Ring-closing metathesis is a variant of the olefin metathesis reaction in which alkylidene moieties are exchanged to form a ring. The most common catalysts for this reaction are complexes of molybdenum or ruthenium.
- 1 Introduction
- 2 Mechanism and Stereochemistry
- 3 Scope and Limitations
- 4 Synthetic Applications
- 5 Comparison to Other Methods
- 6 Experimental Conditions and Procedure
- 7 References
Olefin metathesis involves the exchange of two alkylidene groups to generate two new olefins from one or more starting alkenes. Cleavage of the carbon-carbon double bond is accompanied by the formation of two new carbon-carbon double bonds. This reaction was first observed in 1931, investigated by Du Pont and other manufacturers in the 1950's, and finally defined by Calderon in 1967. Partly due to its relevance to petrochemical industry, olefin metathesis reactions have been investigated extensively. Four general classes of reactions have emerged: cross metathesis, an intermolecular reaction of two alkenes; ring-opening metathesis polymerization (ROMP), in which a cyclic alkene opens to form a polyolefin; ring-opening metathesis (ROM), the opening of a cyclic alkene to form a diene; and ring-closing metathesis (RCM), in which reaction of a diene affords a cyclic alkene and a small olefinic byproduct. RCM is the focus of this article (Eq. 1).(1)
Although initial examples of ring-closing metathesis used poorly defined metal catalysts, subsequent development of Schrock-type molybdenum catalysts such as 1 and Grubbs-type ruthenium catalysts such as 2 - 6 greatly expanded the scope and utility of RCM (Eq. 2). In general, molybdenum catalysts display high activity but are unstable toward air or water; ruthenium catalysts are less active but exhibit good selectivity and functional-group compatibility. RCM has been employed extensively in organic synthesis to establish both saturated and unsaturated rings; the reaction can be used to form carbocycles or heterocycles.(2)
Mechanism and Stereochemistry
The generally accepted mechanism for olefin metathesis involves a series of [2+2] cycloadditions and cycloreversions involving the reactant alkenes and catalytic metal carbenes (Eq. 3). In RCM reactions, cycloaddition of one alkene with the catalyst affords metallacyclobutane intermediate 7 containing a pendant olefin. In a cycloreversion step, a small olefin is expelled and new metal carbene intermediate 8 forms, which still contains a tethered alkene. Intramolecular cycloaddition yields new metallacyclobutane 9, which undergoes cycloreversion to expel the metal carbene catalyst and generate the product cyclic alkene.
Because this mechanism relies on both [2+2] cycloadditions and retro-cycloadditions, in general each step is reversible, resulting in an equilibrium mixture of olefinic products. In RCM reactions, reactants are typically designed so that the desired cyclic alkene is accompanied by a small gaseous olefin such as ethylene or propene, the loss of which drives the reaction forward. Highly dilute conditions discourage intermolecular metathesis and thereby also promote RCM.(3)
NMR and UV-visible spectroscopic studies have clarified some mechanistic details in reactions of Grubbs-type (ruthenium) complexes (Eq. 4). Before [2+2] cycloaddition can occur, loss of a phosphine ligand is required to generate a highly reactive 14-electron complex. Second-generation Grubbs catalysts 3 - 6 employ trans N-heterocyclic carbene ligands to accelerate the phosphine dissociation step.(4)
Ring-closing metathesis has been applied for kinetic resolution and desymmetrization, with the latter more common. Schrock-type complexes with a stereogenic center at molybdenum are more often used as catalysts than ruthenium complexes with chiral ligands. For example, chiral molybdenium complex 10 catalyzes the desymmetrization of vinyl ethers to form dihydropyrans with moderate to good enantioselectivity (Eq. 5).(5)
Further modification of heterocycles established through desymmetrizing RCM affords chiral and nonracemic acyclic structures containing carbon-carbon double bonds. For example, Schrock-type complex 11 catalyzes the cyclization of an allylborane, which undergoes oxidation to yield a chiral diol with very high stereoselectivity and moderate yield (Eq. 6).(6)
Scope and Limitations
The vast majority of olefin metathesis reactions are catalyzed by complexes of either molybdenum (Schrock type) or ruthenium (Grubbs type). Molybdenum catalyst 1 was developed before the Grubbs-type catalysts and is highly active, but sensitivity of this catalyst to air and water limits its applicability. Ruthenium catalysts 2 and 3 are less active and cannot be recycled, but exhibit better functional-group tolerance than the rather indiscriminate catalyst 1.
Second-generation Grubbs catalysts 4 - 6 include a strongly donating N-heterocyclic carbene ligand trans to the phosphine ligand, accelerating phosphine dissociation and increasing their activity relative to 2 and 3. In addition, the substrate scope of these catalysts is greater than that of the first-generation Grubbs-type catalysts.
Ruthenium-based complexes have two general limitations. The first is their tendency to form stable Fischer carbenes in the presence of electron-rich olefins such as enol ethers (Eq. 7). The second is their susceptibility to coordination by Lewis bases, which limits their compatibility with functional groups such as amines and phosphines (however, protection strategies can circumvent this limiation; see below).(7)
When selecting a catalyst for RCM, it is important to consider both the reactivity of the catalyst itself and the structures of the substrate and product. The rate of olefin metathesis is strongly affected by the substitution pattern of the alkene(s), with more substituted alkenes reacting more slowly. Steric hindrance near the reacting alkenes may have an effect similar to alkene substitution.
Synthesis of Carbocycles
Molybdenum catalyst 1 and second-generation ruthenium complexes 4 and 5 are most effective for the synthesis of substituted cyclic alkenes. Terminal alkenes are usually the preferred substrates because of their relatively high reactivity. Although the use of air- and water-sensitive catalyst 1 is undesirable from a practical standpoint, it may be necessary in reactions that establish tetrasubstituted double bonds (Eq. 8).(8)
RCM is effective for the preparation of medium-sized rings such as cyclooctenes. Typical catalysts are ruthenium complexes 2 and 4. Conformational constraints are necessary in the substrate to promote cyclization, but geminal disubstitution between the reactive alkenes is often enough to promote cyclization. Substrates with greater rigidity can give rise to more structurally complex cyclooctenes (Eq. 9). Syntheses of nine- to twelve-membered rings via RCM have been reported, although fewer examples are known.(9)
Macrocycles can be prepared in moderate yields from alkenes with some degree of conformational bias. For example, RCM was applied in a total synthesis of (–)-terpestacin to establish a fifteen-membered ring (Eq. 10).(10)
Functional groups such as halides and boronates can be attached to the reactive alkenes as long as a tolerant catalyst such as 2 is used. The functionalized cyclic products thus prepared can then be employed in cross-coupling reactions (Eq. 11).(11)
Synthesis of Heterocycles
Ring-closing metathesis is not limited to the synthesis of carbocycles. Heterocycles containing a carbon-carbon double bond can also be prepared via RCM. Although nitrogen- and oxygen-containing rings are the most common products, heterocycles containing phosphorus, silicon, boron, sulfur, and other elements have also been prepared. It remains important to consider the substitution pattern of the alkene and the activity and functional-group compatibility of the catalyst.
Grubbs-type catalysts are problematic in cyclizations of amines due to coordination of the Lewis-basic nitrogen to ruthenium. However, amides and other protected amines can be used in RCM reactions with Grubbs-type catalysts (Eq. 12).(12)
As in the synthesis of carbocycles, some level of conformational bias in the substrate is important in the synthesis of medium and large heterocycles. For example, bicyclic carbamates containing an eight-membered ring are efficiently synthesized by ring-closing metathesis (Eq. 13). Synthesis of nitrogen-containing rings containing more than nine atoms tends to be problematic.(13)
Phosphonates containing two or more double bonds cyclize in the presence of catalyst 2. The vinyl group reacts preferentially to form a five- rather than a seven-membered ring (Eq. 14).(14)
Cyclic boronates are formed in cross-metathesis reactions of allylic alcohols and allylboron reagents. Treatment with hydrogen peroxide and sodium hydroxide yields stereodefined allylic diols (Eq. 15).(15)
Unsaturated lactams are a biochemically important class of heterocycles that can be prepared via ring-closing metathesis. Catalyst 1 is effective in the preparation of five- or six-membered lactams, but crotonamides must be used as unsubstituted α,β-unsaturated amides coordinate to molybdenum, preventing reaction (Eq. 16).(16)
Unsubstituted α,β-unsaturated esters can likewise coordinate to the metal center and prevent reaction. Including a Lewis acid such as titanium(IV) isopropoxide in the reaction mixture does not interfere with metathesis and prevents coordination to the catalytic metal, enabling reactions of acrylates (Eq. 17).(17)
Ring-closing metathesis has been applied in a large number of syntheses in both academic and industrial contexts. Product rings range in size from five atoms to macrocycles of twenty atoms or more.(18)
The first large-scale industrial application of ring-closing metathesis was reported in the synthesis of BILN 2061, a protease inhibitor with relevance to hepatitis C. Grela's complex 15 includes both an NHC ligand and an aryl isopropoxy group that readily dissociates from the metal center (Eq. 19).(19)
Comparison to Other Methods
Because of the synthetic importance of the alkene functional group, a variety of olefination methods were developed prior to the advent of olefin metathesis. While some of these have intramolecular, ring-closing variants, others have not been applied generally for the synthesis of cyclic alkenes. Cross-coupling reactions of alkenyl halides or alkenyl nucleophiles, which establish carbon-carbon single bonds adjacent to C-C double bonds, have also emerged as complimentary alternatives to olefination reactions.
Although the Wittig reaction is a popular choice for the synthesis of acyclic olefins, few examples of Wittig ring closures have been reported. The related Horner-Wadsworth-Emmons (HWE) reaction is employed much more often in an intramolecular sense for the synthesis of cyclic olefins (Eq. 20).(20)
The Heck reaction combines an alkene nucleophile with an alkenyl or aryl halide electrophile in the presence of a palladium catalyst and base. When the nucleophile and electrophile are present in a single substrate, Heck reaction affords a cyclic olefin (Eq. 21). Cyclization may occur in an endo or exo mode; exo cyclization is most common in the formation of five-membered rings while endo cyclization is typical in formations of six- and seven-membered rings.(21)
The carbopalladation of allenes is a second palladium-catalyzed method for the synthesis of cyclic alkenes. This versatile reaction can be applied for the synthesis of rings of a variety of sizes, and provides a complementary approach to medium-sized rings of eight to twelve members (Eq. 22).(22)
Experimental Conditions and Procedure
Most ring-closing metathesis reactions are carried out at fairly high dilution of the substrate (10 - 50 mM) with catalyst loadings of 5 - 10 mol % and at slightly elevated temperatures (25 - 110 ºC). Molybdenum catalyst 1 exhibits extreme sensitivity to air and water such that use of a glovebox is ideal. On the other hand, ruthenium catalysts are more stable in air and Schlenck tubes are typically used. Standard workup involves concentration of the reaction mixture, aqueous extraction, and purification via silica gel chromatography, recrystallization, or distillation. Because the standard procedure can leave behind traces of ruthenium, more rigorous workup procedures have been developed that use additional ligands, supercritical fluids, and mesoporous silicates to decrease ruthenium concentrations to extremely low levels.
The diene precursor (80 mg, 0.318 mmol) was dissolved in freshly distilled and degassed CH2Cl2 (20 mL) under an argon atmosphere. The solution was again degassed using argon and ruthenium complex 2 (13 mg, 0.016 mmol, 5 mol %) was added. After 24 h of heating the solution at reflux, the conversion was complete as indicated by TLC (pentanes/ether, 4:1, p-anisaldehyde stain, Rf 0.14). The solvent was removed under vacuum and column chromatography with silica gel (pentanes/ether, 4:1 + 0.1% Et3N) of the crude oil afforded the cyclic enesulfonamide (60 mg, 84%): IR (CH2Cl2) 2864, 1614, 1598, 1350, 1167 cm–1; 1H NMR (500 MHz, C6D6) δ 7.66 (d, J = 8.3 Hz, 2H), 6.80 (d, J = 8.1 Hz, 2H), 6.36 (s, 1H), 4.59 (d, J = 4.1 Hz, 1H), 3.21 (t, J = 9.2 Hz, 2H), 1.88 (s, 3H), 1.81 (t, J = 9.2 Hz, 2H); 13C NMR (125 MHz, C6D6) δ 143.7, 134.5, 131.7, 130.0, 128.5, 111.1, 47.7, 29.9, 21.4; HRMS–FAB (m/z): [M + H]+ calcd for C11H13NO2S, 238.0902; found, 238.0903.
- ↑ Yet, L. Org. React. 2016, 89, 1. (link)
- ↑ Astruc, D. New J. Chem. 2005, 29, 42.
- ↑ Calderon, N. Tetrahedron Lett. 1967, 34, 3327.
- ↑ Hérrison, J.-L.; Chauvin, Y. Makromol. Chem. 1971, 141, 161.
- ↑ Sanford, M. S.; Ulman, M.; Grubbs, R. H. J. Am. Chem. Soc. 2001, 123, 749.
- ↑ Ackermann, L.; Fürsnter, A.; Weskamp, T.; Kohl, F. J.; Herrmann, W. A. Tetrahedron Lett. 1999, 40, 4787.
- ↑ Cefalo, D. R.; Kiely, A. F.; Wuchrer, M.; Jamieson, J. Y.; Schrock, R. R.; Hoveyda, A. H. J. Am. Chem. Soc. 2001, 123, 3139.
- ↑ Jernelius, J. A.; Schrock, R. R.; Hoveyda, A. H. Tetrahedron 2004, 60, 7345.
- ↑ Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; DiMare, M.; O'Regan, M. J. Am. Chem. Soc. 1990, 112, 3875.
- ↑ Huang, J.; Stevens, E. D.; Nolan, S. P.; Petersen, J. L. J. Am. Chem. Soc. 1999, 121, 2674.
- ↑ Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
- ↑ Kirkland, T. A.; Grubbs, R. H. J. Org. Chem. 1997, 62, 7310.
- ↑ Boyer, F.-D.; Hanna, I.; Nolan, S. P. J. Org. Chem. 2001, 66, 4094.
- ↑ Trost, B. M.; Dong, G.; Vance, J. A. J. Am. Chem. Soc. 2007, 129, 4540.
- ↑ Chao, W.; Weinreb, S. M. Org. Lett. 2003, 5, 2505.
- ↑ Renaud, J.; Ouellet, S. G. J. Am. Chem. Soc. 1998, 120, 7995.
- ↑ a b Fu, G. C.; Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 7324.
- ↑ White, J. D.; Hrnciar, P. J. Org. Chem. 2000, 65, 9129.
- ↑ Hanson, P. R.; Stoianova, D. Tetrahedron Lett. 1999, 40, 3297.
- ↑ Micalizio, G. C.; Schreiber, S. L. Angew. Chem., Int. Ed. 2002, 41, 152.
- ↑ Ghosh, A. K.; Cappiello, J.; Shin, D. Tetrahedron Lett. 1998, 39, 4651.
- ↑ Houri, A. F.; Xu, Z.; Cogan, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1995, 117, 2943.
- ↑ Nagata, T.; Nakagawa, M.; Nishida, A. J. Am. Chem. Soc. 2003, 125, 7484.
- ↑ Cabrejas, L. M. M.; Rohrbach, S.; Wagner, D.; Kallen, J.; Zenke, G.; Wagner, J. Angew. Chem., Int. Ed. 1999, 38, 2443.
- ↑ Farina, V.; Shu, C.; Zeng, X.; Wei, X.; Han, Z.; Yee, N. K.; Senanayake, C. H. Org. Process Res. Dev. 2009, 13, 250.
- ↑ Yan, B.; Spilling, C. D. J. Org. Chem. 2008, 73, 5385.
- ↑ Owczarczyk, Z.; Lamaty, F.; Vawter, E. J.; Negishi, E. J. Am. Chem. Soc. 1992, 114, 10091.
- ↑ Ma, S.; Negishi, E. J. Am. Chem. Soc. 1995, 117, 6345.
- ↑ Maynard, H. D.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 4137.
- ↑ Gallou, F.; Saim, S.; Koening, K. J.; Bochniak, D.; Horhota, S. T.; Yee, N. K.; Senanayake, C. H. Org. Process Res. Dev. 2006, 10, 937.
- ↑ McEleney, K.; Allen, D. P.; Holliday, A. E.; Crudden, C. M. Org. Lett. 2006, 8, 2663.
- ↑ Kinderman, S. S.; van Maarseveen, J. H.; Schoemaker, H. E.; Hiemstra, H.; Rutjes, F. P. J. T. Org. Lett. 2001, 3, 2045.