics and on the toluene hydrogenation activity of these catalysts.

EXPERIMENTAL

The Ru(1%)/SiO₂ catalysts have been prepared by dissolving tetraethoxysilane (TEOS) (Merck) in 5 ml of ethanol, and Ruthenium triacetylacetonate (Merck) in 5 ml of acetone. Both solutions, which had previously been heated up to 323 K, were mixed. Then, deionized water was added, so the resulting H₂O/TEOS molar ratio be 10 or 15. It was worked at pH=5, pH=1 (by adding HNO₃) and a pH=11 (by adding aqueous NH₃). The resulting mixture was stirred to 323 K until gelation took place and then it was kept at 323 K during one hour. Finally, the mixture was placed on a heater at 323 K during 12 hours. This allowed to eliminate of the remaining solvent (Xerogel). In order to get the Cu(1%)Ru(1%)/SiO₂ catalysts prepared, a similar procedure was used, the only variation of which being the addition of Cu(NO₃)₂x5H₂O (Merck) to the water that had been added for gelation purposes.

Pores surface area and total volume were obtained from Nitrogen adsorption measurements, at 77 K. A Geminis 2370 sorptometer was used. Samples had been previously reduced at 673 K during one hour and degassed at 373 K over a 2 hour period. Isotherms were recorded up to relative pressure equal to 0.95.

The Thermal Programmed Reduction (TPR) was performed in a TPD-TPR 2900 Micrometrics sorptometer. Samples were heated from room temperature up to 873 K, at a 10 /min heating rate. A H₂(5%)-Ar mixture, 40 cm³/min flux rate, was used as reducer gas. We can read in some bibliographic reports that during the calcination of the supported ruthenium catalysts, some volatile oxides are eliminated, which brings about some metal losses take place during this treatment (4). With the purpose of minimizing ruthenium volatilization, catalysts had not been previously subjected to calcination during this current research work.

Hydrogen chemisorption was carried out by means of an adsorption volumetric equipment furnished with a Validyne CD 23 pressure meter. Samples had previously been reduced at 673 K during one hour, and later degassed during a three-hour-period. Adsorption was done at 353 K, while making use of a H/Ru=1 adsorption stoichiometry. Some of the catalysts were assessed regarding the size of metallic ruthenium particles through transmission electronic microscopy (TEM). By doing so, replicas which were observed on a Jeol JEM-1200 EX microscope were used.

Catalytic activity was measured by means of a flow reactor. The toluene reactive (Merck) was carried over along with hydrogen flux (20 ml/min) up to the reactor. Toluene saturator was kept at 273 K. Catalysts were reduced in situ by means of hydrogen, at 673 K during one hour. Activity measurements were taken at atmospheric pressure and 573 K. Weight of the catalyst used for this purpose amounted to 0.15 g, while the H₂/Toluene molar ratio equaled 100. The converted reactive as well as the products obtained were analyzed by using a gas chromatograph (Schimadzu GC8A), along with a diisodecilftalate column.

RESULTS AND DISCUSSION

The kind of nitrogen adsorption isotherm obtained depend on the gelation pH rather than the amount of water used for hydrolysis. Water amounts higher than the stoichiometrics have been used in this current research work. Therefore, porosity build-up in the catalysts became possible. It has been reported (11) that the higher amount of water brings about solids featuring a larger support crossing-over and, therefore a higher heterogeneity regarding porosity. This fact might be the reason for the slight decrease in surface area because of the increased of used water. Figure 1 shows the nitrogen adsorption isotherms for Ru(1%)/SiO₂ (5,10) and Cu(1%)Ru(1%)/SiO₂ (5,10), both obtained at a pH 5 and a H₂O/TEOS ratio equal to 10. There we can observe that Ru(1%)/SiO₂ catalysts undergo an increase in the adsorbed volume at a high relative pressure and followed by plateau, which is, in turn, a characteristic of solid microporous. The greater the relative pressure is, the more the adsorbed volume increases, as the Type II and Type III isotherms, which features mesoporous or macroporous surfaces. Some catalysts, such as Ru(1%)/SiO₂ showed an adsorption-desorption hysteresis cycle which can be atribuited to capillary condensation in mesoporous solids. The Cu(1%)Ru(1%)/SiO₂ catalysts adsorption isotherms are more similar to the Type I ones, which is why there would be a higher contribution of microporous ones. Changes in the shapes of the nitrogen adsorption isotherms in the case of gels obtained at different pH were also observed. Such behavior has been explained because of changes in the hydrolysis rate (12). This reaction is accelerated and is completed when mineral acids are used as catalysts.

Figura 1. Nitrogen adsorption isotherms at 77 K for a) Ru(1%)/SiO2(5,10) and b) Cu(1%)Ru(1%)/SiO2(5,10)

The results of the nitrogen adsorption at 77 K have been reported in Table 1. The BET surface area, the average porous radius (r_p = 4V_p·1000/S_BET), as well as the percentage due to microporous have been included. We may also observe in this Table that the larger surface area, above 500 m²/g, is achieved for the Ru(1%)/SiO₂ catalysts, which have been obtained under acid pH conditions (1 and 5). No texture related modification was found when taking the amount of added water into consideration. With respect to Cu(1%)Ru(1%)/SiO₂ catalysts, specific surfaces turn out to be lower. However, there is a grater microporous contribution, along with a greater percentage of microporous surface (S_mi). This greater microporosity may be due to the presence of the Cu²⁺ cation in the catalysts preparation media. Small porous radius (r_p) have been achieved: smaller than 8.5 nm. Gels high porosity should improve the supported metals thermal stability.

Table 1. Surface area (S_BET), average pore radius (r_p) and percent surface micropore (S_mi) for different prepared Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ catalysts.

Catalyst	pH	H2O/TEOS	SBET (m2/g)	rp (nm)	Smi (%)
Ru(1%)/SiO2 (1,10)	1	10	626	4.5	15
Ru(1%)/SiO2 (5,10)	5	10	553	4.9	16
Ru(1%)/SiO2 (11,10)	11	10	172	7.2	27
Ru(1%)/SiO2 (1,15)	1	15	598	3.0	11
Ru(1%)/SiO2 (5,15)	5	15	498	5.5	17
Ru(1%)/SiO2 (11,15)	11	15	313	8.5	10
Cu(1%)Ru(1%)/SiO2 (1,10)	1	10	317	2.1	79
Cu(1%)Ru(1%)/SiO2 (5,10)	5	10	469	2.1	47
Cu(1%)Ru(1%)/SiO2 (11,10)	11	10	222	4.1	41
Cu(1%)Ru(1%)/SiO2 (1,15)	1	15	454	1.5	45
Cu(1%)Ru(1%)/SiO2 (5,15)	5	15	595	2.1	31
Cu(1%)Ru(1%)/SiO2 (11,15)	11	15	207	2.4	47

A ruthenium based neutral salt (ruthenium triacetylacetonate) was used as metal precursor for catalysts preparation. This is why metal, depending on the gelation pH will get trapped in a different way in the support. The reason of this is the different interaction taking place between metal and support. The kind of interaction affects the noble metal reducibility. In order to prevent the ruthenium oxides loss from taken place, the TPR of catalysts were carried out without oxidation pretreatment. The Ru(1%)/SiO₂ catalysts show a TPR (Figure 2) featuring two reduction areas, thus indicating the presence of ruthenium with a different thermal stability. There is a profile broad temperature interval, centered at about 523 K, the maximum of which neither moves with the pH nor with the H₂O/TEOS ratio. The ruthenium that has the weakest interaction with the support is reduced to this temperature (13). Such reduction profile amplitude has been attributed to an no homogeneous distribution of the metallic ruthenium particles (14). The other is a sharper reduction profile, centered around 673 K, and it correspond to a ruthenium featuring a stronger interaction with the support surface (13). Such profile maximum moves at a lower temperature as the gelation pH increases, while moving at a higher temperature when the amount of added water for hydrolysis also increases. The latter suggest us that increase in Ru-SiO₂ interaction. This profile brings about the greatest hydrogen consumption, which is why Ru(1%)/SiO₂ catalysts have a greater portion of ruthenium showing a stronger interaction with the support surface. Cu(1%)Ru(1%)/SiO₂ catalysts reduction behavior is slightly different. Likewise, there are two reduction profiles. The one featuring the lower temperature becomes sharper and has the greater hydrogen consumption represents ruthenium with the weakest metal-support interaction. In addition to this, and as it reads in bibliographical reports (15), some Cu²⁺ reduction may take place. Copper catalysts reduction has been observed at temperatures ranging from 553 to 653 K (15, 16). Second reduction profile turns out to be broader and moves at slightly higher temperatures, the lower hydrogen consumption of which might imply a smaller amount of ruthenium strongly trapped by the support. The removal of the organic precursors that have been used in gel preparation may be involved in the reduction profile amplitude. Besides, a different stoichiometry among surface metals may not be put aside in the case of bimetallic catalysts. Reduction profiles that show up at higher temperatures should be related with the small sized ruthenium particles, which, in turn, should be more difficult to reduce. The hydrogen total consumption, computed as the addition of the reduction peak areas per catalyst gram, is larger than the Ru(1%)/SiO₂ catalysts, so copper presence decreases ruthenium reduction capacity. In relation to Ru(1%)/SiO₂ catalysts, hydrogen total consumption increases the lower the preparation pH is. Therefore, these catalysts feature a higher reduction capacity, while in the case of Cu(1%)Ru(1%)/SiO₂ the higher reduction capacity can be found at a higher gelation pH. Both the electronic and geometric effect (17) are able to explain, either by themselves or as a whole, the properties of a bimetallic catalyst. An existing partially oxidized noble metal has a primary role regarding such activity and the selectivity of these catalysts as well.

Figura 2. TPR profiles of representative catalysts. a) Ru(1%)/SiO2(5,10), b) Cu(1%)Ru(1%)/SiO2(5,10), c) Ru(1%)/SiO2(1,10), d)Cu(1%)Ru(1%)/SiO2(1,10)

Hydrogen static chemisorption was applied to catalysts that had been earlier reduced at 673 K, without previous oxidation, while being cooled down at helium atmosphere. The purpose of this was to prevent volatile oxides formation and noble metals from agglomerating. A H/ Ru adsorption stoichiometry equal to 1 was adopted. However, it is advisable to take drawbacks regarding adsorption on bimetallic catalysts into consideration, because of a possible interaction between Ru and Cu. A hydrogen adsorption on copper has not been taken into account, since it has already been reported that Cu/SiO₂ catalysts show a metallic area smaller than 1 m²/g (15) and, thus, hydrogen adsorption on such metal turns out to be quite low. On the other hand, and as it has been stated by Sinfelt (18), there is little surface ruthenium covering in the case of CuRu/SiO₂ catalysts, at low Cu content. At both metals clusters, copper would be placed on surface, which is due to such metals have little miscibility at a bulk state. Because of copper "inactive" characteristics when adsorbing hydrogen, this result becomes a good approach of the composition of surface metallic ruthenium.

The number of atoms of surface ruthenium is smaller compared to the amount of added ruthenium. This may suggest ruthenium build-up and, because of the preparation method that has been used (19), part of this ruthenium might be occluded in the support. The reduction treatment performed at low temperature created the metal particles, but just a part of them moved to the surface. Therefore, there is not an easy way to get ruthenium moved out from the support. Table 2 shows differences with respect to ruthenium dispersion (32 to 11%), which was only achieved under certain conditions, such as having one third of the added noble metal surface exposed for hydrogen adsorption. In the case of Ru (1%) / SiO₂ catalysts, dispersion is slight and decreases as the gelation pH increases. This variation in dispersion may be related to a different kind of interaction between both metal and support precursors, because of the media different pH. The increase in the amount of added water does not critically affect ruthenium dispersion. Regarding Cu(1%)Ru(1%)/SiO₂ catalysts, ruthenium dispersion increases with the gelation pH, while no influence has been found for the H₂O/TEOS ratio. For the case of higher pH (above 2), silica features a negative charged, thus showing a stronger interaction among precursors. Likewise, we must also consider that it is feasible to have bimetallic catalysts undergone some ruthenium chemical modification due to copper presence. There might be some copper close to ruthenium, but some alloy formation is less likely to occur, since both metals show a little miscibility (20).

Table 2. Surface ruthenium atoms, atomic ratio H/Ru and metal particle size by H₂ chemisorption and transmission electronic microcopy for prepared Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ catalysts.

Catalysts	Superficial Ru atoms	H/Ru ratio	dRu, quim. (nm)	dRu, TEM (nm)
Ru(1%)/SiO2 (1,10)	1.91 x 1019	0.32	3.5	4.6
Ru(1%)/SiO2 (5,10)	1.36 x 1019	0.23	4.9	5.4
Ru(1%)/SiO2 (11,10)	0.65 x 1019	0.11	10.3	10.7
Ru(1%)/SiO2 (1,15)	1.07 x 1019	0.18	6.2	7.2
Ru(1%)/SiO2 (5,15)	1.13 x 1019	0.19	5.8	5.1
Ru(1%)/SiO2 (11,15)	0.67 x 1019	0.11	10.0	11.2
Cu(1%)Ru(1%)/SiO2 (1,10)	0.64 x 1019	0.11	10.4	-
Cu(1%)Ru(1%)/SiO2 (5,10)	1.59 x 1019	0.27	4.2	-
Cu(1%)Ru(1%)/SiO2 (11,10)	1.93 x 1019	0.32	3.4	-
Cu(1%)Ru(1%)/SiO2 (1,15)	0.79 x 1019	0.13	8.4	-
Cu(1%)Ru(1%)/SiO2 (5,15)	1.36 x 1019	0.23	4.9	-
Cu(1%)Ru(1%)/SiO2 (11,15)	1.40 x 1019	0.23	4.7	-

The size of the metallic ruthenium particle was assessed through the d_Ru=5/S·r, where S is the metal surface area, and r is the metal specific density. In spite of the fact that the non-immersed metallic particle has been detected by means of chemisorption, while the whole metallic particle can be detected through electronic microscopy, this study found a slight increase in particle size by means of the transmission electronic microscopy. The achieved metallic ruthenium sizes turn out to be small: they range from 3.4 to 11.2 nm. These are expected outcomes, because of the procedure that had been used for catalysts preparation (9). Copper presence slightly decreases the size of metallic ruthenium.

For the purpose of catalytic study, the toluene hydrogenation reaction was used, the suggested reaction mechanism (21) proceeds according to the scheme following: Toluene Æ Methylcyclohexene Æ Methylcyclohexane. Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ catalysts have active sites that are suitable for total hydrogenation, since methylcyclohexane the only product is detected . Reaction carried out to 673 K and atmospheric pressure. A 0.15 g catalyst that had been reduced in situ was used. The results of catalytic activity measurement showed that all catalysts undergo an initial deactivation, which may possible be caused by a non-specified surface deposit. Copper presence slightly decreases initial deactivation. Therefore, it is likely that copper take part in preserving the active site electrons deficient character, or else in keeping the site active structure.

The toluene conversion that has been reported in Table 3 states that Ru(1%)/SiO₂ catalysts are able to convert from 15% to 33% of the reactive. They contain a larger amount of surface exposed ruthenium. Small particles that are supported in oxides become more and more affected the metal-support interaction. Later, some alteration in the electronic properties should be expected and, thus some variations in the metal catalytic features. No effect of the H₂O/TEOS ratio has been observed regarding toluene conversion, except that the increase in pH gelation slightly decreases such conversion. In the case of Ru(1%)/SiO₂ catalysts, the turnover number (TON) has been kept fairly constant at a value close to 0.6 (sec^-1). Aromatic hydrogenation is considered as an insensible reaction with respect to the structure, while the TON value should not change with the size of metal crystal. The researched catalysts show metallic ruthenium sizes larger than 3 nm, which is why it is likely to apply the insensitivity criterion to the structure (21). Catalysts obtained at higher pH and at a higher H₂O/TEOS ratio produce a greater amount of methylcyclohexane. Such outcome could be related to the increase of the size of metallic ruthenium, where the hydrogen adsorption capacity would also increase, thus making the direct hydrogenation of the reactive easier. The effect of the larger specific surface should not be put aside with respect to this outcome, since there is a higher mesoporous and macroporous contribution. This status may improve the reactive fixation on the active sites, thus generating a toluene stronger interaction with the metallic surface. As a result, the retained time on it would increase, which would, in turn, bring about a greater hydrogenation of the aromatic ring.

Table 3. Conversion of toluene (X), Methylcyclohexane moles (n_MCHa) and turnover number (TON) of prepared catalysts to 573 K and 1 atm.

Catalysts	X (%)	nMCHa (moles)	TON (seg-1)
Ru(1%)/SiO2 (1,10)	29	8.30 x 10-8	0.53
Ru(1%)/SiO2 (5,10)	33	9.92 x 10-8	0.64
Ru(1%)/SiO2 (11,10)	21	11.29 x 10-8	0.68
Ru(1%)/SiO2 (1,15)	31	18.41 x 10-8	0.58
Ru(1%)/SiO2 (5,15)	27	15.27 x 10-8	0.57
Ru(1%)/SiO2 (11,15)	15	22.56 x 10-8	0.61
Cu(1%)Ru(1%)/SiO2 (1,10)	6	0.44 x 10-8	0.23
Cu(1%)Ru(1%)/SiO2 (5,10)	8	0.73 x 10-8	0.17
Cu(1%)Ru(1%)/SiO2 (11,10)	10	1.73 x 10-8	0.20
Cu(1%)Ru(1%)/SiO2 (1,15)	4	0.54 x 10-8	0.18
Cu(1%)Ru(1%)/SiO2 (5,15)	5	0.79 x 10-8	0.17
Cu(1%)Ru(1%)/SiO2 (11,15)	10	1.86 x 10-8	0.20

Out of the decrease in toluene conversion, above 10%, in Cu(1%)Ru(1%)/SiO₂ catalysts, with respect to Ru(1%)/SiO₂ catalysts, we may conclude that there is a meaningful copper influence on ruthenium. The small decrease in the amount of the strongly chemisorbed hydrogen on Cu(1%)Ru(1%)/SiO₂ is likely a result of such situation. Since the size of the metallic ruthenium does nor undergo huge changes due to copper addition, we agree with what had been stated by Luyten et. al. (22), who pointed out that there is little ruthenium surface enrichment at low copper contents. Just a small part of the possible Cu-Ru clusters have copper built-in on the their surfaces. Most likely, copper would act as a separator for ruthenium crystals. Such conclusion matches with both metals low miscibility in the bulk. So, the lower capacity of these catalysts, it can be observed that the methylcyclohexane moles show a meaningful decrease, is possibly due to a change in the ensembles size, which would, in turn, prevent hydrogenation from taking place. XPS-based studies on CuRu/Silica catalysts (23) do not show evidence of electronic transfer from Cu to Ru or vice versa, so no important electronic effect regarding catalytic behavior of the Cu-Ru pair is expected. On the other hand, as no formation methylcyclohexene was detected on these catalyst; so, can not assume on them some preferential mechanism regarding a consecutive mechanism in the toluene hydrogenation reaction. Same as what had been obtained with the Ru(1%)/SiO₂ catalysts, TON value in Cu(1%)Ru(1%)/SiO₂ catalysts, which is close to 0.2 (sec-¹) keeps constant. Therefore, variation in the metallic ruthenium diameter, again, will not alter such value.

CONCLUSIONS

Ru(1%)/SiO₂ and Cu(1%)Ru(1%)/SiO₂ gels have been obtained with different texture, surface and catalytic properties by changing the pH, as well as the H₂O/TEOS ratio used in gelation.

Gels show a higher surface area at a lower preparation pH, in spite of the fact that there is an increase in the mesoporous and macroporous fraction. The presence of Cu²⁺ in the reaction media increases Cu(1%)Ru(1%)/SiO₂ catalysts microporosity.

Two reduction stages have been observed, while the amount of water and pH used in their preparation do not appear be an important factor. The shapes and sizes of the reduction profiles are depends on gelation pH and of H₂O/TEOS ratio. Ru(1%)/SiO₂ catalysts feature a greater ruthenium fraction strongly linked to the support when compared to Cu(1%)Ru(1%)/SiO₂ catalysts.

The amount of exposed metallic ruthenium to the surface may change regarding both researched preparation variables. Such amount is not affected by copper addition. The achieved metal dispersions are acceptable for the studied reaction. There is an initial deactivation in Ru (1%)/SiO₂ catalysts. They present a higher toluene conversion and produce a greater amount of methylcyclohexane. The existing copper decreases this initial deactivation, while both conversion and the active centers for the total hydrogenation also decrease. There is a clear effect on the active sites because of copper presence.

ACKNOWLEDGMENTS

The authors would like thank to Dirección de Investigación, Universidad del Bío-Bío, for the financial support to this research.

REFERENCES

1. López, T., Gómez, R., Novaro, O., Ramirtez-Solís, A., Sánchez- Mora, E., Cartillo, S., Poulain, E. y Martinez-Madagán, J.M., J. Catal., 141, 114-123 (1993).         [ Links ]

2. Kotanigawa, T., Actas XV Simposio Iberoamericano de Catálisis, Córdova (Argentina), 1996.         [ Links ]

3. Viniegra, M., Gómez, R. y Gonzalez, R.D., J. Catal., 111, 429 (1988).         [ Links ]

4. Zou, W. y González, R.D., J. Catal., 133, 202 (1992).         [ Links ]

5. Sprock, M., Wu, X. y King, T.S., J. Catal., 138, 617 (1992).         [ Links ]

6. Coq, B., Dutertre, R., Figueras, F. y Tazi, T., J. Catal., 122, 438 (1990).         [ Links ]

7. Corma, A., Melo, F.V., Mendioroz, S. y Fierro, J.L.G. (Editores), in "Studies in Surface Science and Catalysis" 130, 2000, Elsevier Science B.V.         [ Links ]

8. Lambert, C.K. y Gonzalez, R.D., Microporous Materials, 12, 179 (1997).         [ Links ]

9. Psaro, R., y Recchie, S., Catal. Today, 41, 130 (1998).         [ Links ]

10. Niwa, S., Mizukami, F., Isoyama, S., Tsuchiya, T., Shimizu, K., Imai, S. y Imamura, J., J. Chem. Tech. Biotechnol., 36, 236 (1986).         [ Links ]

11. Gonzalez, R.D. y Miura, H., Catal. Rev.-Sci. Eng., 36(1), 145 (1994).         [ Links ]

12. Pecchi, G., Morales, M. y Reyes, P., React. Kinet. Catal. Lett., 61, 237 (1997).         [ Links ]

13. Orellana, F., Plaza de los Reyes, J. y Urizar, S., Bol. Soc. Chil. Quím., 45, 363 (2000).         [ Links ]

14. Marchi, A.J., Fierro, J.L.G., Santamaría, J. y Monzon, A., Appl. Catal. A:General, 142, 375 (1996).         [ Links ]

15. Chambers, A., Jackson, S.D., Stirling, D. y Webb, G., J. Catal., 168, 301 (1997).         [ Links ]

16. Monti, D. y Baiker, A., J. Catal., 83, 329 (1983).         [ Links ]

17. Barbier, J., Dumas, J.M., Geron, C. y Hedrane, H., Applied Catal., 67, L5 (1990).         [ Links ]

18. Sinfelt, J.H., Acc. Chem. Res., 10, 15 (1977).         [ Links ]

19. López, T., Herrera, L., Gómez, R., Zou, W., Robinson, K. y González, R.D., J. Catal., 136, 621 (1992).         [ Links ]

20. Miura, H., Taguchi, H., Sugiyama, K., Matsuda, T. y González, R.D., J Catal., 124, 194 (1990).         [ Links ]

21. Kluson, P., Had, J., Belohav, Z. y Cerveny, L., Applied Catal. A: General, 149, 331 (1997).         [ Links ]

22. Luyten, J.L.M., Eck, M.v., Grandelle J.v. y Hooff, J.H.C.v., J. Phys.Chem., 82(18), 2000 (1978).         [ Links ]

23. Schoenmaker-Stolk, M.C., Verwijs, J.W. y Scholten, J.J.F., Applied Catal., 30, 339 (1987).         [ Links ]