Targeting an Achilles heel in olefin metathesis: A strategy for high-yield synthesis of second-generation Grubbs methylidene catalysts

Abstract

The first clean, high-yield route is presented to methylidenes RuCl₂(L)(PCy₃)(=CH₂) (L = H₂IMes or IMes), key vectors for catalysis and deactivation in many olefin metathesis reactions. (H₂IMes = N,N′-bis(mesityl)imidazolin-2-ylidene; IMes = N,N′-bis(mesityl)imidazol-2-ylidene).

---

The high activity and ease of handling of “second-generation” ruthenium metathesis catalysts (e.g.GII, GII′; Fig. 1) have greatly expanded the scope of olefin metathesis methodologies.¹ Mounting industrial interest underscores the need for more detailed understanding of catalyst behaviour under planned conditions of implementation. Central in this context is the behaviour of the methylidene derivatives that represent the catalyst resting states in ring-closing metathesis (RCM) and cross-metathesis (CM) reactions of terminal olefins. These species have been identified as key vectors for both catalyst decomposition^2,3 and metathesis. To date, difficulties encountered in synthesis of GII=CH₂ (vide infra), compounded by the poor initiation efficiencies characteristic of the second-generation catalysts,^4–6 have restricted direct study of these important intermediates.⁷ Here we report the first clean, high-yield route to methylidene complexes GII=CH₂ (L = H₂IMes) and GII′=CH₂ (L = IMes).

Fig. 1 Metathesis catalysts discussed. IMes = N,N′-bis(mesityl)imidazol-2-ylidene; H₂IMes = N,N′-bis(mesityl)imidazolin-2-ylidene.

The literature route to GII=CH₂ (involving CM of benzylidene GII with ethylene at 50 °C; Scheme 1)⁴ affords the desired product in <40% yield, the balance being unidentified byproducts. The implied rapidity of decomposition is unexpected. The starting complex GII has a reported half-life of >1 month in benzene at 55 °C,⁷ and while the 5.7 h half-life found⁷ for the methylidene target is much shorter, this still exceeds the 1.5 h timescale of the CM reaction by a considerable margin. Of note, however, is the accelerated decomposition observed for various Ru metathesis catalysts in the presence of ethylene.^7–10 We suspected that the low yield of GII=CH₂ might originate in the vulnerability of intermediates formed during ethylene exchange, particularly four-coordinate RuCl₂(H₂IMes)(=CH₂) (C, Scheme 2), metallacyclobutane B, and – perhaps most susceptible⁸ – the unsubstituted metallacyclobutane formed via degenerate CM of GII=CH₂ with ethylene.

Scheme 1 Previously reported route to methylidene GII=CH₂.⁴

Scheme 2 Details of the synthetic sequence of Scheme 1 (L = H₂IMes), highlighting the high-commitment nature of A and C (see text). A parallel sequence involves degenerate exchange of GII=CH₂ with ethylene via an unsubstituted metallacyclobutane analogous to B.

Kinetics studies by the groups of Grubbs⁴ and Chen¹¹ indicate that A (L = H₂IMes) reacts preferentially with olefin, rather than free PCy₃ (a propensity that led Chen to classify the second-generation complexes as “high-commitment” catalysts).¹¹ Indeed, this bias is one factor underlying the exceptional metathesis activity of the N-heterocyclic carbene (NHC) catalysts. In the present context, however, high commitment is detrimental: the bias toward reaction with ethylene results in competitive inhibition of the desired ligand exchange, and prolongs the time that the catalyst spends in its least stable, phosphine-free states. The thermal sensitivity of GII in the presence of ethylene^7–9,10c (as in the synthesis shown in Scheme 1) is evidently much greater than that suggested by model studies in the absence of olefin.

The limitations intrinsic to ethenolysis led us to explore an alternative approach to the second-generation methylidene complexes, based on ligand exchange of GI=CH₂ with free NHCs (H₂IMes, IMes; Scheme 3). While this synthetic strategy may seem counter-intuitive, given that the half-life of GI=CH₂ (40 min at 55 °C in C₆D₆)² is much shorter than that of GII, it is designed to exploit the low-commitment nature of the first-generation complex: that is, its propensity for rapid reaction with free PCy₃^4,11 or, potentially, NHC donors. We considered that fast uptake of the Lewis base, coupled with elimination of the vulnerable metallacyclobutane intermediate, could potentially improve reaction rates and yields.

Scheme 3 Ligand exchange route to second-generation methylidene complexes.

Access to the first-generation methylidene complex GI=CH₂ in high purity is a prerequisite for the planned approach. In our hands, the reported¹² synthesis of GI=CH₂via CM with ethylene resulted in persistent contamination by residual GI, consistent with the equilibrium nature of this reaction.¹³ Essentially quantitative conversions could be readily attained, however, by washing the crude product with pentane to remove styrene, and re-subjecting it to the ethylene treatment. We obtained GI=CH₂ free of GI after the second pass, in an overall isolated yield of 85%.

With clean GI=CH₂ in hand, we turned to installation of the H₂IMes and IMes ligands. Earlier work described the efficiency with which GII′ can be obtained by treating GI with pure IMes.^14–16 Use of isolated IMes, in preference to the in situ-generated carbene, greatly simplified workup and purification, as PCy₃ was then the only adventitious species present at the end of reaction. Building on this precedent, we chose to use pure free IMes and H₂IMes to synthesize the methylidene complexes of interest. The free carbenes are readily accessible via the established procedures,^17,18 and are now also commercially available (Strem).

A potential complication in our intended use of GI=CH₂ as a precursor for ligand exchange is the low room-temperature lability of the PCy₃ ligand,⁴ which necessitates use of elevated temperatures. While both free H₂IMes and free IMes show excellent thermal stability,‡ the vulnerability of GI=CH₂ is evident from the discussion above. On heating a benzene solution of GI=CH₂ and free NHC at 60 °C, however, we found that ligand exchange out-competes decomposition. Thus, NMR-scale experiments in C₆D₆ revealed rapid formation of GII=CH₂ (99% by 22 min; Fig. 2a), with excellent agreement between conversions and in situ yields, as judged from integration against internal standard. Even on increasing reaction times to 45 min, to address any increase in timescale required for preparative-scale experiments, no decomposition was evident. It may be noted that this longer reaction period remains well within the nearly six-hour half-life of the product.⁷GII=CH₂ was obtained in 81% isolated yield, despite some losses incurred by its partial solubility in the cold pentane used to extract the PCy₃ co-product. The corresponding reaction with free IMes likewise enables quantitative formation of GII′=CH₂ (Fig. 2b), which was isolated in similar yields (78%).

Fig. 2 Kinetics of ligand exchange of GI=CH₂ (black line) with isolated free NHCs to afford (a) H₂IMes derivative GII=CH₂ (blue), or (b) IMes derivative GII′=CH₂ (red). Conditions: C₆D₆, 60 °C; conversions measured by integration vs. trimethoxybenzene.

The foregoing describes the first clean, high-yield route to the second-generation Grubbs methylidene complexes GII=CH₂ and GII′=CH₂. A strategy based on ligand exchange of RuCl₂(PCy₃)₂(=CH₂) with free H₂IMes or IMes eliminates the decomposition that severely limits the yields attainable in metathetical exchange of benzylidene GII with ethylene. Use of the isolated free NHCs (both of which are now commercially available) also contributes to high purity with minimal workup, as the only byproduct in the reaction is readily-removed PCy₃. Given the importance of these methylidene resting-state species as vectors for both metathesis and catalyst deactivation, we anticipate that these straightforward, high-yield routes to GII=CH₂ and GII′=CH₂ will aid significantly in clarifying key reaction pathways in olefin metathesis.

Acknowledgements

This work was supported by the National Science and Engineering Research Council (NSERC) of Canada.

Notes and references

1. For recent leading reviews, see: (a) X. Miao, P. H. Dixneuf, C. Fischmeister and C. Bruneau, Green Chem., 2011, 13, 2258–2271; (b) G. C. Vougioukalakis and R. H. Grubbs, Chem. Rev., 2010, 110, 1746–1787; (c) M. Mori, Materials, 2010, 3, 2087–2140; (d) C. Samojlowicz, M. Bieniek and K. Grela, Chem. Rev., 2009, 109, 3708–3742; (e) S. Monfette and D. E. Fogg, Chem. Rev., 2009, 109, 3783–3816.
2. M. Ulman and R. H. Grubbs, J. Org. Chem., 1999, 64, 7202–7207.
3. P. W. N. M. van Leeuwen and J. C. Chadwick, Homogeneous Catalysts: Activity – Stability – Deactivation, Wiley-VCH, Weinheim, 2011, ch. 10.
4. M. S. Sanford, J. A. Love and R. H. Grubbs, J. Am. Chem. Soc., 2001, 123, 6543–6554.
5. D. E. Fogg and H. M. Foucault, in Comprehensive Organometallic Chemistry III, ed. R. H. Crabtree and D. M. P. Mingos, Elsevier, Oxford, 2007, vol. 11, pp. 623–652.
6. L. Cavallo, J. Am. Chem. Soc., 2002, 124, 8965–8973.
7. A rare direct study of GII=CH₂ involved thermolysis in benzene in the presence and absence of ethylene. See: S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day and R. H. Grubbs, J. Am. Chem. Soc., 2007, 129, 7961–7968.
8. E. F. van der Eide and W. E. Piers, Nat. Chem., 2010, 2, 571–576.
9. W. J. van Rensburg, P. J. Steynberg, W. H. Meyer, M. M. Kirk and G. S. Forman, J. Am. Chem. Soc., 2004, 126, 14332–14333.
10. For studies demonstrating the detrimental effect of ethylene on the rates and yields of Ru-catalyzed cross-metathesis and ring-closing metathesis of olefins, see: (a) K. A. Burdett, L. D. Harris, P. Margl, B. R. Maughon, T. Mokhtar-Zadeh, P. C. Saucier and E. P. Wasserman, Organometallics, 2004, 23, 2027–2047; (b) Z. Lysenko, B. R. Maughon, T. Mokhtar-Zadeh and M. L. Tulchinsky, J. Organomet. Chem., 2006, 691, 5197–5203; (c) S. Monfette, M. Eyholzer, D. M. Roberge and D. E. Fogg, Chem.–Eur. J., 2010, 16, 11720–11725 . In contrast, initiation rates for the Piers phosphonium alkylidene catalysts are dramatically increased by addition of a catalytic amount of ethylene. See: ; (d) E. M. Leitao, E. F. van der Eide, P. E. Romero, W. E. Piers and R. McDonald, J. Am. Chem. Soc., 2010, 132, 2784–2794.
11. C. Adlhart and P. Chen, Helv. Chim. Acta, 2003, 86, 941–949.
12. P. Schwab, R. H. Grubbs and J. W. Ziller, J. Am. Chem. Soc., 1996, 118, 100–110.
13. In comparison, an equilibrium yield of just 20% GI=CH₂ was reported on treating vinylalkylidene complex RuCl₂(PCy₃)₂(=CHCH=CPh₂) with 100 psi C₂H₄ at 50 °C in CD₂Cl₂. See ref. 12.
14. M. Scholl, T. M. Trnka, J. P. Morgan and R. H. Grubbs, Tetrahedron Lett., 1999, 40, 2247–2250.
15. J. Huang, E. D. Stevens, S. P. Nolan and J. L. Petersen, J. Am. Chem. Soc., 1999, 121, 2674–2678.
16. X. Bantreil and S. P. Nolan, Nat. Protoc., 2011, 6, 69–77.
17. A. J. Arduengo, R. Krafczyk, R. Schmutzler, H. A. Craig, J. R. Goerlich, W. J. Marshall and M. Unverzagt, Tetrahedron, 1999, 55, 14523–14534.
18. A. J. Arduengo, H. V. R. Dias, R. L. Harlow and M. Kline, J. Am. Chem. Soc., 1992, 114, 5530–5534.

---

Footnotes

† Electronic supplementary information (ESI) available: Full experimental details and NMR spectra. See DOI: 10.1039/c2cy20213a

‡ ¹H NMR experiments showed no decomposition of H₂IMes or IMes over 10 h at 60 °C in C₆D₆, as indicated by integration against an internal standard.

---