Strategies for the Generation of Molecularly Imprinted Polymeric Nitroxide Catalysts

ORGANIC LETTERS Strategies for the Generation of Molecularly Imprinted Polymeric Nitroxide Catalysts 2005 Vol. 7, No. 22 4879-4882 Christopher D. Anderson, Kenneth J. Shea,* and Scott D. Rychnovsky* Department of Chemistry, UniVersity of California-IrVine, 516 Rowland Hall, IrVine, California 92697-2025 srychnoV@uci.edu Received July 23, 2005 ABSTRACT Two strategies for preparing catalytically active molecularly imprinted nitroxide-containing polymers are outlined. Both strategies rely upon the thermal rearrangement chemistry of tertiary amine N-oxides. To this end, several polymers were prepared and the polymeric nitroxides were revealed by oxidation with m-CPBA. All of the resulting polymeric catalysts proved to be competent mediators of the oxidation of alcohols. Whereas enzymatic processes achieve selectivity by imposing geometric constraints on a given substrate, organic reactions are typically subject to the intrinsic bias of a substrate. One strategy for attaining selectivity by mimicking the geometric constraints imposed by enzymes is the use of molecularly imprinted polymers (MIPs).1 MIPs are typically highly crosslinked networks with defined and accessible “binding sites”. These sites result from the presence of a removable template molecule during the polymerization event. After polymerization, the template can be removed (typically by chemical or physical means) to leave a polymer that retains the shape and complementary polarity of the template molecule (Figure 1). To date, MIPs have been successfully utilized to effect enantioselective hydrolysis,2 diastereoselective and regioselective reduction,3 and enantioselective alkylation,4 among (1) Wulff, G. Chem. ReV. 2002, 102, 1-28 and references therein. (2) (a) Sellergren, B.; Shea, K. J. Tetrahedron: Asymmetry 1994, 5, 1403-1406. (b) Sellergren, B.; Karmalkar, R. N.; Shea, K. J. J. Org. Chem. 2000, 65, 4009-4027. (3) Bystrom, S. E.; Boerje, A.; Akermark, B. J. Am. Chem. Soc. 1993, ¨ 115, 2081-2083. (4) (a) Wulff, G.; Vietmeier, J. Makromol. Chem. 1989, 190, 17171726. (b) Wulff, G.; Vietmeier, J. Makromol. Chem. 1989, 190, 17271735. 10.1021/ol051749n CCC: $30.25 Published on Web 09/28/2005 © 2005 American Chemical Society other transformations. Substrate-selective metal-catalyzed oxidation has also been accomplished by the use of MIPs.5 Nitroxides, such as TEMPO (1), are stable free radicals that are efficient catalysts for the oxidation of alcohols to Figure 1. General strategy for the generation of molecularly imprinted polymers. Figure 2. Proposed mechanism of TEMPO-mediated oxidation under alkaline conditions. ketones, aldehydes, and carboxylic acids (Figure 2). These catalysts have found widespread use in synthetic chemistry.6 Several studies have focused on the mechanism of this process, and the currently accepted mechanism under alkaline reaction conditions is depicted in Figure 2.7 The utility of this organocatalyst has attracted our group and others to investigate the use of chiral nitroxides for the kinetic resolution of secondary alcohols; however, these endeavors have met with limited success.8 Our longstanding interest in the chemistry of nitroxides, coupled with the potential of molecularly imprinted polymers for achieving selectivity, led us to initiate a project investigating the interface of these two areas. To this end, we decided to approach the generation of nitroxide polymers from a unique perspective. To date, several immobilized TEMPO moieties have been prepared, including silicasupported TEMPO,9 MCM-41-supported TEMPO,10 sol-gel TEMPO,11 PEG-TEMPO,12 polynorbornene-derived TEMPO,13 and polyamine TEMPO.14 These are all effective catalysts for the oxidation of alcohols to aldehydes and can be (5) (a) Efendiev, A. A. Macromol. Symp. 1994, 80, 289-313. (b) Efendiev, A. A.; Orudzhev, D. D.; Shakhtakhinsky, T. N.; Kabanov, V. A. In Homogeneous and Heterogeneous Catalysis; Yermakov, Y., Likholobov, V.; Eds.; VNU Science: Utrecht, 1986; pp 717-725. (6) Merbouh, N.; Bobbitt, J. M.; Brueckner, C. Org. Prep. Proced. Int. 2004, 36, 3-31. (7) (a) Golubev, V. A.; Borislavskii, V. N.; Alexandrov, A. L. Bull. Acad. Sci. USSR, DiV. Chem. Sci. 1977, 1874-1881. (b) Golubev, V. A.; Sen, V. D.; Rozantsev, E. G. Bull. Acad. Sci. USSR, DiV. Chem. Sci. 1979, 1927Ä 1931. (c) Semmelhack, M. F.; Schmid, C. R.; Cortes, D. A. Tetrahedron ´ Lett. 1986, 27, 1119-1122. (8) (a) Formaggio, F.; Bonchio, M.; Crisma, M.; Peggion, C.; Mezzato, S.; Polese, A.; Barazza, A.; Antonello, S.; Maran, F.; Broxterman, Q. B.; Kaptein, B.; Kamphuis, J.; Vitale, R. M.; Saviano, M.; Benedetti, E.; Toniolo, C. Chem. Eur. J. 2002, 8, 84-93. (b) Kashiwagi, Y.; Kurashima, F.; Kikuchi, C.; Anzai, J.; Osa, T.; Bobbitt, J. M. Chem. Commun. 1999, 1983-1984. (c) Naik, N.; Braslau, R. Tetrahedron 1998, 54, 667-696. (d) Rychnovsky, S. D.; McLernon, T. L.; Rajapakse, H. J. J. Org. Chem. 1996, 61, 1194-1195. (e) Ma, Z.; Huang, Q.; Bobbitt, J. M. J. Org. Chem. 1993, 58, 4837. (9) (a) Bolm, C.; Fey, T. Chem. Commun. 1999, 1795-1796. (b) Fey, T.; Fischer, H.; Bachmann, S.; Albert, K.; Bolm, C. J. Org. Chem. 2001, 66, 8154-8159. (10) Brunel, D.; Fajula, F.; Nagy, J. B.; Deroide, B.; Verhoef, M. J.; Veum, L.; Peters, J. A.; van Bekkum, H. Appl. Catal., A 2001, 213, 7382. (11) (a) Ciriminna, R.; Blum, J.; Avnir, D. Pagliaro, M. Chem. Commun. 2000, 1441-1442. (b) Ciriminna, R.; Bolm, C.; Fey, T.; Pagliaro, M. AdV. Synth. Catal. 2002, 344, 159-163. (12) (a) Pozzi, G.; Cavazzini, M.; Quici, S.; Benaglia, M.; Dell′Anna, G. Org. Lett. 2004, 6, 441-443. (b) Ferreira, P.; Hayes, W.; Phillips, E.; Rippon, D.; Tsang, S. C. Green Chem. 2004, 6, 310-312. (13) Tanyeli, C.; Gumus, A. Tetrahedron Lett. 2003, 44, 1639-1642. ¨ ¨¸ 4880 Figure 3. Allylamines. removed from reaction mixtures by filtration. However, these catalysts were prepared with monomers incapable of imprinting. In contrast to these methods, we sought to use 2,2,6,6tetramethylpiperidine derivatives to template the polymer and then selectively remove the N-substituent to reveal a catalytically active nitroxide with an imprinted cavity adjacent to the N-O• catalyst site. One feature of enzymes that is desirable to emulate is the transition-state stabilization imparted by the structure of the active site. As such, our initial plan was to template structures that would effectively mimic N-oxide 2. Unfortunately, it was found that all of the 2,2,6,6-tetramethylpiperidine N-oxide derivatives we attempted to synthesize decomposed thermally above 0 °C, precluding the use of such N-oxides as templates for the desired MIPs. Thus an alternate strategy of imprinting amine structures, such as allylic amines 4 and O-alkyl hydroxylamines 9b, 16, and 17, was adopted. These templates omit the zwitterionic character of the proposed transition structure 2 but maintain an appropriately positioned templating element. Initially, allylic amines 4a-c were targeted to evaluate the potential of this strategy for the preparation of catalytically active nitroxide polymers (Figure 3). An alkylation approach to these molecules was identified as the most efficient means of attaining the desired structures. 1-Bromo2-hexene reacted with amine 6 at elevated temperature to provide hydroxy amine 7 in 53% yield, which could be derivatized to acetate 3b and methacrylate 4b (Scheme 1). Scheme 1. Preparation of Tertiary Amine Template 4b Allylamine 4a and allylic amine 4c were prepared by an analogous sequence. To validate this strategy, 3a was treated (14) Dijksman, A.; Arends, I. W. C. E.; Sheldon, R. A. Chem. Commun. 2000, 271-272. Org. Lett., Vol. 7, No. 22, 2005 with m-CPBA to provide acetoxy TEMPO 12a as the only product by GC analysis. This occurs via initial generation of N-oxide 8 followed by Meisenheimer [2,3]-rearrangement15 to provide an O-allyl hydroxylamine 9a, which undergoes oxidation to provide another N-oxide, 10a (Scheme 2). Decomposition by a Cope-like elimination generates a Scheme 2. Scheme 3. Synthesis of Diisopropylidene Galactose Template 16 Proposed Mechanism of Template Removal and washing, the resultant colorless polymers were treated with excess mCPBA to reveal the pink-colored imprinted TEMPO catalysts.19, 20 hydroxylamine 11a, which is subsequently oxidized to the nitroxide 12a. The one limitation of this strategy is the inability to easily access allylic amines 5 in enantiopure form, which would provide entry into chiral templates. Given the limitation of the allylamine templates, we envisaged that O-alkylated hydroxylamines might be suitable templates for MIPs. Additionally, this method would allow direct access into chiral templates. To this end, iodo-galactose 13 was photolyzed in the presence of nitroxide 14 and HSnBu3 to give silyl-protected hydroxylamine 15 in 71% yield.16 Conversion to a polymerizable methacrylate derivative 16 was achieved in 70% yield over two steps. O-Alkyl hydroxylamines 9b and 17 were prepared by a similar sequence. As a validation of this approach for generating nitroxides, hydroxylamine 16 was treated with m-CPBA to give nitroxide 12 in quantitative yield. A proposed mechanism of oxidative removal of the O-alkyl hydroxylamine template moiety is illustrated in Scheme 2. With the desired polymerizable templates in hand, templates 4a-c, 9b, 16, and 17 (5 mol % template) were each copolymerized with ethylene glycol dimethacrylate (EGDMA) and methyl methacrylate (MMA) under free-radical conditions using acetonitrile as porogen.17,18 After crushing (15) Albini, A. Synthesis 1993, 263-277 and references therein. (16) Weigel, T. M.; Liu, H.-w. Tetrahedron Lett. 1988, 29, 42214224. (17) Spivak, D.; Gilmore, M. A.; Shea, K. J. J. Am. Chem. Soc. 1997, 119, 4388-4393. (18) The polymerization mixture was composed of 80 mol % EDGMA, 14 mol % MMA, 5 mol % template, 1 mol % AIBN, and a volume of CH3CN equal to the four other components. Org. Lett., Vol. 7, No. 22, 2005 After surveying known TEMPO oxidation protocols with the catalyst derived from 4b, Anelli’s procedure (NaOCl, NaHCO3, KBr, H2O, CH2Cl2, 0 °C) emerged as the best method for use with this polymeric catalyst.21 The oxidations of primary alcohols were rapid (