Influence of hydrothermal treatment on the acid and redox functions of a Ga/HZSM5 catalyst

Influence of hydrothermal treatment on the acid and redox functions of a Ga/HZSM5 catalyst

M.G.F. Rodrigues¹, P. Magnoux² and M. Guisnet²

¹Universidade Federal da Paraíba, Centro de Ciências e Tecnologia, Departamento de Engenharia Química,

Programa de Pós-Graduação em Engenharia Química, Avenida Aprígio Veloso, 882, Bodocongó - 58.109-970,

Phone: [phone], Fax: [phone], Campina Grande - Paraíba, Brazil.

E-mail: [e-mail]

²Université de Poitiers, Laboratoire de Catalyse en Chimie Organique, UMR CNRS 6503, 40, Avenue du

Recteur Pineau 86022, Phone: [phone], Fax: [phone], Poitiers Cedex, France.

(Received April 18, 1998; Accepted: August 4, 1999)

Abstract - After steaming at 530° C for 30 minutes and for 6 hours, a Ga/HZSM5 catalyst was characterized by three model reactions (meta-xylene isomerization, propane aromatization and methylcyclohexane transformation). The activity and selectivity of this catalyst were compared to those of a fresh Ga/HZSM5 catalyst. It was demonstrated that hydrothermal treatment provokes a significant decrease in the dehydrogenation activity of the gallium species and a small decrease in the protonic acidity of the catalyst.

Keywords: Ga/HZSM5 catalyst, propane, deactivation by steaming.

INTRODUCTION

Aromatization of short-chain alkane has recently been reviewed (Ono, 1992; Guisnet et al., 1992; Guisnet and Gnep, 1996). Aromatization occurs mainly by a bifunctional scheme on Ga/HZSM5 catalyst with Ga species catalyzing the dehydrogenation of the alkane reactants and of the naphtene intermediates and the acid sites catalyzing the oligomerization of alkenes and the cyclization of oligomers. However, their activity and selectivity for aromatics are obtained at obtained at high temperatures (500 to 600° C) imposed by thermodynamics (Derouane et al., 1994) under conditions where deactivation may readily occur. Deactivation of zeolite catalysts originates from coking which causes pore blockage and active site coverage or poisoning (Beekman and Froment, 1979; Beekman and Froment, 1980; Froment, 1989; Rodrigues et al., 1996) and from the migration and sintering of the Ga species in the Ga/HZSM5 (Kanazinov et al., 1996; Dooley et al., 1992; Hamid et al., 1994).

Structural alterations in the Ga/HZSM5 catalyst (dealumination, degalliation) may result from exposure to steam produced by burning coke during oxidation regeneration of coked catalysts (Hamid et al., 1994). These structural alterations of the Ga/HZSM5 catalysts can affect the dehydrogenation activity, or the acid activity or both. Consequently, it is very important to specify the effect of the steam on the acid and dehydrogenation activities. Three model reactions, meta-xylene isomerization, propane aromatization and methycyclohexane transformation were used to specify the effect of a hydrothermal treatment on the dehydrogenation and acid activities of Ga/HZSM5 samples.

EXPERIMENTAL METHODS

The Ga/HZSM5 catalyst (2wt% Ga) was prepared by impregnation of a HZSM5 zeolite (PQ Zeolites, Si/Al = 40) with Ga(NO₃)₃. Prior to use, the Ga/HZSM5 catalyst was treated in a dry hydrogen flow for 12 hours at 600° C and then under water at 530° C (1 ml.h^-1, P(H₂O) = 0.9 bar ) for 30 minutes or for 6 hours. These treated catalysts (200 mg for each treatment) were divided in three lots and characterized by model reactions as shown in the table 1. The model reactions were carried out in a flow reactor under the following conditions:

The reaction products were analyzed on-line by gas chromatography with a 30 m fused silica DB Wax capillary column for meta-xylene isomerization and with a 50 m fused silica Plot Al₂O₃/KOH capillary column for the others reactions.

RESULTS AND DISCUSSION

Three model reactions were used to characterize the Ga/HZSM5 catalyst before and after hydrothermal treatment: meta-xylene isomerization at 350° C, propane aromatization at 530° C and methylcyclohexane transformation at 500° C. Table 2 shows that the initial rates of reactant transformation were always lower when the catalysts were treated under water. The activities also decreased with time for treatment under water.

Hydrothermal treatment affects the acid sites. This is shown clearly by the decrease in the rate of meta-xylene isomerization (1.1 and 2.5 times lower at treatment times of 30 minutes and 6 hours, respectively). It is well know that this reaction occurs on protonic sites by the following mechanism (Guisnet, 1985; Poustma, 1976), and it is very unlikely that the reaction rate is affected by redox sites.

Para- and ortho-xylenes are practically the only products of meta-xylene transformation. The para/ortho ratio does not depend on the treatment (under H₂ or H₂ + H₂O). In all cases, it was close to 1.5 independent of conversion level. The decrease in meta-xylene activity is certainly due to the dealumination of the zeolite using hydrothermal treatment. However, the aluminium extra framework species formed during this dealumination doesn’t affect xylene selectivity.

The initial rate of propane transformation decreases during the hydrothemal treatment (Table 2). Initial activity becomes 3 and 6 times lower with 30 minutes and 6 hours of treatment under water, respectively. In propane aromatization, propene resulting from propane dehydrogenation (D) on the gallium sites and methane, ethane and ethylene resulting from cracking (C) on the acid sites are the only primary products. A higher value of C/D are obtained after hydrothermal treatment (Figure 1). This suggests that the dehydrogenative sites (Ga species) are more affected by the treatment than the acid sites.

Initial activity of methylcyclohexane transformation also decreases during hydrothermal treatment (Table 2). Time of treatment also affects this activity (1.25 and 2.9 times lower at 30 minutes and 6 hours, respectively). Methylcyclohexane is directly transformed into C₁-C₆ cracking products on acid sites (A_C) and C₆-C₈ aromatics on gallium sites (A_D).

Figure 2 shows the evolution of the dehydrogenation and cracking activities as a function of time under water treatment at 530^oC. After a treatment of 30 minutes, the cracking and dehydrogenation activities decrease similarly. However, at a treatment time of 6 hours, the loss of the dehydrogenation activity is much more important than those of the cracking activity (85% compared to 50%). Then, the dehydrogenation function is more affected than the acid function by the hydrothermal treatment.

The dehydrogenation cracking rate ratio (A_D/A_C) is equal to 2 after hydrogen treatment and after 30 minutes of water treatment as compared to 4 to 6 after a hydrothermal treatment of 6 hours (Figure 3).

As in the case of propane aromatization, hydrothermal treatment affects the dehydrogenation sites more than affects the acid sites. However, in the Ga/HZSM5 sample treated during 6 hours under water, the A_D/A_C ratio is lower than it is on pure protonic zeolite HZSM5 (Figure 3). This suggests that, due to the presence of gallium species, the dehydrogenation activity is not totally suppressed after this treatment.

The decrease in dehydrogenation activity can be due the degalliation of the zeolite framework (Hamid et al., 1994), but also to the presence of extra framework aluminium species formed during the dealumination of the zeolite which allows access of the reactants to the gallium species.

CONCLUSION

Hydrothemal treatment of a Ga/HZSM5 catalyst at high temperatures causes a decrease in the activity of the acid sites and a more significant decrease in the dehydrogenation activity of the gallium species. Steaming can lead to degalliation and to dealumination of the zeolite with the formation of extra framework species responsible for a partial blockage of access to the active sites (gallium and protonic sites).

REFERENCES

Ono Y., Catalysis Review – Science Engineering, 34(3), 179 (1992).

Guisnet, M., Gnep, N.S. and Alario F., Appl. Catal. A, 89, 1 (1992).

Guisnet, M. and Gnep, N.S., Catal. Today, 31, 275 (1996).

Derouane, E.G., Hamid, S.B.A., Ivanova, I.I., Blom, N. and Nielsen, P.E.H., J. Molecular Catalysis, 86, 371 (1994).

Beekman, J.W. and Froment, G.F., Ind. Eng. Chem. Fund., 18, 245 (1979).

Beekman, J.W. and Froment, G.F., Chem. Eng. Sci., 35, 805 (1980).

Froment, G.F., Studies in Surface Science and Catalysis, 6, 1 (1989).

Rodrigues, M.G.F, Barré, M., Magnoux, P., Choudary, V.R and Guisnet, M., J. Chim. Phys., 93, 337 (1996).

Kanazinov, V., Price, G. L. and Tynliev, G., Zeolite, 12,846 (1996).

Dooley, K.M., Chang, C. and Price, G.L., Appl. Catal., 84, 17 (1992).

Hamid, S.B.A., Derouane, E.G., Meriaudeau, P. And Naccache C., Studies in Surface Science and Catalysis, 88, 183 (1994).

Hamid, S.B.A., Derouane, E.G., Demortier, G., Riga J. and Yarne, M.A., Appl. Catal., A: General, 108, 85 (1994).

Guisnet, M. in: Catalysis by Acids and Bases, Studies in Surface Science and Catalysis, 20, 283 (1985).

Poutsma, M.L. in: Zeolite Chemistry and Catalysis, A. Rabo (Ed.), Am. Chem. Soc. Monography, Washington, 171, 437, (1976).

Publication Dates

• Publication in this collection
  24 Apr 2000
• Date of issue
  Mar 2000

History

• Received
  18 Apr 1998
• Accepted
  04 Aug 1999