On the acidity and/or basicity of USY zeolites after basic and acid treatment

On the acidity and/or basicity of usy zeolites after basic and acid treatment

V. Calsavara¹, N. R. C. F. Machado¹, J. L. Bernardi Jr¹ and E. F. Sousa-Aguiar^2-3

¹Departamento de Engenharia Química, Universidade Estadual de Maringá, Av. Colombo, 5790,

CEP 87020-900, Fone [phone], Fax: [phone], Maringá - PR, Brazil.

E-mail: [e-mail] and [e-mail]

²CENPES/PETROBRAS, DICAT, Rio de Janeiro - RJ, Brasil.

³Escola de Química, UFRJ, DPO, Rio de Janeiro - RJ, Brasil.

(Received: November 8, 1998; Accepted: September 22, 1999)

Abstract - The isopropanol decomposition reaction was used to evaluate the catalytic activity of ultrastable (USY) zeolites with different degrees of dealumination, treated in strongly alkaline medium at various temperatures and contact times. This treatment resulted in the reinsertion of non-framework aluminium, a result of the ultrastabilization process. The samples obtained were also submitted to an acid treatment, leaching the non-framework aluminium that had not been reinserted. The results obtained at 723K showed a large reduction in the acidic activity of the alkaline-treated zeolite, as the treatment conditions became more severe (the longer the treatment time or the higher the temperature, the higher the degree of dealumination). On the other hand, treated samples displayed some isopropanol dehydrogenation activity (basic sites). However, this activity was not very significant and did not depend on the alkaline treatment or ultrastabilization conditions used. The effect of reaction temperature and acid leaching on activity is also shown.

Keywords: ultrastable zeolites, alkaline treatment, reinsertion, acid leaching, acidity, basicity.

INTRODUCTION

Zeolites are crystalline aluminosilicates with SiO₄ and AlO₄ tetrahedra as their primary structural units. The connection of several tetrahedra generates the secondary building units (SBU) of the zeolites, like the four-member or six-member rings (Szostak, 1989). Combinations of SBU produce different polyhedra. In the case of the Y zeolite, eight truncated octahedra (or b cavities), connected by hexagonal prisms, form the unit cell. Each b cavity has 1.6 Å and 2.8 Å openings (too small for the majority of organic molecules). However, the arrangement of the eight octahedra creates a supercavity (or a cavity) with a 12-ring opening of 7.4 Å (Meier et al., 1996). This configuration of the Y zeolite is comprised of 192 tetrahedra, with the following typical composition (Breck, 1984):

Na₅₆[(AlO₂)₅₆(SiO₂)₁₃₆].250H₂O.

The silica-to-alumina ratio of Y zeolites can vary from 3 to about 6. Higher ratios are obtained by means of different dealumination processes reviewed by Scherzer (1984). The process that consists of hydrothermal treatment after ammonia exchange, proposed by McDaniel and Maher (1976) and called ultrastabilization, was responsible for the wide use of Y zeolites in oil hydrocracking. The ultrastable Y zeolites (USY) show great acid strength for catalytic cracking processes by virtue of the structural changes undergone during thermal treatment. The dealumination occurring in the ultrastabilization process leads to the formation of non-framework aluminium species and structural defects such as hydroxyl nests, as well as secondary mesopores from the partial destruction of the crystalline structure (Scherzer, 1984).

The reinsertion of the non-framework aluminium in the framework was first reported by Breck and Skeels (1980) and has usually been studied by placing ultrastable Y zeolites in contact with highly alkaline solutions (Man and Klinowski, 1988; Sousa-Aguiar et al., 1989; Barrie et al., 1991). In the proposed mechanism (Liu et al., 1986), the dissolution of the extra-framework species in the strongly basic medium supplies the tetrahedral aluminate anions that then enter the framework at the hydroxyl nests or directly replace the framework silicon. The method leads to novel zeolites, which have the same structure and silica-to-alumina ratios similar to those of the parent Y zeolite, but different silicon and aluminium distributions within the framework and thus modified thermal stability and catalytic activity. The reinsertion process was verified in previous work by Calsavara et al. (1996), using several characterization techniques. Two ultrastable Y zeolites with different degrees of dealumination were treated with NaOH solution at various temperatures and contact times. Filtrate analysis showed the transference of the solubilized non-framework aluminium species to the solid phase at a rate that increased as both temperature and dealumination degree increased. Reinsertion of this aluminium in the lattice was verified by the reduction in the structural silica-to-alumina ratios determined by ²⁹Si NMR and FTIR. The reduction in mesoporosity pointed to a aluminium reinsertion in the structural defects. The experimentapoints were well fitted to a crystallization model that made it possible to estimate the activation energy of the reinsertion process.

The strongly modified acidity of the zeolite after the basic treatment permits the supposition of basicity in the resulting material. It is then interesting to verify the catalytic activity of this zeolite by using a model reaction like isopropanol decomposition, which occurs on acid sites with dehydration, as well as on basic sites with dehydrogenation. This reaction has been widely used in the study of zeolite reactivity, with a special focus on basic sites (Yashima et al., 1974; Hathaway and Davis, 1989; Bezoukhanova and Kalvachev, 1994).

Sobrinho et al. (1993) and Sousa-Aguiar et al. (1989) studied the combination of the ultrastabilization of Y zeolites with the acid leaching of the non-framework aluminium species which arose during the ultrastabilization treatment. The main characteristic of this procedure is the emptying of the pores that are partially blocked by those species.

During the process of alumina reinsertion in ultrastable zeolites by basic treatment, it was previously verified (Calsavara et al., 1996) that there are still non-framework aluminium species in the material, even after prolonged basic treatment. Acid leaching could remove these aluminium species that had not been reinserted, eliminating pore blockage and thus enabling easier access to the zeolitic pore structure.

EXPERIMENTAL

The samples utilized in this study were obtained according to procedures previously described (Calsavara et al., 1996). The original NaY was ultrastabilized at 823K (USY1) and 923K (USY2). Treatment with NaOH solution was done at 323K and 353K with contact times of 1 and 24 hours. The main characteristics of the samples are listed in Table 1.

The USY1 samples treated for 1 and 24 hours at 353K also underwent sulphuric acid leaching, according to the procedure of Sobrinho et al. (1993), but with a pH of 3.0 just after the addition of acid.

Catalytic tests were carried out in a 4.5 mm ID stainless steel reactor, placed inside a cylindrical furnace, using a catalyst sample of 0.30g (dry basis). Nitrogen was used as the carrier gas, fed at 30 mL/min during activation as well as in the reaction. In the activation step, temperature was gradually raised up to 873K and then kept there for 90 minutes. At that point the system was cooled to the temperature of reaction.

Isopropanol was fed by a gear pump. The mass flow rate was 2.87 g/h, unless specified otherwise. The gas flow rate was monitored by means of a bubble flowmeter. As soon as the nitrogen flow rate could be kept constant independently of the pressure drop in the catalytic bed, the proportion of gaseous products could be determined.

Samples of the liquid product were taken regularly by means of a condensation trap and analysed by a gas chromatograph which employed a 10% carbowax-20M/chromo W-HP column (6ft x 1/8 in) at 373K and a thermal conductivity detector. The results shown correspond to samples collected after about 90 minutes of reaction, except in Figure 3 where a time variation is presented.

RESULTS AND DISCUSSION

In Table 1 one can observe the reduction in the area of mesopores after the alkaline treatment. However, there are still structural defects, represented by the residual mesoporosity. The basic treatment nearly recovered the sodium content of the NaY, indicating a recomposition of the zeolitic structure. The aluminium reinsertion in the structure is confirmed by the reduction in the structural silica-to-alumina ratio. However, as this ratio has not reached the values of the original NaY, there is still non-framework aluminium, which may be removed by acid leaching.

The results obtained for some samples (after basic treatment) in the isopropanol decomposition at 723K and a space time of 6.27 h . g_catalyst /mol isopropanol (IPA) are summarized in Table 2.

One can observe an almost complete dehydration to propene without significant acetone generation in the case of both starting zeolites (USY1 and USY2), showing high levels of activity and stability for acid reactions. In the alkaline-treated samples, however, the existence of some basic sites is revealed by the presence of acetone, in spite of the fact that acid activity was still more significant than basic activity. This indicates that the aluminium reinsertion by basic treatment did not generate a large number of basic sites, but significant alterations in the USY after reinsertion did occur, depending on treatment conditions.

Considering the samples that had been treated in alkaline medium for one hour, one can observe a large reduction in the acid activity of the more dealuminated zeolite (USY2) treated at 353K, compared to the same zeolite treated at 323K as well as to the USY1 (less dealuminated) treated at 353K. An equivalent reduction in acid activity occurred by extending the treatment of zeolite USY1 at 353K from 1 to 24 hours. The same was not observed for the USY2 zeolite treated at 323K. These observations suggest that the acid activity of the USY zeolites remains after alkaline treatment if it is not severe. Further, the loss of acid activity is more pronounced when the starting zeolite has undergone a higher temperature ultrastabilization (deeper dealumination). This loss of activity can be related to the aluminium reinsertion in the framework during alkaline treatment, as discussed elsewhere (Calsavara et al., 1996), since the behaviour observed here matches the higher aluminization rate for the most dealuminated zeolite and for the higher treatment temperature. Further, the reduction in acidity is also related to a probable ion exchange during basic treatment, with the substitution of sodium ions for hydrogen ions in the zeolitic structure.

To verify the possible influence of reaction temperature on acid strength and distribution of products, in Table 3 and Figure 1 the influence of the isopropanol decomposition reaction temperature on product formation for the USY1 zeolite, treated at 353K for one hour with a reaction space time of 6.27 h . g_catalyst / mol isopropanol (IPA), is shown.

It is evident from Figure 1 that acid activity for isopropanol decomposition increases as temperature increases, until almost completely converted at 723K, but with no modification of the distribution of products.

A study which varied the mass of catalyst in the reactor while keeping constant the flow rate of the reagent (by varying contact time) and the temperature (673K) was also performed. The results, for the USY1 zeolite treated at 353K for one hour, are shown in Table 4 as well as in Figure 2. One may observe a direct relation between conversion and space time.\

The effect of acid leaching of the USY1 samples treated in basic medium for 1 and 24 hours at 353K on the isopropanol decomposition reaction is summarized in Table 5. Two different isopropanol feed rates were used: 2.87 g/h and 5.26 g/h, corresponding to space times of 6.27 and 3.43 h . g_catalyst / mol isopropanol, respectively. These results showed that acid leaching was important in recovering the activity of the zeolite that had previously undergone a more severe basic treatment (353K/24h). H⁺ ions were exchanged for Na⁺ ions in the zeolitic structure, thereby recovering acidity. The Na₂O content in this sample dropped from 11.9 wt% to 7.4 wt%. In the sample treated for only 1h at 353K, which had a Na₂O content of 8.7 wt% after acid leaching, the activity level continued high.

In Table 5, as in the other tables, the data correspond to product samples taken after about 90 minutes of reaction. However, when the zeolites submitted to acid leaching were tested, the isopropanol conversion showed a large variation over time (unlike the others). This variation is displayed in Figure 3.

Comparing the data in Figure 3 to those for corresponding samples that had not undergone acid leaching (in Table 2), one may observe a considerable deactivation over time of the zeolite treated in alkaline medium for one hour and subsequently acid leached. This deactivation starts earlier when the space time is reduced. This effect was not observed with the zeolite that was treated in alkaline medium for 24 hours, which showed a very high acid activity, attaining almost complete conversion even at the lowest space time.

This increase in activity could have been caused by the acid leaching of the aluminium species that blocked the zeolite micropores to a greater extent in the USY1 zeolite, which was treated in alkaline medium for a longer period of time and so showed a greater reduction in mesopores (Table 1) that generated micropores. With the reduction in pore diameter there was blockage by non-reinserted aluminium. When this aluminium was removed by acid leaching, the pores were emptied. The withdrawal of aluminium is confirmed by the slight increase in the overall silica-to-alumina ratio: from 5.6 to 6.2 for the USY1 treated for 1h at 353K and from 5.3 to 5.5 for the sample treated for 24h.

In an overall view of Tables 2 to 5, one observes that the zeolites obtained by alkaline treatment of USY do not show considerable activity for basic catalysis under the conditions studied, since a significant isopropanol dehydrogenation was not achieved. The conversion to acetone was always around 1%, with some variations that led to an increase in the acetone fraction in products (and thus in dehydrogenation selectivity) just in those tests where there was a smaller overall isopropanol conversion. Regarding the di-isopropyl ether formation by intermolecular dehydration, it appeared only in some cases, and the fraction obtained was always much smaller than the water fraction. The ether fraction increased under milder conditions with a lower isopropanol conversion. This allows us to conclude that the conditions used in this study are effective enough for more complete dehydration at an intramolecular level.

CONCLUSIONS

The alkaline treatment of USY zeolites gives rise to some basic sites, as indicated by the appearance of acetone in the isopropanol decomposition reaction. However, the activity of these sites is not significant, since conversion to acetone was always around 1%. Furthermore, this activity does not seem to depend on ultrastabilization and alkaline treatment conditions.

The residual acid activity of the USY zeolites after alkaline treatment is inversely proportional to the severity of this treatment or of the ultrastabilization process, which can be related to the aluminium reinsertion mechanism.

The new zeolites resulting from the mild alkaline treatment show good stability in their acid activity.

Acid leaching led to an increase in activity for the sample that had undergone a longer alkaline treatment and a regeneration of the acid sites by the exchange of H⁺ ions for Na⁺ ions in the zeolitic structure.

The aluminium reinsertion process increases the sodium content to the levels of the parent NaY zeolite. However, the resulting zeolite still has good acidity, and taking into account the differences between the structural and overall silica-to-alumina ratios, sodium must be incorporated in the non-framework aluminium that has not been reinserted or leached.

REFERENCES

Barrie, P.J., Gladden, L.F. and Klinowski, J., Neutron diffraction studies of realuminated zeolite Y. J. Chem. Soc. Chem. Commun., p. 592 (1991).

Bezoukhanova, C.P. and Kalvachev, Y.A., Alcohol reactivity on zeolites and molecular sieves. Catal. Rev. - Sci. Eng., 36, 1, 125 (1994).

Breck, D.W., Zeolite molecular sieves: structure, chemistry and use. Robert E. Krieger, Malabar, FL, USA (1984).

Breck, D.W. and Skeels, G.W., Zeolite chemistry IV: Evidence for the elimination and subsequent reinsertion of framework aluminium during the stabilization of NH₄⁺ -exchanged zeolite Y. 5^th International Conference on Zeolites, p. 335 (1980).

Calsavara, V., Sousa-Aguiar, E.F. and Machado, N.R.C.F., Reactivity of USY extraframework alumina in alkaline medium. Zeolites, 17, 340 (1996).

Hathaway, P.E. and Davis, M.E., Base catalysis by alkali-modified zeolites. I- Catalytic activity. J. Catal., 116, 263 (1989).

Liu, X., Klinowski, J. and Thomas, J.M., Hydrothermal isomorphous insertion of aluminium into the framework of zeolite Y: A convenient method of modifying the siting of Al and Si in faujasitic catalysts. J. Chem. Soc. Chem. Commun., p. 582 (1986).

Man, P.P. and Klinowski, J., Quadrupole nutation²⁷ Al nmr studies of isomorphous substitution of aluminium in the framework of zeolite Y. Chemical Physics Letters, 147, 581 (1988).

McDaniel, C.V. and Maher, P.K., Zeolite stability and ultrastable zeolites. In: Rabo, J.A., ed. Zeolite chemistry and catalysis. ACS, Washington, D.C., USA (1976).

Meier, W.M., Olson, D.H. and Baerlocher, Ch., Atlas of zeolite structure types. Zeolites, 17, 105 (1996).

Scherzer, J., The preparation and characterization of aluminum-deficient zeolites. In: Whyte Jr., T.E. et al. eds. Catalytic Materials: relationship between structure and reactivity. ACS, Washington, D.C., USA (1984).

Sobrinho, E.V., Cardoso, D., Gusmão, J., Sousa-Aguiar, E.F. and Menezes, S.M.C., Desaluminização cíclica da zeólita Y. Anais 7^o Seminário Brasileiro de Catálise. p. 160 (1993).

Sousa-Aguiar, E.F.,Duarte, M.A.I.,Mascarenhas, Y.P. and Martins, R.L., Evidências de reinserção de alumínio na rede cristalina de zeólitas ultra-estáveis por meio de tratamento com soluções alcalinas. Anais 5° Seminário Brasileiro de Catálise, Guarujá. p. 43 (1989).

Szostak, R., Molecular Sieves: Principles of synthesis and identification. Van Nostrand Reinhold Catalysis Series, New York, NY (1989).

Yashima, T., Suzuki, H. and Hara, N., Decomposition of 2-propanol over alkali cation exchanged zeolites. J. Catal., 33, 486 (1974).

Publication Dates

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

History

• Received
  08 Nov 1998
• Accepted
  22 Sept 1998