Increased efficiency in cross-metathesis reactions of sterically hindered olefins

Article abstract of DOI:10.1021/ol702624n

Efficiency in olefin cross-metathesis reactions is affected upon reducing the steric bulk of N-heterocyclic carbene ligands of ruthenium-based catalysts. For the formation of disubstituted olefins containing one or more allylic substituents, the catalyst

Full text of DOI:10.1021/ol702624n

ORGANIC
LETTERS
2008
Vol. 10, No. 3
441-444
Increased Efficiency in Cross-Metathesis
Reactions of Sterically Hindered Olefins
Ian C. Stewart, Christopher J. Douglas, and Robert H. Grubbs*
The Arnold and Mabel Beckman Laboratory of Chemical Synthesis,
DiVision of Chemistry and Chemical Engineering, California Institute
of Technology, Pasadena, California 91125
rhg@caltech.edu
Received October 29, 2007
ABSTRACT
Efficiency in olefin cross-metathesis reactions is affected upon reducing the steric bulk of N-heterocyclic carbene ligands of ruthenium-based
catalysts. For the formation of disubstituted olefins containing one or more allylic substituents, the catalyst bearing N-tolyl groups is more
efficient than the corresponding N-mesityl catalyst. In contrast, the formation of trisubstituted olefins is more efficient using the N-mesityl-
containing catalyst. A hypothesis to explain this dichotomy is described.
Olefin cross-metathesis (CM)1 is a powerful synthetic tool
for the preparation of functionalized alkenes due in large
part to the advent of catalysts 12a and 22b that contain an
N-heterocyclic carbene (NHC) ligand (Figure 1).3 These
boronates,3i,j and silanes3k to be utilized as reactive CM
partners. We have described a general model for selectivity
in CM,3h but a number of limitations still hinder its
widespread application in organic synthesis. One of those
issues is the tolerance of steric conjestion on or around the
reactive olefin moiety.4a Wagener and co-workers have
described the challenge of even simple allylic methyl
(2) (a) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953-956. (b) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A.
H. J. Am. Chem. Soc. 2000, 122, 8168-8179.
(3) For leading references on CM methodology development, see: (a)
O’Leary, D. J.; Blackwell, H. E.; Washenfelder, R. A.; Grubbs, R. H.
Tetrahedron Lett. 1998, 39, 7427-7430. (b) Chatterjee, A. K.; Grubbs, R.
H. Org. Lett. 1999, 1, 1751-1753. (c) Chatterjee, A. K.; Morgan, J. P.;
Scholl, M.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122, 3783-3784.
(d) Chatterjee, A. K.; Choi, T. L.; Grubbs, R. H. Synlett 2001, 1034-
1037. (e) Choi, T. L.; Chatterjee, A. K.; Grubbs, R. H. Angew. Chem.,
Int. Ed. 2001, 40, 1277-1279. (f) Grela, K.; Bieniek, M. Tetrahedron
Lett. 2001, 42, 6425-6428. (g) Chatterjee, A. K.; Sanders, D. P.; Grubbs,
R. H. Org. Lett. 2002, 4, 1939-1942. (h) Chatterjee, A. K.; Choi, T. L.;
Sanders, D. P.; Grubbs, R. H. J. Am. Chem. Soc. 2003, 125, 11360-11370.
(i) Morrill, C.; Grubbs, R. H. J. Org. Chem. 2003, 68, 6031-6034. (j)
Morrill, C.; Funk, T. W.; Grubbs, R. H. Tetrahedron Lett. 2004, 45, 7733-
7736. (k) BouzBouz, S.; Boulard, L.; Cossy, J. Org. Lett. 2007, 9, 3765-
3768.
Figure 1. Ruthenium-based olefin metathesis catalysts.
more active catalysts allow 1,1-disubstituted olefins,3c R,â-
unsaturated carbonyls,3e and vinyl phosphonates,3d sulfones,3f
(1) For recent reviews, see: (a) Connon, S. J.; Blechert, S. Angew. Chem.,
Int. Ed. 2003, 42, 1900-1923. (b) Handbook of Metathesis; Grubbs, R.
H., Ed.; Wiley-VCH: Weinheim, Germany, 2003.
10.1021/ol702624n CCC: $40.75
© 2008 American Chemical Society
Published on Web 01/05/2008

substituents in CM with both Ru- and Mo-based catalyts.4b
Herein we investigate the effect of reducing the steric bulk
of the NHC ligand on the efficiency of sterically challenging
CM reactions.
(entry 4). The allylic amide in entry 5, readily prepared by
an Overman rearrangment,6 was challenging for both cata-
lysts investigated. Notably, catalyst 3 performed better in
all cases examined.
We recently reported the enhanced reactivity of cata-
lyst 3 for the formation of tetrasubstituted olefins by ring-
closing metathesis (RCM).5 On the basis of this improved
efficiency we thought that this catalyst, which bears N-tolyl
rather than N-mesityl groups in the NHC ligand, might
also exhibit increase efficiency in CM reactions. As an ini-
tial assay, we looked at the cross metathesis of but-3-en-2-
yl benzoate (4) with cis-1,4-diacetoxy-2-butene to form
allylic benzoate 5 (eq 1). Phosphine-containing catalyst 1
We next investigated the efficiency of catalysts 2 and 3
in CM reactions of alcohol 6 (Table 2). Good to excellent
Table 2. CM Reactions of Tertiary Alcohol 6
afforded the desired product in only 38% yield,3h whereas
the phosphine-free catalyst 2 furnished 59% of compound
5, likely due to increased catalyst lifetime. Catalyst 3, how-
ever, proved to be most active in this series, producing the
desired CM product in an impressive 87% yield. With this
promising initial result, we embarked on a systematic com-
parison of catalysts 2 and 3 in a variety of sterically
challenging CM reactions.
The formation of disubstituted olefins bearing bulky
substituents in the allylic postion proceeds in low yields with
phosphine-containing catalyst 1. Both phosphine-free cata-
lysts 2 and 3 exhibited increased efficiency in CM of a
number of challenging substrates with 5-acetoxy-1-pentene
(Table 1). Allylic methyl groups have been reported to
a E/Z > 20:1 in all cases. b Isolated yields. c Isolated as a 3:2 mixture of
diastereomers.
yields of the cross product were obtained with an allylic
phenyl (entry 2) and benzoate substituent (entry 3). Inclusion
of additional allylic substituents led to lower yields, but
catalyst 3 afforded the densely substituted product in good
yield (entry 4). Again, N-tolyl catalyst 3 outperformed
N-mesityl catalyst 2 in all cases.
In stark contrast to the results listed in Tables 1 and 2,
the formation of trisubstituted olefins by CM is less efficient
using catalyst 3 compared with 2 (Table 3). Reaction of
Table 1. Formation of Disubstituted Olefins Bearing Allylic
Substituents
Table 3. Formation of Trisubstituted Olefins by CM
a Isolated yields.
a E/Z > 20:1 in all cases. b Isolated yields.
methylenecyclohexane with 5-acetoxy-1-pentene afforded the
undergo CM reactions with reduced efficiency,4 but clearly
this substituent is easily tolerated by catalyst 3, as the desired
product is obtained in 98% yield (entry 2). Even the very
bulky OTBDPS group is accommodated with this catalyst
cross product in 78% and 60% yield with catalysts 2 and 3,
(4) (a) Ulman, M.; Grubbs, R. H. Organometallics 1998, 17, 2484-
2489. (b) Courchay, F. C.; Baughman, T. W.; Wagener, K. B. J. Organomet.
Chem. 2006, 691, 585-594.
442
Org. Lett., Vol. 10, No. 3, 2008

respectively (entry 1). The addition of an allylic substituent
to the 1,1-disubstituted olefin partner drastically reduced the
efficiency of the CM reaction as the desired product was
obtained in a mere 17% yield using 2, whereas 3 afforded
none of the trisubstituted olefin in this case.
The divergence in relative efficiencies of these two
catalysts for the formation of di- and trisubstituted olefins
suggests that at least two parameters are at play here. There
are both productive and unproductive olefin metatheses
occurring in any given CM reaction. For example, if a 1,2-
disubstituted metallacyclobutane is formed, cycloreversion
will generate a ruthenium methylidene and the desired CM
product (Path A in Figure 2). However, if olefin coordination
Figure 3. Relevant π-complexes in the formation of trisubstituted
olefins by CM. [Ru] ) RuCl2.
ing a similar number of turnovers as 2, but the smaller N-tolyl
ligand leads to an increase in the number of unproductive
reactions, resulting in lower yields for the desired CM
product.
This steric-based argument suggests that increasing the
steric bulk of the NHC ligand should increase the yields for
the formation of trisubstituted olefins by CM. In support of
this hypothesis, catalyst 7,9 which displays N-2,6-diisopro-
pylphenyl substituents, affords excellent yields, and more
importantly higher than those with catalysts 2 or 3, of the
desired products (eqs 2 and 3). The addition of an allylic
substituent to a 1,1-disubstituted olefin (e.g., Table 3, entry
2) complicates this trend, from which catalyst 2 emerges as
the most efficient (eq 4).
Figure 2. Productive and unproductive CM pathways.
leads to a 1,3-disubstituted metallacyclobutane (Path B),
collapse of this intermediate does not result in a productive
CM reaction, but does constitute a catalyst turnover event.
Thus both selectiVity of metallacyclobutane formation and
the total number of catalyst turnoVer eVents (i.e., catalyst
stability) influence the efficiency of cross-metathesis reac-
tions.
These issues of regioselectivity are especially important
in the formation of trisubstituted olefins by CM. While the
smaller NHC ligand in 3 allows larger reactants to be
accommodated in the formation of disubstituted olefins by
CM, it likely also favors unproductive pathways with 1,1-
disubstituted olefins. Specifically, the relative selectivity for
the formation of π-complex B over A7 is likely lower than
the selectivity of D versus C due to the smaller N-tolyl
containing ligand (Figure 3).8 In other words, by decreasing
the size of the NHC ligand the rate of unproductive cross-
metathesis pathways (Path B) may be increased relative to
productive pathways (Path A). Catalyst 3 could be perform-
(5) Stewart, I. C.; Ung, T.; Pletnev, A. A.; Berlin, J. M.; Grubbs, R. H.;
Schrodi, Y. Org. Lett. 2007, 9, 1589-1592.
(6) Overman, L. E. Angew. Chem., Int. Ed. Engl. 1984, 23, 579-586.
(7) For simplicity only the syn conformation of the two N-tolyl
substituents for A and B is shown in Figure 3.
In summary, by reducing the steric bulk of the NHC ligand
in ruthenium-based metathesis catalysts, increased efficien-
cies for the formation of sterically challenging disubstituted
olefins was observed. The formation of trisubstituted olefins
by CM, however, is more efficient using bulkier NHC
ligands, likely due to the selectivity of productive versus
unproductive pathways.
(8) A cis approach is implied in Figure 3, though aside from the
significant increase in reactivity with catalyst 2 compared to 3 in a number
of sterically demaning examples, we have no evidence for this geometry in
CM reactions. This geometry is supported by isolated π-complexes, for
example, see: Anderson, D. R.; Hickstein, D. D.; O’Leary, D. J.; Grubbs,
R. H. J. Am. Chem. Soc. 2006, 128, 8386-8387. The influence of ligand
sterics may also be manifested in ruthenacyclobutane intermediates, but to
a lesser extent due to the trans arrangement of metallacycle and the NHC
ligand. For pertinent examples, see: Wenzel, A. G.; Grubbs, R. H. J. Am.
Chem. Soc. 2006, 128, 16048-16049. Romero, P. E.; Piers, W. E. J. Am.
Chem. Soc. 2005, 127, 5032-5033. Romero, P. E.; Piers, W. E. J. Am.
Chem. Soc. 2007, 129, 1698-1704.
(9) Courchay, F. C.; Sworen, J. C.; Wagener, K. B. Macromolecules
2003, 36, 8231-8239.
Org. Lett., Vol. 10, No. 3, 2008
443

Acknowledgment. The authors thank the NIH (postdoc-
toral fellowships to I.C.S. and C.J.D. and Grant 5RO1GM-
31332 to R.H.G.), Dr. Patricio Romero (Caltech) for helpful
discussions, and Materia Inc. for their generous donation of
catalysts used in these studies.
Supporting Information Available: Experimental pro-
cedures and characterization of all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL702624N
444
Org. Lett., Vol. 10, No. 3, 2008