Selective acylation of 2 methoxynaphthalene by large pore zeolites: catalyst selection through molecular modeling

https://doi.org/10.1016/S1476-9271(02)00093-2Get rights and content

Abstract

The selective acylation of 2-methoxynaphthalene (2-MON) is commercially very important to produce selectively 2-acyl-6-methoxynaphthalene (2,6-AMON), which is a precursor to Naproxen, an anti-inflammatory drug. Most of the laboratory investigations conducted with different solid acids show that the undesirable products are formed in large quantities. Thus, various molecular modeling techniques were used to investigate selectivity towards desired 2,6-AMON isomer over undesired 1,2-AMON, in four large-pore zeolites, namely, mordenite (MOR), zeolite L (LTL), zeolite beta (BEA) and ITQ-7 (ISV). The qualitative results were obtained by using simple molecular graphics (MG) and structural fitting approach. The quantitative results were obtained by incorporating the interaction of atoms of the molecules and those of the zeolite frameworks. From diffusion energy profile calculations the diffusion energy barriers for self-diffusion of 2-MON and the acylated isomers were obtained. From these energy barrier values the selectivity offered by zeolites towards desired product was determined and it was found to be in the order of ISV>BEA>MOR>LTL. Hybrid Quantum Mechanics–Molecular Mechanics (QM/MM) approach was used to study the effect of Brønsted acidity on the activity and selectivity offered by zeolites. The interaction of the reactant and product species with the acidic protons at T3 and T9 sites in BEA, having different acidities was studied by this method. The QM energy values indicate that acidity affected the catalytic activity but not the regioselectivity towards the desired 2,6-AMON isomer.

Introduction

Aromatic acylation is of paramount importance in various areas of the fine chemical industry. The Friedel–Crafts acylation of aromatics is a method of choice for the synthesis of aromatic ketones, which are important intermediates in the production of fine chemicals and pharmaceuticals (Kouwenhoven and van Bekkum, 1997). One of such important reactions is the acylation of 2-methoxynaphthalene (2-MON) to give 2-acyl-6-methoxynaphthalene (2,6-AMON) which is the key intermediate in the production of a popular anti-inflammatory drug, (S)-2-(6-methoxy 2-naphthyl) propanoic acid, commonly known as Naproxen (Walker et al., 1978, Chan, 1993).

Classical catalysis of Friedel–Crafts acylation reaction employs various Lewis acids such as AlCl3, FeCl3 or TiCl4 as catalysts (Gunnewegh et al., 1995, Cantrell, 1967). The industrial applicability of this reaction is severely hampered due to a number of problems inherent to the classical homogeneous catalysis of Friedel–Crafts acylation, namely, (i) the requirement of far more than molar amounts of the acylating agent, which will be completely destroyed during the separation work-up; (ii) the requirement of stoichiometric amount of Lewis acid catalysts, which will form stable complexes with the resulting ketones, thus making separation of products difficult; (iii) the formation of considerable amount of by-products and toxic wastes.

With modern stringent environmental restrictions, and in consonance with green chemistry principles, replacement of Lewis acids with solid heterogeneous catalysts has great industrial relevance. In fact, the above limitations can be easily overcome by means of non-conventional heterogeneous catalysis with solid acids that do not form complexes with the resulting ketones and thus, can be easily separated and reused. Various solid acid catalysts have been found effective in carrying out the acylation of 2-MON, namely, La-Ce-Y (Chiche et al., 1986), zeolite beta (Kim et al., 2000), other zeolites (Corma et al., 1989, Finiels et al., 1993, Richard et al., 1993), H-MCM-41 (Gunnewegh et al., 1998), cation exchanged K10 Montmorillonite (Choudhary et al., 1998), alumina pillared clays, sulfated zirconia (Yadav and Krishnan, 1998), etc.

The acylation of 2-MON (1), with acetic anhydride over heterogeneous catalysts, leads to the formation of two major products; a bulky, 1-acyl, 2-methoxynaphthalene (1,2-AMON) (2) molecule and a linear, 2-acyl, 6-methoxynaphthalene (2,6-AMON) (3). The former is kinetically controlled, while the latter is thermodynamically favored. The acylation occurs preferentially at position 1, which is a kinetically controlled position (Scheme 1).

So, at lower conversion level, with no geometrical constrains, the selectivity towards 1,2-AMON is higher (Yadav and Krishnan, 1998). However, when contact time increases, the conversion as well as selectivity towards 2,6-AMON isomer increases, which is attributed to the secondary reactions, which involve the protodeacylation of 1,2-AMON as well as the migration of acyl group from position 1 to the 6-position carbon of 2-MON, by a transacylation mechanism. Here, the interplay between the electronic and steric factors control the selectivity and thus, shape-selective properties of zeolites can be utilized effectively to achieve the selectivity towards linear, desired 2,6-AMON isomer. Since the molecules are bulky, large pore zeolites should be ideal, but then the pore size and acidity of the catalyst will play a critical role.

In the present work, we applied various molecular modeling techniques to screen the potential zeolite catalysts for the acylation of 2-MON and to select a suitable zeolite catalyst for the selective production of 2,6-AMON isomer. Structural fitting and molecular graphics (MG) techniques were used to study qualitatively the interrelation between the size and shape of the molecules 1 to 3 and the selectivity offered by the different zeolites due to their different pore geometries (Fig. 1). These qualitative results were made more quantitative by incorporating the interaction of atoms in the molecules and those in the zeolite frameworks. The diffusion characteristics of molecules 13 were studied in four different 12-member, large-pore zeolites, namely,

  • 1

    mordenite (MOR) with an elliptical one-dimensional channel;

  • 2

    zeolite L (LTL) with a circular channel and two-dimensional cages;

  • 3

    zeolite beta (BEA) with circular three-dimensional channel system and;

  • 4

    ITQ-7 (ISV) with circular three-dimensional channel system similar to zeolite beta, but pore size slightly smaller than it.

Force-field based energy minimization technique was used, being computationally less demanding than the molecular dynamics or Monte Carlo simulations, for the large molecules such as 2-MON and its acylated isomers. The diffusivities of the molecules and the shape-selective properties of the zeolites were investigated by analyzing the results of energy minimization calculations. In this paper, we also report an investigation of the comparative sorption of the AMON isomers in these four zeolites. The calculations indicate that the different zeolite frameworks bind different AMON isomers in differing manner and to different extents, thereby demonstrating that the type of modeling approach applied in this work can indeed be used to investigate the catalyst selectivity (Fig. 2).

Hybrid Quantum Mechanics–Molecular Mechanics (QM/MM) approach by using a cluster model under the influence of periodic environment was employed to study the interaction of various reactant and product species with the acidic protons in zeolite beta. The study includes QM/MM calculations at two crystallographically distinct ‘T’ sites, T3 and T9, having different acidities. By analyzing the results of QM/MM cluster calculations, the effect of Brønsted acidity on the activity of zeolite and the selectivity offered by it towards desired 2,6-AMON isomer was investigated. This approach was found to be useful in synthesis of MTBE from tert-butanol and methanol (Desai, 1998, Desai and Yadav, 1998) and acylation of phenol (Bhatt, 2000, Bhatt and Yadav, 2000).

Section snippets

Methods

Computational studies reported here were carried out using software programs, insight II and discover supplied by Molecular Simulations Inc., USA. The force field energy minimization calculations were done with the discover program, using Consistent Valence force field (CVFF) of Hagler et al. (1979). The interaction energy of the molecule with the zeolite framework was calculated using the force field energy expression (Hagler et al., 1979) that contained the terms corresponding to deformation

Molecular orbital (MO) calculations

The experimental results of the acylation of 2-MON and the deacylation of product ketones suggest that C-1 is the most reactive position and that 2,6-AMON is more stable than 1,2-AMON; which parallels the electrophillic substitution of naphthalenes having an electron donating substituent at the 2-position. To obtain numerical information on the relative reactivity of the carbons of 2-MON and relative stability of the product ketones, semi-empirical MO calculations (MOPAC) on 2-MON, 2,6-AMON and

Conclusion

MG method provides a preliminary screening of possible zeolite catalysts for a specific reaction before further detailed calculations or experiments are performed. This method has been applied to study the structural fitting of 2-MON and its acylated products inside various zeolites to understand their shape-selective properties. From the analysis of the results of MG studies and structural fitting studies, it is inferred that MOR, BEA and ISV are potential catalysts for the shape-selective

Acknowledgements

S.P.P. thanks University Grants Commission (UGC) for award of JRF, which enabled this work to be carried out. G.D.Y. acknowledges support from the Darbari Seth Professorship Endowment and Michigan State University for hosting him as the Johansen Crosby Visiting Professor of Chemical Engineering (2001–2002).

References (28)

  • A. Corma et al.

    Appl. Catal.

    (1989)
  • R.C. Deka et al.

    J. Mol. Graphics Modeling

    (1998)
  • A. Finiels et al.

    Stud. Surf. Sci. Catal.

    (1993)
  • E.A. Gunnewegh et al.

    Stud. Surf. Sci. Catal.

    (1995)
  • J.A. Horsley et al.

    J. Catal.

    (1994)
  • S.D. Kim et al.

    J. Mol. Catal. A: Chem.

    (2000)
  • F. Richard et al.

    Stud. Surf. Sci. Catal.

    (1993)
  • G.D. Yadav et al.

    Stud. Surf. Sci. Catal.

    (1998)
  • Bhatt, J.B., 2000. M. Chem. Engg. Thesis, University of...
  • Bhatt, J.B., Yadav, G.D., 2000. Paper presented at CHEMCON-2000, Indian Institute of Chemical Engineers, Calcutta,...
  • A. Berreghis et al.

    Catal. Lett.

    (2000)
  • T.S. Cantrell

    J. Org. Chem.

    (1967)
  • Chan, A.S.C., 1993. Method for preparing alpha-arylpropanoic acids. US Patent...
  • A. Chatterjee et al.

    J. Chem. Soc. Farady Trans.

    (1995)
  • Cited by (0)

    View full text