Synthesis and Application of Polymer-Supported Chiral Brønsted Acid Organocatalysts

Abstract

A new methodology for the immobilization of chiral Brønsted acids has been developed. The resulting heterogeneous organocatalysts have been employed in multiple consecutive catalysis cycles in the asymmetric organocatalytic transfer hydrogenation. The new catalyst system can not only be easily recovered from the reaction mixture in a tea-bag approach but it can also be easily reused in several catalytic cycles without loss of reactivity and selectivity.

Polymer-supported catalysts have attracted considerable attention as they are well known to have several advantages over the homogeneous catalysts in asymmetric transformations.1 These include easy separation and isolation, efficient recovery and reuse as well as the possibility to perform the reactions in a continuous process in flow reactors. However, polymer-supported catalysts often suffer from lower catalytic activity and enantioselection compared to their homogeneous counterparts. In an ideal case the advantages of homogeneous asymmetric catalysis, high catalytic activity and selectivity, and those of heterogeneous catalysis, easy isolation and recycling, are combined.

In recent years, metal-free catalysts2–6 have received great attention and have been widely employed in enantioselective catalysis. In particular, BINOL-derived phosphoric acids have emerged as important, versatile and powerful catalysts for an increasing range of enantioselective transformations.7–9 For instance, they have been demonstrated to show excellent catalytic activity in asymmetric activation of aldimines and ketimines as well as carbonyl functionalities.

Typically, these catalysts have to be prepared in a multi-step reaction sequence and in terms of operational simplicity and practicability it would be desirable to recover and to reuse them as often as possible. Herein we report a new methodology to immobilize chiral Brønsted acids onto an insoluble polymer support.

For our initial study we selected BINOL phosphoric acids with styryl substituents in the 3,3′-positions as a model catalyst for catalytic asymmetric transfer hydrogenations.10,11 The incorporated styrene group would allow a straightforward immobilization by cationic, anionic or radical polymerization.

Thus, we prepared catalyst 3a and applied this catalyst initially in the homogeneous Brønsted acid-catalyzed transfer hydrogenation of quinoline 1a using Hantzsch dihydropyridine as hydride source. The resulting tetrahydroquinoline 4a was obtained in good yield and with moderate enantioselectivity (Scheme 1).

Scheme 1. Transfer hydrogenation of quinoline with 3a.

The following “heterogenization” of the catalyst was achieved by cross-linking radical copolymerization of catalyst 3a with styrene and divinylbenzene and resulted in the polymer-supported Brønsted acid catalyst 5a (Scheme 2).

Scheme 2. Cross-linking of catalyst 3a.

In contrast to most polystyrene-based heterogeneous systems, which have to be removed by filtration, we decided to design and prepare the polymer-supported Brønsted acid catalyst in the form of a polymer-stick A. This catalyst stick could subsequently be incorporated in an open reaction container B and the combined system would be placed in a reaction vessel C to perform multiple sequential reactions. Thus, after completion of the reaction the separation of the catalyst can be simply achieved by pulling the container out of the reaction vessel in a tea-bag fashion (Figure 1 B) and apply it again in the next cycle (Figure 1 C).

Figure 1. Polymer supported Brønsted acid catalyst A; polymer-stick in a reaction container B; Brønsted acid catalyst stick at work C.

With the polymer-supported Brønsted acid catalyst 5a in hand we fabricated the newly designed catalyst system B and examined it in the asymmetric reduction of quinoline 1a by employing the same reaction conditions we used for the homogeneous catalyst 3a. The reactions were performed in toluene at 60 °C with a catalyst loading of 5 mol%. After each reaction and separation the immobilized catalyst system was washed and subsequently reused in the next reaction cycle. As shown in Figure 2 the polymer stick is highly stable and can be readily recovered and be reused for at least 10 times. Most importantly and in contrast to many immobilized Lewis base catalysts, we did not observe a noteworthy loss of reactivity and enantioselectivity. Over 10 reaction cycles the new heterogeneous catalyst system exhibited an excellent performance which is comparable to that of the related homogeneous reaction (Scheme 1).

Figure 2. Recovery and re-use of polymer-supported catalyst 5a in the asymmetric transfer hydrogenation of quinoline 1a.

With this encouraging preliminary result, we decided to prepare BINOL phosphates with bulkier substituents in the 3,3′-positions of the BINOL skeleton since better enantioselectivities had typically been observed for such bulkier catalysts. At the same time we chose the linkage for immobilization to be placed into the 6,6′-positions of the catalyst The synthesis route to prepare catalysts 3b and 3c is outlined in Scheme 3. Bromination of commercially available BINOL gave the 6,6′-dibromo-BINOL 8 in excellent yield. Subsequent protection of the hydroxy groups resulted in derivative 9 which was effectively transferred into the dicarboxylic acid by applying a well established lithiation and carboxylation procedure. The following acid-catalyzed esterification with simultaneous deprotection of the hydroxy groups resulted in compound 10, which was directly brominated to give the dibromo diol 11. Compound 11 was subjected to a double Suzuki reaction to obtain 12 which was transformed into the dialdehyde 13 following a reduction and oxidation protocol. Finally, Wittig–Horner reaction and phosphorylation resulted in the desired Brønsted acid catalysts 3b and 3c.

Scheme 3. Preparation of BINOL phosphates 3b and 3c.

The new catalysts 3b and 3c were then tested in the homogeneous enantioselective reduction of benzoxazine 6. The reactions were conducted in chloroform at room temperature with 5 mol% of catalyst (Scheme 4). From the results obtained it can be concluded that the introduction of substitutents in the 6,6′-positions of the BINOL skeleton has no considerable effect on the chemical reactivity but a significant impact on the enantioselectivity. Astonishingly, the application of catalyst 3b bearing a vinyl group in the 6,6′-positions provided the product with only moderate enantioselectivity (64% ee). In contrast and to our delight catalyst 3c performed well and gave the desired product 7 with an excellent enantiomeric excess of 96% ee.

Scheme 4. Transfer hydrogenation of benzoxazine 6 applying catalyst 3b and 3c.

Having established the optimized catalyst 3c we proceeded with the cross-linking. The polymerization was performed in neat styrene and divinylbenzene at 80–90 °C for 40 min to give the immobilized catalyst 5c (Scheme 5).12

Scheme 5. Cross-linking of catalyst 3c to afford polymer stick 5c.

The catalytic ability of the polymer-supported catalyst 5c was subsequently evaluated in multiple transfer hydrogenations of benzoxazine. The reactions proceeded smoothly at room temperature to give the product in good isolated yields and with excellent enantioselectivities. The polymer stick was successfully recycled and reused for at least 12 cycles without loss of activity and selectivity (Figure 3). Again, the catalytic activity and the asymmetric induction are comparable to those of the homogeneous reaction demonstrating the efficiency and applicability of this new catalyst system.

Figure 3. Multiple use of catalyst 5c in the transfer hydrogenation of benzoxazine 6.

Next we examined the stability of the Brønsted acid catalyst stick 5c by performing the transfer hydrogenation consecutively in various solvents. As shown in Figure 4, the catalyst system is not only stable in chlorinated, aromatic and apolar non-protic solvents but provides the product in overall good yields and with excellent enantioselectivities. While the reduction in chlorinated and aromatic solvents afforded the product in 94% ee, the selectivities in dibutyl ether (88% ee) are slightly reduced.

Figure 4. Consecutive transfer hydrogenations in different solvents.

In summary, we have for the first time succeeded in immobilizing chiral BINOL-derived phosphoric acids onto a solid support. The great potential of this new heterogeneous Brønsted acid catalyst system was demonstrated in the catalytic enantioselective transfer hydrogenation of quinolines and benzoxazines. The polymer-supported chiral Brønsted acid catalysts were found to be stable and as effective as their homogeneous counterparts. Furthermore, we were able to design and to develop a new catalyst system which, based on tea-bag approach, can be simply removed from the reaction mixture and can be reused in multiple consecutive catalytic cycles without any loss of enantioselectivity, and more importantly reactivity.

The newly developed catalyst system additionally represents a new approach toward compartmentalization and site-separation of catalysts and allows and facilitates future applications of combined catalysis where, for instance, several heterogeneous catalysts are placed in the same reaction vessel.

Experimental Section

Preparation of the Polymer

To phosphate 3c (3.8 mg) dissolved in CHCl₃ (0.2 mL) was added 1 mL of styrene/divinylbenzene (1/2) and AIBN. The resulting mixture was placed in a tube and the copolymerization was carried out at 80–90 °C until the solution in the tube solidified and the polymer-supported catalyst could be removed from the tube in the form of a stick. The resulting polymer stick 5c was washed with chloroform to remove the residual monomer.

General Procedure for the Transfer Hydrogenation of Benzoxazine

Benzoxazine 6, immobilized catalyst 5c (5 mol%) and Hantzsch dihydropyridine (1.25 equiv.) were suspended in CHCl₃ in a screw-capped vial. The resulting mixture was allowed to stir at room temperature for 20–24 h. The polymer-supported catalyst was removed from the reaction mixture, washed with CHCl₃ and reused for the next reaction. The solvent of the product mixture was removed under reduced pressure and the residue was purified by column chromatography on silica gel to afford product 7.

Acknowledgements

This work was supported by a grant from the European Research Council (ERC 209437). We gratefully acknowledge the DFG, RWTH Aachen University and Evonik-Degussa for financial support.