A general model for selectivity in olefin cross metathesis

Article abstract of DOI:10.1021/ja0214882

In recent years, olefin cross metathesis (CM) has emerged as a powerful and convenient synthetic technique in organic chemistry; however, as a general synthetic method, CM has been limited by the lack of predictability in product selectivity and stereoselectivity. Investigations into olefin cross metathesis with several classes of olefins, including substituted and functionalized styrenes, secondary allylic alcohols, tertiary allylic alcohols, and olefins with α-quaternary centers, have led to a general model useful for the prediction of product selectivity and stereoselectivity in cross metathesis. As a general ranking of olefin reactivity in CM, olefins can be categorized by their relative abilities to undergo homodimerization via cross metathesis and the susceptibility of their homodimers toward secondary metathesis reactions. When an olefin of high reactivity is reacted with an olefin of lower reactivity (sterically bulky, electron-deficient, etc.), selective cross metathesis can be achieved using feedstock stoichiometries as low as 1:1. By employing a metathesis catalyst with the appropriate activity, selective cross metathesis reactions can be achieved with a wide variety of electron-rich, electron-deficient, and sterically bulky olefins. Application of this model has allowed for the prediction and development of selective cross metathesis reactions, culminating in unprecedented three-component intermolecular cross metathesis reactions.

Full text of DOI:10.1021/ja0214882

Published on Web 08/20/2003
A General Model for Selectivity in Olefin Cross Metathesis
Arnab K. Chatterjee, Tae-Lim Choi, Daniel P. Sanders, and Robert H. Grubbs*
Contribution from the Arnold and Mabel Beckman Laboratories for Chemical Synthesis,
DiVision of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125
Received December 31, 2002; E-mail: rhg@caltech.edu
Abstract: In recent years, olefin cross metathesis (CM) has emerged as a powerful and convenient synthetic
technique in organic chemistry; however, as a general synthetic method, CM has been limited by the lack
of predictability in product selectivity and stereoselectivity. Investigations into olefin cross metathesis with
several classes of olefins, including substituted and functionalized styrenes, secondary allylic alcohols,
tertiary allylic alcohols, and olefins with R-quaternary centers, have led to a general model useful for the
prediction of product selectivity and stereoselectivity in cross metathesis. As a general ranking of olefin
reactivity in CM, olefins can be categorized by their relative abilities to undergo homodimerization via cross
metathesis and the susceptibility of their homodimers toward secondary metathesis reactions. When an
olefin of high reactivity is reacted with an olefin of lower reactivity (sterically bulky, electron-deficient, etc.),
selective cross metathesis can be achieved using feedstock stoichiometries as low as 1:1. By employing
a metathesis catalyst with the appropriate activity, selective cross metathesis reactions can be achieved
with a wide variety of electron-rich, electron-deficient, and sterically bulky olefins. Application of this model
has allowed for the prediction and development of selective cross metathesis reactions, culminating in
unprecedented three-component intermolecular cross metathesis reactions.
Introduction
factors: first, low catalyst activity to effect a reaction without
a strong enthalpic driving force (such as ring-strain release in
Olefin cross metathesis (CM) is a convenient route to
functionalized and higher olefins from simple alkene precursors.
Cross metathesis has recently gained prominence due to the
availability of catalysts with varied activities, such as 1,1 2,2
and 33 (Figure 1). These catalysts have expanded the variety of
functional groups amenable to CM and have demonstrated the
ability to prepare highly substituted olefins by CM, often in a
stereoselective manner. The installation of structural elements
within complex natural products and the synthesis of reagents
for further synthetic transformations can now be accomplished
by CM using active and functional group tolerant metathesis
catalysts. However, cross metathesis remains an underrepre-
sented area of olefin metathesis when compared to ring-opening
metathesis polymerizations (ROMP)4 and ring-closing metath-
esis (RCM).5 This has been predominantly a result of several
ROMP) or the entropic advantage of intramolecular reactions
(such as RCM), second, low product selectivity for the CM
product, and, third, poor stereoselectivity in the newly formed
olefin. While the development of increasingly active catalysts
has resolved many of these concerns, the inability to accurately
predict selectivity of cross metathesis reactions remains a
pertinent issue for the practical application of cross metathesis.
By placing sterically large and electron-withdrawing groups
near the reacting olefin, we expected to be able to improve CM
product selectivity and stereoselectivity. As a result of our
investigations with catalysts 1 and 2, a significant number of
new substrates classes that participate in selective olefin cross
metathesis reactions have been discovered.6 While a descriptive
model of selective CM processes had not yet been disclosed,7
we observed that several different types of olefins could be
properly matched to provide highly selective CM reactions.
(1) (a) Scholl, S.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
(b) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc. 2001,
123, 6543.
(6) (a) Chatterjee, A. K.; Grubbs, R. H. Org. Lett. 1999, 1, 1751. (b) Chatterjee,
A. K.; Morgan, J. P.; Scholl, M.; Grubbs, R. H. J. Am. Chem. Soc. 2000,
122, 3783. (c) Choi, T.-L.; Chatterjee, A. K.; Grubbs, R. H. Angew. Chem.,
Int. Ed. 2000, 39, 1277. (d) Chatterjee, A. K.; Choi, T.-L.; Grubbs, R. H.
Synlett 2001, 1034. (e) Choi, T.-L.; Lee, C. W.; Chatterjee, A. K.; Grubbs,
R. H. J. Am. Chem. Soc. 2001, 123, 10417. (f) Goldberg, S. D.; Grubbs,
R. H. Angew. Chem., Int. Ed. 2002, 41, 807. (g) Chatterjee, A. K.; Sanders,
D. P.; Grubbs, R. H. Org. Lett. 2002, 4, 1939. (h) Toste, F. D.; Chatterjee,
A. K.; Grubbs, R. H. Pure Appl. Chem. 2002, 74, 7. (i) Chatterjee, A. K.;
Toste, F. D.; Choi, T.-L.; Grubbs, R. H. AdV. Synth. Catal. 2002, 344,
634. (j) Chatterjee, A. K.; Grubbs, R. H. Angew. Chem., Int. Ed. 2002, 41,
3171.
(7) The following review on olefin metathesis comments on the lack of a
general model for selectivity in CM: Schuster, M.; Blechert, S. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2036; Angew. Chem. 1997, 109, 2124. For
a recent review on cross metathesis, see: Connon, S. J.; Blechert, S. Angew.
Chem., Int. Ed. 2003, 42, 1900.
(2) (a) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2039; Angew. Chem. 1995, 107, 2179. (b) Schwab,
P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 100. (c)
Belderrain, T. R.; Grubbs, R. H. Organometallics 1997, 16, 4001.
(3) (a) Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; Dimare, M.;
O’Regan, M. J. Am. Chem. Soc. 1990, 112, 3875. (b) Bazan, G. C.;
Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V. C.; O’Regan, M.
B.; Thomas, J. K.; Davis, W. M. J. Am. Chem. Soc. 1990, 112, 8378. (c)
Bazan, G. C.; Oskam, J. H.; Cho, H. N.; Park, L. Y.; Schrock, R. R. J.
Am. Chem. Soc. 1991, 113, 6899.
(4) For a recent review, see: Frenzel, U.; Nuyken, O. J. Polym Sci. Part A:
Polym. Chem. 2002, 40, 2895.
(5) (a) Fu¨rstner, A. Angew. Chem. 2000, 112, 3013; Angew. Chem., Int. Ed.
2000, 39, 3140. (b) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001,
34, 18. (c) Kotha, S.; Sreenivasachary, N. Ind. J. Chem. B, 2001, 40, 763.
9
11360
J. AM. CHEM. SOC. 2003, 125, 11360-11370
10.1021/ja0214882 CCC: $25.00 © 2003 American Chemical Society

Selectivity in Olefin Cross Metathesis
A R T I C L E S
Figure 1. Commonly used olefin metathesis catalysts.
Scheme 1. Homodimerization in Cross Metathesis
sterically hindered and/or electron-deficient olefins falling into
Types II through IV.
Olefin Categorization for Metathesis Catalysts of Various
Activities. Table 1 categorizes all reported CM substrates for
catalysts 1, 2, and 3 and provides chemists with two basic
functions. First, it provides a starting reference point for the
design of potentially selective CM reactions. Second, for those
working to develop more active metathesis catalysts, it provides
a challenging set of olefins that are not active with current
catalysts in CM (i.e., Type IV olefins). Up to this point,
methodology developed in the area of olefin metathesis has been
marked by repeated use of the most active catalyst available.
While simple modification of the steric or electronic properties
of an olefin (i.e., by choice of protecting groups) is often
sufficient to change the reactivity of an olefin and result in a
selective reaction, CM selectivity may also be achieved by
simply choosing a catalyst with differing activity, as will be
shown specifically in the case of styrene CM. Since these
catalysts are commercially available, it is straightforward to
employ the most selective catalyst. As new types of olefins are
used in CM, assigning them an appropriate olefin type will allow
them to be used more effectively in selective CM.
Figure 2. Olefin categorization and rules for selectivity.
These trends from our CM results, in addition to the trends
derived from the rapidly expanding body of CM literature,
provide the foundation for an empirical model for product
selective CM. In this article, we will present a general model
based on the categorization of olefin reactivity which can be
used to predict both selective and nonselective cross metathesis
reactions for a number of commercially available metathesis
catalysts with varying activities.
Olefin Reactivity and Product Selectivity in Cross Me-
tathesis. Given the various possible alkylidene intermediates
and the numerous primary and secondary metathesis pathways
involved in a cross metathesis reaction, it is difficult to
accurately predict how the complex interplay of steric and
electronic factors will determine the ability of various sets of
olefins to participate in selective CM reactions. Due to the
multitude of factors influencing olefin reactivity in cross
metathesis, a more straightforward, empirical ordering or
categorization of olefin reactivity is required. The most con-
venient way to rank olefin reactivity is to examine their ability
to homodimerize (Scheme 1). However, instead of simply
looking at the absolute ability of an olefin to undergo ho-
modimerization, we looked at its ability to undergo homodimer-
ization relatiVe to other olefins and describe olefins on a gradient
scale of their propensity to undergo homodimerization, and
importantly, the subsequent reactivity of their homodimers. This
analysis leads to a general model that comprises four distinct
olefin types which can be used to predict both selective and
nonselective CM reactions (Figure 2).
Type I olefins are categorized as those able to undergo a rapid
homodimerization and whose homodimers can participate in CM
as well as their terminal olefin counterpart. Type II olefins
homodimerize slowly, and unlike Type I olefins, their ho-
modimers can only be sparingly consumed in subsequent
metathesis reactions. Type III olefins are essentially unable to
be homodimerized by the catalyst but are still able to undergo
CM with Type I and Type II olefins. Type IV olefins are not
able to participate in CM with a particular catalyst but do not
inhibit catalyst activity toward other olefins. Outside these
categories are olefins that deactivate the catalyst. In general, a
reactivity gradient exists from most active type (Type I olefin)
to least active Type (Type IV), with sterically unhindered,
electron-rich olefins categorized as Type I and increasingly
An olefin’s general reactivity (as shown in Figure 2) with a
given catalyst determines the role of secondary metathesis, that
is, subsequent reactions of a product olefin with the propagating
catalyst. Predicting the ability (or inability) of the catalyst to
perform secondary metathesis on a newly formed CM olefin is
important for the development of selective CM reactions.
Efficient secondary metathesis occurs when all components in
the reaction are readily accessible to the metal alkylidene
complex, including homodimers and the CM product. The key
to CM reaction selectivity is minimizing the number of
undesirable CM side products (such as the homodimers of the
starting olefins) either by avoiding their initial formation or by
ensuring that they are fully consumed in secondary metathesis
events. It is also important that the desired cross product not
be redistributed into a statistical product mixture by these same
secondary metathesis events.
Throughout this paper, examples of both selective and
nonselective CM reactions with combinations of olefins from
various types (categorized in Table 1) will be presented to
illustrate how the general model shown in Figure 2 accounts
for CM selectivity with a wide variety of substrates and how it
can be used to predict selectivity in new CM reactions. Specific
attention will be paid to how the manipulation of the rates of
homodimerization and the role of secondary metathesis in CM
through the modification of the steric and electronic properties
9
J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003 11361

A R T I C L E S
Chatterjee et al.
Table 1. Olefin Categories for Selective Cross Metathesis
Scheme 2. Equilibration of Cross Products
product. However, the reaction with catalyst 1 yields a much
higher amount of the more thermodynamically favorable trans
isomer, presumably due to secondary metathesis on the CM
product formed in the reaction. Therefore, the increased trans
ratio is simply an effect of the higher activity of catalyst 1
toward the product than catalyst 2. This brings up the compli-
cated and multifaceted role of secondary metathesis. As will
be shown in subsequent sections, in some instances, we observe
good product selectivity due to the stability of the cross product
toward redistribution by secondary metathesis but poor stereo-
selectivity due to the inability of the cross product to readily
undergo cis/trans isomerization via these same secondary
metathesis processes.
of an olefin or the appropriate choice of catalyst can lead to
selectivity in cross metathesis.
Nonselective Cross Metathesis with Two Type I Olefins.
When two Type I olefins are used in a CM reaction, the rates
of homodimerization are similar and the reactivities of the
homodimers and cross products toward secondary metathesis
events are high. In these reactions, the desired cross product
will be equilibrated with the various homodimers through
secondary metathesis reactions (Scheme 2). This will result in
a statistical product mixture. For these reactions, one must use
nearly 10 equiv of one CM partner to provide 90% of the CM
product (Scheme 3). More generally, when two olefins of the
same type are combined, nonselective product mixtures are
usually obtained. For example, two Type II olefins (such as
methyl vinyl ketone and methyl acrylate) can react with each
other but will generally undergo nonselective CM, albeit with
reduced yields due to the lower overall olefin reactivity.6e
A good example of nonselective CM is the reaction of allylic
alcohols using either catalyst 1 or 2 (Scheme 4). The CM
reaction between allylbenzene and an allylic alcohol equivalent
affords the CM product 4 in 80% isolated yield with either
catalyst 1 or 2, with good to moderate stereoselectivity. Since
both reaction partners are Type I olefins, four allyl acetate
equivalents are required to provide an 80% yield of the CM
Selective Cross Metathesis. To avoid the statistical product
distributions produced by these inefficient reactions, one can
design selective CM reactions using olefins from two different
types, whose rates of dimerization are significantly different
and/or slower than CM product formation. The first approach
toward selective CM involves the reaction of a Type I olefin
with a less reactive Type II or Type III olefin that undergoes
homodimerization at a significantly lower rate, if at all. In this
reaction, although the Type I olefin may initially homodimerize,
the product distribution is driven toward the desired cross
product as ethylene is driven from the system (preventing the
regeneration of terminal olefins) and the Type I homodimer
readily undergoes secondary metathesis with the Type II/III
olefin (Scheme 5). This desired cross product will not be
9
11362 J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003

Selectivity in Olefin Cross Metathesis
A R T I C L E S
Scheme 3. Statistical Distribution of CM Products
Scheme 4. Nonselective Olefin Cross Metathesis
Scheme 5. Primary Reactions in Cross Metathesis of Type I with Type II/III
Scheme 6. Cross Metathesis of Type I Olefin with Methyl
Fumarate
if protecting groups were required to provide stereoselectivity
in this system, similar to our previous work with catalyst 2.8
We hypothesized that increasing steric bulk through the addition
of protecting groups would decrease CM reactivity but would
increase the trans selectivity. Interestingly, although the reactiv-
ity trends were as expected, the product stereoselectivity trends
were opposite to our hypothesis. Greater selectivity was
observed with the unprotected alcohol (entry 3) than with a
bulky protecting group, such as a tert-butyldiphenylsilyl ether
(Entry 5). It is not clear why a smaller protecting group (entry
2 vs entry 5) allows for greater trans selectivity but may in part
be due to greater steric accessibility of the cross product with
the smaller protecting group to isomerization via secondary
metathesis.
Selective Cross Metathesis of Type I with Type III Olefins.
During the course of our earlier studies with catalyst 2, we found
that olefins with fully substituted allylic carbons/quaternary
centers did not participate in CM.8 They did not eliminate
activity of the catalyst but simply did not participate in the
reaction (Type IV). However, due to the greater activity of
catalyst 1, CM reactions of quaternary allylic olefins with
R-olefins are now possible with excellent stereoselectivity due
to their great steric bulk (Table 3). These reactions are useful
because they are able to install highly substituted carbons in a
stereodefined manner.
We were pleased to discover that these reactions are the first
example of exclusive trans olefin selectivity in CM based solely
on the sterics of alkyl substituents. For example, an unprotected
tertiary alcohol (entry 1) provides an excellent yield of the CM
product with only the trans isomer observed by 1H NMR. Alkyl
substituents have also been explored in the reaction and work
quite nicely with catalyst 1 with a variety of terminal olefins
(entries 2 and 3). There are several key differences observed
when comparing the selectivity of tertiary allylic substrates in
CM (Table 2) with quaternary allylic substrates (Table 3). While
the steric bulk of tertiary allylic substrates leads to the initial
equilibrated to a statistical product mixture due to the inability
of the catalyst to efficiently convert the cross product to other
products (i.e., the homodimer of the Type II/Type III olefin)
via secondary metathesis.
Selective Cross Metathesis of Type I with Type II Olefins.
For example, reactions between Type I terminal olefins and
Type II olefins such as R,â-unsaturated carbonyl olefins,
including acrylates, acrylamides, acrylic acid, and vinyl ketones
result in highly selective CM reactions with high stereoselec-
tivity (E/Z > 20:1).6b,c In this case, methyl acrylate is a Type II
olefin because it undergoes homodimerization to a small extent
under the reaction conditions, allowing for selective reactions
with Type I olefins.6b To illustrate the low reactivity of the
homodimers of these Type II olefins, when methyl fumarate,
the trans homodimer of methyl acrylate, is resubjected to catalyst
1 and 5-hexenyl acetate, no CM product is observed (Scheme
6). This is similar to our previously reported results using the
homodimer of 2-vinyl-1,3-dioxolane with catalyst 2.8
Secondary allylic alcohols are another class of Type II olefins
which can be utilized in CM with moderate to high cross product
yields and good stereoselectivity8 without using large stoichio-
metric excesses of one reagent (Table 2). It is interesting that
the addition of a methyl group at the allylic position leads to
much greater trans selectivity, compared to the 7:1 E/Z ratio
obtained with a primary allylic alcohol. Similar results were
also obtained using catalyst 2.8 Interestingly, most of the trans
selectivity obtained in entry 1 is retained when a simple R-olefin
is used in the reaction (entry 2). We decided to also investigate
(8) Blackwell, H. E.; O’Leary, D. J.; Chatterjee, A. K.; Washenfelder, R. A.;
Bussmann, D. A.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122, 58.
9
J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003 11363

A R T I C L E S
Chatterjee et al.
Table 2. Secondary Allylic Alcohol Cross Metathesis
1
a Determined by H NMR. b Reaction performed at 23 °C.
Table 3. Quaternary Allylic Olefin Cross Metathesisa
1
a 3-5 mol % of catalyst 1 used, CH2Cl2, 40 °C, 12 h. b Only E isomer observed by H NMR.
Scheme 7. Altering Cross Metathesis Selectivity Using Steric Effect
formation of moderately trans product distribution, it also
hinders secondary metathesis with the result being moderate
trans selectivity. However, the greater steric bulk of the
quaternary allylic substrates presumably leads to the initial
formation of an exclusively trans product distribution which is
unchanged by secondary metathesis events. In addition, the
homodimerization of the quaternary allylic olefins is negligible
in many cases, allowing for greater CM product selectivity. For
example, with 2-methyl-2-vinyl-1,3-dioxolane (Table 3, entries
4 and 5), there was no observation of the dimer of the cyclic
ketal, characteristic of a Type III olefin. However, when the
tertiary alcohol in entry 1 was used in the CM at 40 °C, there
was a background dimerization of the tertiary alcohol. When a
similar tertiary alcohol was used, a reduced yield of CM product
was observed (Scheme 7). In addition, the independent dimer-
ization of this tertiary allylic alcohol can be performed in
excellent yields. However, if this isolated dimer is subsequently
reacted with a terminal olefin, no CM product is observed. This
indicates that once the dimer is formed, it does not undergo
secondary metathesis presumably due to steric bulk. In addition,
we found homodimers of olefins with tertiary allylic carbons
were not accessible for secondary metathesis either. As a result
of these observations, unprotected tertiary allylic alcohols can
be considered to be Type II olefins. Interestingly, this undesired
dimerization could be suppressed to a large extent by increasing
steric bulk via the addition of a silyl protecting group, effectively
9
11364 J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003

Selectivity in Olefin Cross Metathesis
A R T I C L E S
Table 4. Cross Metathesis of 1,1-Disubstituted Olefins
1
1
a Determined by H NMR, confirmed by H NMR NOE experiments.
changing the substrate to a Type III olefin (Scheme 7). This
simple substrate modification results in increased CM selectivity
by suppressing the unwanted homodimerization side reaction.
These reactions represent the control of product and stereose-
lectivity in CM based purely on steric considerations, which
may be manipulated through the judicious choice of protecting
groups.
Other examples of CM with Type III olefins, such as 1,1-
disubstituted olefins, using catalyst 1 are shown in Table 4.6a,b
Even functionalized 1,1-disubstituted olefins are excellent
substrates for CM which can give high stereoselectivity (Table
4, entries 2 and 4). Highly active catalyst 1 has also been shown
to perform secondary metathesis on trisubstituted prenyl olefins6g
(Entry 4) allowing for a regio- and stereoselective formal allylic
oxidation of one of the terminal methyl groups. The reason
methacrylic acid (entry 3) gives relatively poor yield and E/Z
ratio is unclear. The results in Table 4 demonstrate the ability
to use olefins with functionalities of several different oxidation
states in the CM reactions.
Selective Cross Metathesis of Type II with Type III
Olefins. A second approach to selective CM processes utilizes
two olefins which both dimerize at much slower rates than the
formation of productive cross-metathesis product. The inability
of Type III olefins to homodimerize allows them to also undergo
selective reactions with Type II CM partners. In these reactions,
formation of cross product dominates if the rates of homodimer-
ization of Type II olefins and secondary metathesis of the CM
products are very slow. For example, most 1,1-disubstituted
olefins will readily perform selective CM with terminal (Type
I) olefins as well as acrylates6e (Type II) and acrolein acetals6a
(Type II) but will not homodimerize. However, given the low
reactivity of some Type III olefins, reduced CM yields may be
obtained. Also, since the undesired dimers of Type II olefins
are essentially inactive for further CM (unlike Type I dimers),
stoichiometric excesses of the less reactive Type III olefin may
be required to produce high yields of cross product, such as
the CM of acrylates carried out in neat isobutylene.6g
acrylate (Type II) and ethyl vinyl ketone (Type II) compete with
the CM formation, resulting in a considerable amount of
homodimer side products. However, by capping with a â-methyl
group, the dimerization is suppressed and higher selectivity
toward CM product is obtained. Another strategy is to utilize
slow addition of the Type II olefin in order to maintain a low
concentration of the more reactive Type II olefin, thereby,
minimizing the amount of dimerization (Table 5, entry 7).
Cross Metathesis with Olefins that Bridge Type Catego-
ries. While protection of tertiary allylic alcohols has been shown
to change their CM behavior leading to enhanced CM selectiv-
ity, other olefins, such as styrenes, bridge the broad type
categories outlined previously. The CM behavior of styrenes
depends strongly on their substitution patterns, providing far
more flexibility in terms of functionalization/substitution and
catalyst choice in order to achieve selective reactions. Styrenes
represent one of the classes of olefins used widely in CM with
ill-defined catalyst systems,9 as well as 210 and 3,11 because of
high trans selectivity in the CM product. In all these cases, the
dimerization of styrene to stilbene was reported to be slow,
allowing for moderate to good selectivities in CM (Type II).
However, catalyst 1 showed a significantly different reactivity
and prompted further investigation into this family of CM
substrates. For example, with 2.5 mol % of catalyst 1, the
dimerization of styrene to E-stilbene was achieved in 94%
isolated yield. Consequently, the CM reaction of styrene with
a terminal olefin employing catalyst 1 produces a statistical
product distribution (Table 6, entry 1) and is different from the
results using catalyst 3, as reported by Crowe and Zhang (entry
2).11 To confirm that homodimerized styrene can undergo
efficient secondary metathesis during CM with catalyst 1,
E-stilbene was successfully used as a styrene surrogate in CM
with an allylic acetate equivalent (Entry 7). The reaction
produced a statistical ratio of CM products. This is unprec-
(9) Warwel, S.; Winkelmuller, W. J. Mol. Catal. 1985, 28, 247.
(10) (a) Feher, F. J.; Soulivong, D.; Eklund, A. G.; Wyndham, K. D. Chem.
Commun. 1997, 1186. (b) Biagini, S. C. G.; Gibson, S. E.; Keen, S. P. J.
Chem. Soc., Perkin Trans. 1 1998, 2485. (c) Huwe, C. M.; Woltering, T.
J.; Jiricek, J.; Weitz-Schmidt, G.; Wong, C.-H. Bioorg. Med. Chem. 1999,
773. (d) Seshadri, H.; Lovely, C. J. Org. Lett. 2000, 2, 327. (e) Eichelberger,
U.; Mansourova, M.; Henning, L.; Findeisen, M.; Giesa, S.; Muller, D.;
Welzel, P. Tetrahedron 2001, 57, 9737. (f) Yammamoto, Y.; Takahashi,
M.; Miyaura, N. Synlett 2002, 128.
In a similar manner, cross metathesis was carried out in neat
3,3-dimethyl-1-butene to drive the CM between the Type III
olefin and various Type II olefins (Table 5, entries 1-3). Entries
5 and 6 illustrate the importance of the relative rates of
dimerization of the Type II olefins. Even with 4 equiv of
2-methyl-1-heptene (Type III), the rates of dimerization of ethyl
(11) (a) Crowe, W. E.; Zhang, Z. J. J. Am. Chem. Soc. 1993, 115, 10998. (b)
Crowe, W. E.; Goldberg, D. R.; Zhang, Z. J. Tetrahedron Lett. 1996, 37,
2117.
9
J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003 11365

A R T I C L E S
Chatterjee et al.
Table 5. Cross Metathesis between Type II and Type III Olefins
1
1
a E/Z ) 2 determined by H NMR, confirmed by H NOE experiments. b Vinyldioxolane (3 equiv) added slowly in four equal parts over a 6 h period.
1
1
E/Z ) 3 determined by H NMR, confirmed H NOE experiments.
Table 6. Cross Metathesis of Styrenes with Terminal Olefins
1
a Only E isomer observed by H NMR.
edented since catalysts 2 and 3, as well as ill-defined catalysts,
are not able to efficiently use stilbene as a CM partner, providing
another example of the unique reactivity of catalyst 1. These
results clearly indicate that styrene is a Type I olefin for the
more active catalyst 1; however, for catalyst 3 employed by
Crowe and Zhang, styrene is a Type II olefin. Styrene CM
clearly demonstrates how matching the olefin with an appropri-
ate catalyst (shown in Table 1) can be equally as important as
matching olefins in order to achieve selective CM.
In addition to the possibility of using a less active catalyst to
9
11366 J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003

Selectivity in Olefin Cross Metathesis
A R T I C L E S
Table 7. Styrene Cross Metathesis with Acrylate Esters
1
1
a Only E isomer observed by H NMR. b Determined by H NMR.
achieve selectivity, alterations in styrene structure allow for
selective CM reactions with terminal olefins using catalyst 1,
by changing styrene from a Type I olefin to a Type II or Type
III olefin. For example, the use of 2-bromostyrene as the CM
partner leads to selective formation of the CM product (entry
3). By simply using an excess of this substituted styrene, we
achieved near quantitative conversion of hexenyl acetate. In the
case of 2-bromostyrene, both the steric bulk of the bromine atom
and its electron-withdrawing character contribute to make it a
Type II olefin. Crowe and Zhang also were able to incorporate
ortho-substituted styrenes in CM with catalyst 3 but found that
their reactivity is low with terminal olefins (entry 4).11 This
may be due to low catalyst activity toward electronic-deficient
styrene substrates, since the accompanying terminal olefin was
dimerized efficiently. Due to the higher reactivity of 1, we only
observed reduced activity when multiple electron-withdrawing
substituents were present. For example, 2,6-difluorobenzene was
subjected to CM conditions, and only moderate yields of CM
product were isolated (entry 6).
With the differences in reactivity observed with substituted
styrenes, we began to investigate CM of styrenes with a variety
of non-Type I olefins. We had previously disclosed that a variety
of Type II olefins such as R,â-unsaturated esters, amides,
ketones, and acids are excellent CM partners with terminal
olefins.6b,c In addition, it has been demonstrated by Crowe and
Goldberg that CM of π-conjugated olefins, such as acrylonitriles,
was not compatible with CM with styrenes using catalyst 3
because they possessed similar electronic properties. They
suggested that CM required matching of a more nucleophilic,
electron-rich olefin with either styrene or acrylonitrile.12 How-
ever, in contrast to Crowe’s work with catalyst 3, we found
that styrenes are excellent CM partners with electron-deficient
R,â-unsaturated carbonyls, such as acrylates using catalyst 1
(Table 7). This CM complements Wittig or Horner-Wad-
sworth-Emmons chemistry since unprotected benzaldehydes
work well (entry 7). Similar results were also obtained using
acrylamidesandvinylphosphonatesasthe“enone”component.6c,d,i
We discovered that the reactivity trends of styrenes were
different in these CM reactions when compared to terminal
olefin CM in Table 6. While Type I styrenes (Table 7, entries
1 and 6) were excellent CM partners with acrylates, it was
observed that ortho-substituted styrenes that did not dimerize
readily (Type II or Type III olefins) were also not good CM
partners with acrylates (entries 3-4, 9-11). Simple ortho-alkyl
groups did not reduce the reaction yield with acrylates (entry
2); however, when electron-withdrawing functionalities (entry
4) or significant steric bulk (entry 3) was added to the ortho
position, CM yields fell dramatically due to the low reactivity
of both cross partners with the catalyst. Therefore, for proper
CM selectivity, the two olefins in CM need to have a difference
in rate of reaction with the metal alkylidene complex (e.g., the
olefins must be of different types and, preferably, one should
be able to readily react with the catalyst to make an active
alkylidene).
As additional evidence for alteration in styrene reactivity
based on substitution patterns, we investigated ortho-phenol
styrene derivatives. These are interesting substrates for catalytic
reactions, since several ortho-phenol derivatives form stable
benzylidene complexes.13 However, instead of inhibiting catalyst
activity, a variety of protected phenols are active for catalytic
CM (Scheme 8). We found that small protecting groups, such
as acetate, allowed for excellent CM with acrylates (Type I +
(12) Crowe, W. E.; Goldberg, D. R. J. Am. Chem. Soc. 1995, 117, 5162.
9
J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003 11367

A R T I C L E S
Chatterjee et al.
Scheme 8. ortho-Phenol Cross Metathesis
Scheme 9. Chemoselective CM Based on Olefin Categorization
Scheme 10. Chemoselective Cross Metathesis Using Catalysts 2 and 3
Type II, respectively). The balance of the material in this
reaction was recovered as the stilbene dimer. This protection
pattern is similar to other unhindered styrenes in Table 4.
However, when a larger protecting group is employing, such
as tert-butyldimethylsilyl, CM of the hindered styrene with
acrylate (Type III + Type II, respectively) gives poor yields.
In contrast, the reaction provides a very good yield with allyl
acetate equivalents (Type III + Type I, respectively). In this
case, this substrate reacts like Type III due to steric bulk and is
very selective in CM with Type I R-olefins. Collectively, the
disclosed work in styrene CM provided an important contribu-
tion to the development of the categorization model described
in this paper. Both the steric and/or electronic effects of
substituents on the styrene, as well as the catalyst choice, can
be manipulated to achieve CM selectivity.
Regioselective and Chemoselective Cross Metathesis. The
model outlined so far provides a foundation for predicting
product selectivity in CM, but it also can be applied to the
prediction of chemoselective CM, which is critical for dif-
ferentiating olefins in the synthesis of complex molecules. The
CM of a Type I or Type II olefin in the presence of a Type IV
olefin is a good example of this type of chemoselectivity. For
example, using catalyst 2, a disubstituted R,â-unsaturated
carbonyl containing olefin (Type IV) is not affected, allowing
for selective reactions between a Type I olefin dimer and a Type
I olefin (Scheme 9).
Importantly, the CM model outlined in Figure 2 also is
consistent with results previously reported by other groups. For
example, Blechert, et al. used steric constraints and heteroatom
functionality to demonstrate that a highly substituted allylamine
(Type IV for catalyst 2) could be benign to CM in the presence
of two Type I olefins (Scheme 10).14 Interestingly, in the same
report by Blechert, catalyst 3 was used to effect a highly
selective CM reaction of that protected allylamine (Type III
for catalyst 3) with allylsilanes (Type I) in excellent yield
(Scheme 10). In addition, this is one of the first examples of
using steric bulk at the allylic carbon to obtain high olefin
stereoselectivity and is comparable to the results we observed
in Tables 1 and 2 with catalyst 1.
Similarly, Crowe and Zhang performed a selective CM
between a Type I terminal olefin and styrene (Type II for
catalyst 3) in the presence of a 1,1-disubstituted olefin (Type
IV).11a As demonstrated previously in our lab,6a,g with the more
active catalyst 1, 1,1-disubstituted olefins are a Type III olefin
(14) Brummer, O.; Ruckert, A.; Blechert, S. Chem.sEur. J. 1997, 3, 441.
(15) Feng, J.; Schuster, M.; Blechert, S. Synlett 1997, 129.
(16) (a) Pietraszuk, C.; Marciniec, B.; Fischer, H. Organometallics 2000, 19,
913. (b) Kujawa-Welten, M.; Pietraszuk, C.; Marciniec, B. Organometallics
2002, 21, 840.
(17) Blanco, O. M.; Castedo, L. Synlett 1999, 557.
(18) Faure, S.; Piva-Le Blanc, S.; Piva, O. Tetrahedron Lett. 1999, 40, 6001.
(19) Langer, P.; Holtz, E. Synlett 2001, 110.
(20) Zhang, L.; Herndon, J. W. Tetrahedron Lett. 2002, 43, 4471.
(21) Maishal, T. K.; Sinha-Mahapatra, D. K.; Paramjape, K.; Sarkar, A.
Tetrahedron Lett. 2002, 43, 2263.
(22) Grela, K.; Bieniek, M. Tetrahedron Lett. 2001, 42, 6425.
(23) Imhof, S.; Randl, S.; Blechert, S. Chem. Commun. 2001, 1692.
(24) BouzBouz, S.; Cossy, J. Org. Lett. 2001, 3, 1451.
(25) BouzBouz, S.; De Lemos, E.; Cossy, J. AdV. Synth. Catal. 2002, 344, 627.
(13) (a) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. J. Am.
Chem. Soc. 2000, 122, 8168. (b) Wakamatsu, H.; Blechert, S. Angew.
Chem., Int. Ed. 2002, 41, 2403.
9
11368 J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003

Selectivity in Olefin Cross Metathesis
A R T I C L E S
Scheme 11. Three Component Olefin Cross Metathesis
Table 8. Three Component Olefin Cross Metathesisa
1
a Using 5-7 mol % of 1 in 0.1-0.2 M refluxing CH2Cl2, 12 h. b E/Z ) 8:1 by H NMR. c Reaction at 23 °C. d Method A ) added all components at
one time. Method B ) added component Z and then added component Y after 4 h.
Scheme 12. One-Pot Three Component Olefin Cross Metathesis
that is active for CM to form trisubstituted olefins. This shows
that while more active catalysts have a larger set of CM active
olefins (Type I, II, III), it is useful to identify Type IV olefins
for less active catalysts, to help elucidate possible chemoselec-
tive CM reactions (Scheme 10). While electronic and steric
parameters of olefins account as key contributing factors in ways
olefins are classified, other factors are often implied in
determining selectivity, including chelating ability of certain
functional groups to metal catalysts. For example, the effects
of carbonyl groups, such as acetate protecting groups, and allylic
heteroatoms have been implied to alter reactivity in CM. It is
for these reasons that a comprehensive empirical model is
necessary that can simplistically account for all of these
observations in the CM methodology literature.
somer. This reaction is successful because the Type III and Type
II olefins react at a much slower rate with each other than their
respective reactions with the Type I olefin. In addition, the
product olefin from the diene-methyl vinyl ketone CM does
not readily undergo secondary metathesis reactions with isobu-
tylene, allowing for a selective reaction.
The formation of a kinetic CM product also allows for
chemoselective coupling, where a one-pot sequential CM
reaction can occur (Scheme 12). For example, if two CM
partners that can react with each other are used in a three-
component reaction, such as styrene and methyl vinyl ketone,
then a sequential addition strategy is necessary to avoid the
unwanted side reaction between these components. Using these
two types of three-component CM reactions, a variety of
asymmetrically substituted dienes have been prepared (Table
8). These reactions illustrate the use of olefin categorization to
effectively predict proper three-component reactivity. In theory,
any combination of a Type I diene with a Type II and a Type
III olefin would provide a three-component product (Method
A, entries 1-4). As mentioned previously, if two Type I olefins
need to be coupled, then this coupling must be performed after
coupling of the Type II olefin (Method B, entries 5 and 6). The
reactions add a new level of complexity to olefin metathesis
Three-Component Cross Metathesis Reactions. In addition
to describing selectivity in the simple homologation of two
olefins in a CM reaction, the olefin classification in Table 1
also provides an opportunity to develop new reactions, such as
multicomponent processes. While a three-component reaction
has always been theoretically possible, the large mixture of
compounds that would form via nonselective processes has made
this a challenging method to develop. However, with the current
model of selective CM described here, two important concepts
have been learned. First, under selective CM conditions,
secondary metathesis of the resulting olefins can be significantly
slower than productive CM. Second, by using two olefins that
do not perform CM with each other or do so only very slowly,
then a third diene-containing olefin can be functionalized at both
olefinic sites to provide an unsymmetrical product (Scheme 11).
In such a reaction, olefins of three different types may be
converted predominantly into one product as a single stereoi-
9
J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003 11369

A R T I C L E S
Chatterjee et al.
Compound 5. cis-2-Butene-1,4-diacetate (160 µL, 0.9 mmol) and
2-benzyloxy-3-butene (90 µL, 0.51 mmol) were added simultaneously
via syringe to a stirring solution of 1 (11 mg, 0.015 mmol, 2.8 mol %)
in CH2Cl2 (2.5 mL). The flask was fitted with a condenser and refluxed
under nitrogen for 12 h. The reaction mixture was then reduced in
volume to 0.5 mL and purified directly on a silica gel column (2 × 10
cm), eluting with 9:1 hexane:ethyl acetate (500 mL). Pale oil was
obtained (48 mg, 0.19 mmol, 38% yield). Spectra was compared to
reported compound; see Blackwell, H. E.; O’Leary, D. J.; Chatterjee,
A. K.; Washenfelder, R. A.; Bussmann, D. A. J. Am. Chem. Soc. 2000,
122, 58. Rf ) 0.36 (9:1 hexane:ethyl acetate).
reactions and are possible due to development of a better
understanding of CM reactivity patterns.
Conclusions
In conclusion, a general empirical model for olefin reactivity
in cross metathesis has been developed for the prediction of
CM selectivity, in terms of olefin product selectivity, regiose-
lectivity, and chemoselectivity. A general ranking of olefin
reactivity in CM is achieved by categorizing olefins by their
relative ability to undergo homodimerization via cross metathesis
and the susceptibility of their homodimers toward secondary
metathesis reactions. Product selectivity can be achieved by
suppressing the rate of homodimerization of one component and
controlling the rate of secondary metathesis on the desired cross
product. These rates can be controlled through the choice of
olefins with significantly different activities, which can be
modified by altering their steric and electronic properties through
substituents, functionalities, or protecting groups. In addition,
an appropriate choice of olefin metathesis catalyst is critical
for product selectivity, regioselectivity, or chemoselectivity. This
empirical approach toward understanding cross metathesis
selectivity by categorizing the reactivity provides a convenient
starting point for the prediction and design of new, selective
CM reactions, such as multicomponent CM processes.
Compound 24. To an oven dried, 100 mL Fischer-Porter bottle
with Teflon stir bar, ruthenium metathesis catalyst 1 (14 mg, 0.017
mmol, 7.0 mol %) was added. The bottle was capped with a rubber
septum and flushed with dry nitrogen and cooled to -78 °C.
1,5-Hexadiene (85 µL, 0.72 mmol) and methyl vinyl ketone (20 µL,
0.24 mmol) were injected into the bottle. Once the substrates were
frozen, a pressure regulator was attached to the bottle. The bottle was
evacuated and backfilled with dry nitrogen 3 times. Subsequently,
isobutylene (10 mL) was condensed into the bottle. The bottle was
backfilled to ∼2 psi with nitrogen, sealed, and allowed to slowly warm
to room temperature, at which time it was transferred to an oil bath at
40 °C. After stirring for 12 h, the bottle was removed from the oil bath
and allowed to cool to room temperature. The isobutylene was slowly
vented off at room temperature until the pressure apparatus could be
safely disassembled. The reaction mixture was then reduced in volume
to 0.5 mL and purified directly on a silica gel column (2 × 10 cm),
eluting with 10:1 hexane:ethyl acetate. Clear oil was obtained (32 mg,
Experimental Section
1
0.21 mmol, 89% yield). H NMR (300 MHz, CDCl3, ppm): δ 6.78
General Information. Analytical thin-layer chromatography (TLC)
was performed using silica gel 60 F254 precoated plates (0.25 mm
thickness) with a fluorescent indicator. Flash column chromatography
was performed using silica gel 60 (230-400 mesh) from EM Science.
All other chemicals were purchased from the Aldrich or TCI America
and used as delivered unless noted otherwise. CH2Cl2 was purified by
passage through a solvent column prior to use. Catalyst 1 and 2 were
stored and manipulated on the bench. NMR spectra were recorded on
a Varian Mercury 300 MHz NMR.
(1H, dt, J ) 15.9, 6.6 Hz), 6.07 (1H, dt, J ) 15.9, 1.5 Hz), 5.12-5.06
(1H, m), 2.26-2.14 (7H, m), 1.69 (3H, s), 1.60 (3H, s). Rf ) 0.53 (9:1
hexane:ethyl acetate). Spectra matches that of a previous characteriza-
tion; see Coxon, J. M.; Garland, R. P.; Hartshorn, M. P. Aust. J. Chem.
1972, 25, 353.
Compound 25. 1,5-Hexadiene (70 µL, 0.59 mmol) and methyl vinyl
ketone (25 µL, 0.30 mmol) were added simultaneously via syringe to
a stirring solution of 1 (18 mg, 0.021 mmol, 7.1 mol %) in CH2Cl2
(2.0 mL) under a nitrogen atmosphere. The flask was fitted with a reflux
condenser stirred at 40 °C with a continuous flow of nitrogen for 3 h.
At that point, a solution of styrene (25 mL, 0.30 mmol) and catalyst 1
(16 mg, 0.019 mmol, 6.2 mol %) in CH2Cl2 was cannula transferred.
The reaction mixture was stirred at 40 °C for an additional 8 h. The
resulting solution was then reduced in volume to 0.5 mL and purified
directly on a silica gel column (2 × 10 cm), eluting with 15:1 hexane:
ethyl acetate to provide cross product (Rf ) 0.33 in 9:1 hexane:ethyl
Representative Procedure. Olefin A (1.0 mmol) and Olefin B (1.0
mmol) were added via syringe to a stirring solution of 2 (0.05 mmol,
5.0 mol %) in CH2Cl2 (2.5 mL). The flask was fitted with a condenser
and refluxed under nitrogen for 12 h. The reaction mixture was then
reduced in volume to 0.5 mL and purified directly on a silica gel column
(2 × 10 cm), eluting with hexane:ethyl acetate (500 mL).
Representative Procedure Using One Olefin as Solvent. Olefin
A (0.28 mmol) was added via syringe to a stirring solution of 1 (18
mg, 0.021 mmol, 7.6 mol %) in 3,3-dimethylbutene (1.5 mL, excess)
under a nitrogen atmosphere. The flask was stirred under a continuous
flow of nitrogen for 12 h at room temperature (23 °C). The reaction
mixture was then reduced in volume to 0.5 mL and purified directly
on a silica gel column (2 × 10 cm), eluting with hexane:ethyl acetate
to provide cross product.
1
acetate) as a clear oil (28 mg, 0.14 mmol, 47% yield). H NMR (300
MHz, CDCl3, ppm): d 7.35-7.21 (5H, m), 6.87-6.79 (1H, m), 6.42
(1H, d, J ) 15.9 Hz), 6.27-6.10 (2H, m), 2.41 (4H, app s), 2.26 (3H,
s). Spectra match those of a previously characterized compound; see
Johns, A.; Murphy, J. A.; Sherburn, M. S. Tetrahedron 1989, 45, 7835.
Compound 4. cis-2-Butene-1,4-diacetate (160 µL, 1.0 mmol) and
allylbenzene (55 µL, 0.50 mmol) were added simultaneously via syringe
to a stirring solution of 2 (11 mg, 0.014 mmol, 2.7 mol %) in CH2Cl2
(2.5 mL). The flask was fitted with a condenser and refluxed under
nitrogen for 12 h. The reaction mixture was then reduced in volume to
0.5 mL and purified directly on a silica gel column (2 × 10 cm), eluting
with 9:1 hexane:ethyl acetate (500 mL). Pale oil was obtained (76 mg,
80% yield, trans/cis as determined by integration of peaks at 4.73 and
4.55 ppm). 1H NMR (300 MHz, CDCl3, ppm): δ 7.34-7.17 (5H, m),
5.92 (1H, m), 5.65 (1H, m), 4.55 (2H, app d), 3.41 (2H, d, J ) 3.3
Hz), 2.06 (3H, unresolved s). 13C NMR (75 MHz, CDCl3, ppm): δ
171.4, 135.1, 134.0, 129.2, 129.1, 126.8, 125.8, 65.5, 60.8, 39.2, 21.6.
Rf ) 0.53 (9:1 hexane:ethyl acetate); HRMS (EI) calcd for C12H14O2
[M - H]+ 189.0916, found 189.0916.
Acknowledgment. The authors would like to thank Prof.
Helen E. Blackwell, Dr. John P. Morgan, Dr. Steven D.
Goldberg, Dr. Brian T. Connell, Prof. F. Dean Toste, and Prof.
Justin P. Gallivan for helpful discussions and encouragement.
The authors gratefully acknowledge the generous funding
provided by the National Institutes of Health.
Supporting Information Available: Experimental procedures
and nuclear magnetic resonance and mass spectroscopic data
for all new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA0214882
9
11370 J. AM. CHEM. SOC. VOL. 125, NO. 37, 2003