Enol Ethers as Substrates for Efficient Zand Enantioselective Ring-Opening / Cross-Metathesis Reactions Promoted by Stereogenic-at-Mo Complexes :

The first examples of catalytic enantioselective ring-opening/cross-metathesis (EROCM) reactions that involve enol ethers are reported. Specifically, we demonstrate that catalytic EROCM of several oxaand azabicycles, cyclobutenes and a cyclopropene with an alkylor aryl-substituted enol ether proceed readily in the presence of a stereogenic-at-Mo monopyrrolide-monoaryloxide. In some instances, as little as 0.15 mol % of the catalytically active alkylidene is sufficient to promote complete conversion within 10 minutes. The desired products are formed in up to 90% yield and >99:1 enantiomeric ratio (er) with the disubstituted enol ether generated in >90% Z selectivity. The enol ether of the enantiomerically enriched products can be easily differentiated from the terminal alkene through a number of functionalization procedures that lead to the formation of useful intermediates for chemical synthesis (e.g., efficient acid hydrolysis to afford the enantiomerically enriched carboxaldehyde). In certain cases, enantioselectivity is strongly dependent on enol ether concentration: larger equivalents of the cross partner leads to the formation of products of high enantiomeric purity (versus near racemic products with one equivalent). The length of reaction time can be critical to product enantiomeric purity; high enantioselectivity in reactions that proceed to >98% conversion in as brief a reaction time as 30 Supporting Information Available. Experimental procedures and spectral data for substrates and products (PDF). This material is available free of charge via the Internet at http://pubs.acs.org. NIH Public Access Author Manuscript J Am Chem Soc. Author manuscript; available in PMC 2013 February 8. Published in final edited form as: J Am Chem Soc. 2012 February 8; 134(5): 2788–2799. doi:10.1021/ja210946z. N IH PA Athor M anscript N IH PA Athor M anscript N IH PA Athor M anscript seconds can be nearly entirely eroded within 30 minutes. Mechanistic rationale that accounts for the above characteristics of the catalytic process is provided. Introduction Advances in catalytic olefin metathesis during the last two decades have transformed the way in which a great number of organic molecules can be prepared.1 Cyclic structures of nearly any size and/or variety as well as a considerable number of unsaturated acyclic molecules are rendered easily accessible through this remarkable class of transformations.2 Nonetheless, major advances remain to be achieved if catalytic olefin metathesis is to reach its true potential.3 Discovery and development of catalysts that promote olefin metathesis efficiently and can control stereoselectivity – that is, furnish high Z or E selectivity and/or enantioselectivity – stands as a critical and challenging objective. Since the first efficient cases of catalytic enantioselective olefin metathesis were reported in 1998 (ring-closing),4,5 ring-opening/cross-metathesis (ROCM) processes have received significant attention.6,7 Chiral Mo alkylidenes and Ru-based carbenes have been developed for desymmetrization of cyclic alkenes, generating carboor heterocyclic dienes enantioselectively. In 2007, we reported the first application of catalytic enantioselective ring-opening/cross-metathesis (EROCM) to the total synthesis of natural product baconipyrone C.8 In spite of the above advances, a number of shortcomings remain unaddressed. One deficiency concerns the limited range of cross partners utilized: nearly all reactions have been with aryl-substituted alkenes. Furthermore, several problems relating to stereochemical control stand unresolved. High enantiomeric ratios are observed in a number of EROCM reactions; in most instances, however, either E alkenes are formed predominantly or exclusively6a–d,7a–c,e or a mixture of olefin isomers is generated with little or no stereochemical control.7d,f Another problem concerns the extensive reaction times (up to 120 hours)7e–f that might be required for achieving high conversion; a more facile transformation is achievable but only at the expense of higher catalyst loadings. The issue of catalyst efficiency is a growing concern with transformations that are promoted by complexes derived from ruthenium, which is a relatively rare and increasingly precious metal. Enol ethers are easily accessible cross partners that might be used in catalytic EROCM reactions9 to afford versatile enantiomerically enriched products; there are, nonetheless, only a small number of cases where such O-substituted alkenes have been utilized as cross partners in intermolecular olefin metathesis reactions; all reported processes have been catalyzed by an achiral complex.10 A disclosure by Ozawa in 2000 outlines a Ru-catalyzed transformation involving norbornene and phenylvinyl ether; the desired product was obtained in only 17% yield and, notably, with 85% Z selectivity.11 Subsequent studies by Rainier offer five additional examples of transformations of ethylvinyl ether or enol acetate with 7-oxaor 7-azanorbornenes, promoted by achiral Ru-based carbenes; nearly equal mixtures of Z and E alkenes were uniformly generated.12 Thus, to the best of our knowledge, catalytic ROCM or EROCM reactions that involve an enol ether and which proceed with effective control of alkene stereoselectivity and/or enantioselectivity have not been disclosed. We have developed stereogenic-at-Mo and W complexes that display the unique ability to catalyze a range of olefin metathesis reactions with unique levels of efficiency and stereoselectively.13 We have demonstrated that the Moor W-based monopyrrolidemonoaryloxides promote ring-closing metathesis of dienes14 or enynes15 in high yield and with exceptional enantioselectivity. Subsequent investigations led us to Yu et al. Page 2 J Am Chem Soc. Author manuscript; available in PMC 2013 February 8. N IH PA Athor M anscript N IH PA Athor M anscript N IH PA Athor M anscript establish that Z-selective EROCM16 and homocoupling of terminal alkenes17 as well as ethenolysis of Z-1,2-disubstituted olefins18 can be performed. We have put forth the first examples of Z-selective cross-metathesis processes with enol ethers serving as one set of cross partners.19 Most recently we have shown that the corresponding tungsten complexes can be used to promote Z-selective macrocyclic ringclosing metathesis.20 In the case of catalytic EROCM, highly enantioand Z-selective, reactions proved to be largely restricted to aryl olefins (i.e., styrenyl derivatives). Such a drawback diminishes utility, since the resulting aryl-substituted alkenes offer a limited range of possibilities for functionalization.21 Here, we report a protocol for efficient, enantioand Z-selective ROCM involving an alkylor aryl-substituted enol ether in combination with oxaor azabicyclic alkenes, cyclobutenes or a cyclopropene. Transformations are promoted by 0.15–3.0 mol % of a stereogenic-at-Mo monopyrrolide-monoaryloxide and proceed to completion, typically at 22 °C, within 30 minutes. The desired cyclic or acyclic dienes, containing an easily differentiable terminal alkene and a vinyl ether in addition to one or more tertiary or quaternary carbon stereogenic centers, are generated in 61–90% yield, 85:15–99:1 enantiomeric ratio (er) and 84:16 to >98:2 Z:E selectivity. We show that, depending on the structure of the cyclic alkene, the amount of the terminal olefin present and the reaction time can have a significant influence on the enantioselectivity and efficiency of the EROCM (but not enol ether stereochemistry). As will be detailed below, such attributes are mechanistically important, offering valuable insights regarding the inner workings of this emerging class of olefin metathesis catalysts.13–20 Results and Discussion 1. Preliminary studies with commonly used Ruand Mo based complexes We began by probing the ability of widely used Ruand Mo-based complexes to catalyze a representative ROCM reaction with an enol ether. We selected silyl-protected oxabicycle 4 as the substrate and examined processes involving commercially and/or easily accessible nbutylvinyl ether 5a or p-methoxyphenylvinyl ether 5b (1.1 equiv vs 4). As the data in entry 1 of Table 1 illustrate, with Ru carbene 1, 69% disappearance of the cyclic alkene is observed after 30 minutes but only 32% of the desired product (6a) is isolated; the remainder of the substrate is likely consumed through oligomerization. Furthermore, the resulting disubstituted enol ether moiety of the 2,4,6-trisubstituted pyran is generated as a 3:1 mixture of E and Z isomers. With the sterically more congested enol ether 5b as the cross partner (entry 2, Table 1), 95% of 4a disappears within 30 minutes but oligomerization of the cyclic alkene represents the predominant pathway. In a similar fashion, with Mobased diolate 2, the relatively strained alkene is consumed rapidly but the desired ROCM product is isolated in 10–14% yield (entries 3–4, Table 1), and the product enol ether is generated either non-selectively (1:1 Z:E; entry 3) or with moderate preference for the Z isomer (80% Z, entry 4). Finally, when chiral Mo diolate 3 is used, disappearance of the starting materials is not detected even after 12 hours. 2. Stereogenic-at-Mo monopyrrolides as catalysts for EROCM reactions with enol ethers and utility in chemical synthesis i. Initial examination of various Mo-based monopyrrolides—Next, we investigated whether stereogenic-at-Mo monopyrrolide-monoalkoxide or aryloxides might not only promote EROCM reactions efficiently and with high enantioselectivity, but deliver the resulting disubstituted enol ether with a high degree of Z selectivity as well. We thus discovered that reactions with Mo-based monopyrrolide monoalkoxide 7 and enol ethers 5a or 5b proceed with significantly higher efficiency than observed with the complexes shown Yu et al. Page 3 J Am Chem Soc. Author manuscript; available in PMC 2013 February 8. N IH PA Athor M anscript N IH PA Athor M anscript N IH PA At